Search is not available for this dataset
pipeline_tag
stringclasses 48
values | library_name
stringclasses 205
values | text
stringlengths 0
18.3M
| metadata
stringlengths 2
1.07B
| id
stringlengths 5
122
| last_modified
null | tags
listlengths 1
1.84k
| sha
null | created_at
stringlengths 25
25
|
|---|---|---|---|---|---|---|---|---|
automatic-speech-recognition
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-large-xls-r-300m-spanish-custom
This model was trained from scratch on the common_voice dataset.
It achieves the following results on the evaluation set:
- eval_loss: 0.2245
- eval_wer: 0.2082
- eval_runtime: 801.6784
- eval_samples_per_second: 18.822
- eval_steps_per_second: 2.354
- epoch: 0.76
- step: 8400
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- num_epochs: 10
- mixed_precision_training: Native AMP
### Framework versions
- Transformers 4.16.0.dev0
- Pytorch 1.10.1+cu102
- Datasets 1.17.1.dev0
- Tokenizers 0.11.0
|
{"tags": ["generated_from_trainer"], "datasets": ["common_voice"], "model-index": [{"name": "wav2vec2-large-xls-r-300m-spanish-custom", "results": []}]}
|
glob-asr/base-spanish-asr
| null |
[
"transformers",
"pytorch",
"wav2vec2",
"automatic-speech-recognition",
"generated_from_trainer",
"dataset:common_voice",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
{}
|
glob-asr/test-asr-sp-model
| null |
[
"transformers",
"pytorch",
"wav2vec2",
"feature-extraction",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
automatic-speech-recognition
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-large-xls-r-300m-guarani-small
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4964
- Wer: 0.5957
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 30
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| No log | 6.65 | 100 | 1.1326 | 1.0 |
| 1.6569 | 13.32 | 200 | 0.5264 | 0.6478 |
| 1.6569 | 19.97 | 300 | 0.5370 | 0.6261 |
| 0.2293 | 26.65 | 400 | 0.4964 | 0.5957 |
### Framework versions
- Transformers 4.16.2
- Pytorch 1.10.0+cu111
- Datasets 1.18.3
- Tokenizers 0.11.0
|
{"language": ["gn"], "license": "apache-2.0", "tags": ["generated_from_trainer", "robust-speech-event", "gn", "hf-asr-leaderboard"], "datasets": ["common_voice"], "model-index": [{"name": "wav2vec2-large-xls-r-300m-guarani-small", "results": []}]}
|
glob-asr/wav2vec2-large-xls-r-300m-guarani-small
| null |
[
"transformers",
"pytorch",
"tensorboard",
"wav2vec2",
"automatic-speech-recognition",
"generated_from_trainer",
"robust-speech-event",
"gn",
"hf-asr-leaderboard",
"dataset:common_voice",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null | null |
{}
|
glob-asr/wav2vec2-large-xls-r-300m-spanish-medium
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
automatic-speech-recognition
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-large-xls-r-300m-spanish-small
This model is a fine-tuned version of [jhonparra18/wav2vec2-large-xls-r-300m-spanish-custom](https://huggingface.co/jhonparra18/wav2vec2-large-xls-r-300m-spanish-custom) on the common_voice dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3596
- Wer: 0.2105
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 0.1971 | 0.79 | 400 | 0.2169 | 0.2077 |
| 0.2293 | 1.58 | 800 | 0.2507 | 0.2418 |
| 0.2065 | 2.37 | 1200 | 0.2703 | 0.2459 |
| 0.1842 | 3.16 | 1600 | 0.2716 | 0.2495 |
| 0.1634 | 3.95 | 2000 | 0.2695 | 0.2510 |
| 0.1443 | 4.74 | 2400 | 0.2754 | 0.2435 |
| 0.1345 | 5.53 | 2800 | 0.3119 | 0.2654 |
| 0.1267 | 6.32 | 3200 | 0.3154 | 0.2573 |
| 0.1237 | 7.11 | 3600 | 0.3251 | 0.2666 |
| 0.1118 | 7.91 | 4000 | 0.3139 | 0.2503 |
| 0.1051 | 8.7 | 4400 | 0.3286 | 0.2573 |
| 0.0964 | 9.49 | 4800 | 0.3348 | 0.2587 |
| 0.0946 | 10.28 | 5200 | 0.3357 | 0.2587 |
| 0.0897 | 11.07 | 5600 | 0.3408 | 0.2590 |
| 0.0812 | 11.86 | 6000 | 0.3380 | 0.2560 |
| 0.079 | 12.65 | 6400 | 0.3304 | 0.2415 |
| 0.0753 | 13.44 | 6800 | 0.3557 | 0.2540 |
| 0.0717 | 14.23 | 7200 | 0.3507 | 0.2519 |
| 0.0691 | 15.02 | 7600 | 0.3554 | 0.2587 |
| 0.0626 | 15.81 | 8000 | 0.3619 | 0.2520 |
| 0.0661 | 16.6 | 8400 | 0.3609 | 0.2564 |
| 0.0582 | 17.39 | 8800 | 0.3818 | 0.2520 |
| 0.0556 | 18.18 | 9200 | 0.3685 | 0.2410 |
| 0.0515 | 18.97 | 9600 | 0.3658 | 0.2367 |
| 0.0478 | 19.76 | 10000 | 0.3701 | 0.2413 |
| 0.0486 | 20.55 | 10400 | 0.3681 | 0.2371 |
| 0.0468 | 21.34 | 10800 | 0.3607 | 0.2370 |
| 0.0452 | 22.13 | 11200 | 0.3499 | 0.2286 |
| 0.0399 | 22.92 | 11600 | 0.3647 | 0.2282 |
| 0.0393 | 23.72 | 12000 | 0.3638 | 0.2255 |
| 0.0381 | 24.51 | 12400 | 0.3359 | 0.2202 |
| 0.0332 | 25.3 | 12800 | 0.3488 | 0.2177 |
| 0.033 | 26.09 | 13200 | 0.3628 | 0.2175 |
| 0.0311 | 26.88 | 13600 | 0.3695 | 0.2195 |
| 0.0294 | 27.67 | 14000 | 0.3624 | 0.2164 |
| 0.0281 | 28.46 | 14400 | 0.3688 | 0.2113 |
| 0.0274 | 29.25 | 14800 | 0.3596 | 0.2105 |
### Framework versions
- Transformers 4.16.0.dev0
- Pytorch 1.10.1+cu102
- Datasets 1.17.1.dev0
- Tokenizers 0.11.0
|
{"tags": ["generated_from_trainer"], "datasets": ["common_voice"], "model-index": [{"name": "wav2vec2-large-xls-r-300m-spanish-small", "results": []}]}
|
glob-asr/wav2vec2-large-xls-r-300m-spanish-small
| null |
[
"transformers",
"pytorch",
"wav2vec2",
"automatic-speech-recognition",
"generated_from_trainer",
"dataset:common_voice",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-large-xls-r-300m-spanish-large
This model is a fine-tuned version of [tomascufaro/xls-r-es-test](https://huggingface.co/tomascufaro/xls-r-es-test) on the common_voice dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1431
- Wer: 0.1197
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 10
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 20
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 300
- num_epochs: 5
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 0.1769 | 0.15 | 400 | 0.1795 | 0.1698 |
| 0.217 | 0.3 | 800 | 0.2000 | 0.1945 |
| 0.2372 | 0.45 | 1200 | 0.1985 | 0.1859 |
| 0.2351 | 0.6 | 1600 | 0.1901 | 0.1772 |
| 0.2269 | 0.75 | 2000 | 0.1968 | 0.1783 |
| 0.2284 | 0.9 | 2400 | 0.1873 | 0.1771 |
| 0.2014 | 1.06 | 2800 | 0.1840 | 0.1696 |
| 0.1988 | 1.21 | 3200 | 0.1904 | 0.1730 |
| 0.1919 | 1.36 | 3600 | 0.1827 | 0.1630 |
| 0.1919 | 1.51 | 4000 | 0.1788 | 0.1629 |
| 0.1817 | 1.66 | 4400 | 0.1755 | 0.1558 |
| 0.1812 | 1.81 | 4800 | 0.1795 | 0.1638 |
| 0.1808 | 1.96 | 5200 | 0.1762 | 0.1603 |
| 0.1625 | 2.11 | 5600 | 0.1721 | 0.1557 |
| 0.1477 | 2.26 | 6000 | 0.1735 | 0.1504 |
| 0.1508 | 2.41 | 6400 | 0.1708 | 0.1478 |
| 0.157 | 2.56 | 6800 | 0.1644 | 0.1466 |
| 0.1491 | 2.71 | 7200 | 0.1638 | 0.1445 |
| 0.1458 | 2.86 | 7600 | 0.1582 | 0.1426 |
| 0.1387 | 3.02 | 8000 | 0.1607 | 0.1376 |
| 0.1269 | 3.17 | 8400 | 0.1559 | 0.1364 |
| 0.1172 | 3.32 | 8800 | 0.1521 | 0.1335 |
| 0.1203 | 3.47 | 9200 | 0.1534 | 0.1330 |
| 0.1177 | 3.62 | 9600 | 0.1485 | 0.1304 |
| 0.1167 | 3.77 | 10000 | 0.1498 | 0.1302 |
| 0.1194 | 3.92 | 10400 | 0.1463 | 0.1287 |
| 0.1053 | 4.07 | 10800 | 0.1483 | 0.1282 |
| 0.098 | 4.22 | 11200 | 0.1498 | 0.1267 |
| 0.0958 | 4.37 | 11600 | 0.1461 | 0.1233 |
| 0.0946 | 4.52 | 12000 | 0.1444 | 0.1218 |
| 0.094 | 4.67 | 12400 | 0.1434 | 0.1206 |
| 0.0932 | 4.82 | 12800 | 0.1424 | 0.1206 |
| 0.0912 | 4.98 | 13200 | 0.1431 | 0.1197 |
### Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu102
- Datasets 1.18.2.dev0
- Tokenizers 0.11.0
|
{"license": "apache-2.0", "tags": ["generated_from_trainer", "es", "robust-speech-event"], "datasets": ["common_voice"], "model-index": [{"name": "wav2vec2-large-xls-r-300m-spanish-large", "results": []}]}
|
glob-asr/wav2vec2-xls-r-300m-spanish-large-LM
| null |
[
"transformers",
"pytorch",
"safetensors",
"wav2vec2",
"automatic-speech-recognition",
"generated_from_trainer",
"es",
"robust-speech-event",
"dataset:common_voice",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# xls-r-es-test-lm
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - ES dataset.
It achieves the following results on the test set with lm model:
- Loss: 0.1304
- WER: 0.094
- CER: 0.031
It achieves the following results on the val set with lm model:
- Loss: 0.1304
- WER: 0.081
- CER: 0.025
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 7.5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- num_epochs: 10.0
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 2.9613 | 0.07 | 500 | 2.9647 | 1.0 |
| 2.604 | 0.14 | 1000 | 1.8300 | 0.9562 |
| 1.177 | 0.21 | 1500 | 0.3652 | 0.3077 |
| 1.0745 | 0.28 | 2000 | 0.2707 | 0.2504 |
| 1.0103 | 0.35 | 2500 | 0.2338 | 0.2157 |
| 0.9858 | 0.42 | 3000 | 0.2321 | 0.2129 |
| 0.974 | 0.49 | 3500 | 0.2164 | 0.2031 |
| 0.9699 | 0.56 | 4000 | 0.2078 | 0.1970 |
| 0.9513 | 0.63 | 4500 | 0.2173 | 0.2139 |
| 0.9657 | 0.7 | 5000 | 0.2050 | 0.1979 |
| 0.9484 | 0.77 | 5500 | 0.2008 | 0.1919 |
| 0.9317 | 0.84 | 6000 | 0.2012 | 0.1911 |
| 0.9366 | 0.91 | 6500 | 0.2024 | 0.1976 |
| 0.9242 | 0.98 | 7000 | 0.2062 | 0.2028 |
| 0.9138 | 1.05 | 7500 | 0.1924 | 0.1863 |
| 0.921 | 1.12 | 8000 | 0.1935 | 0.1836 |
| 0.9117 | 1.19 | 8500 | 0.1887 | 0.1815 |
| 0.9064 | 1.26 | 9000 | 0.1909 | 0.1839 |
| 0.9118 | 1.32 | 9500 | 0.1869 | 0.1830 |
| 0.9121 | 1.39 | 10000 | 0.1863 | 0.1802 |
| 0.9048 | 1.46 | 10500 | 0.1845 | 0.1791 |
| 0.8955 | 1.53 | 11000 | 0.1863 | 0.1774 |
| 0.8947 | 1.6 | 11500 | 0.1907 | 0.1814 |
| 0.9073 | 1.67 | 12000 | 0.1892 | 0.1853 |
| 0.8927 | 1.74 | 12500 | 0.1821 | 0.1750 |
| 0.8732 | 1.81 | 13000 | 0.1815 | 0.1768 |
| 0.8761 | 1.88 | 13500 | 0.1822 | 0.1749 |
| 0.8751 | 1.95 | 14000 | 0.1789 | 0.1715 |
| 0.8889 | 2.02 | 14500 | 0.1819 | 0.1791 |
| 0.8864 | 2.09 | 15000 | 0.1826 | 0.1794 |
| 0.886 | 2.16 | 15500 | 0.1788 | 0.1776 |
| 0.8915 | 2.23 | 16000 | 0.1756 | 0.1719 |
| 0.8689 | 2.3 | 16500 | 0.1769 | 0.1711 |
| 0.879 | 2.37 | 17000 | 0.1777 | 0.1739 |
| 0.8692 | 2.44 | 17500 | 0.1765 | 0.1705 |
| 0.8504 | 2.51 | 18000 | 0.1699 | 0.1652 |
| 0.8728 | 2.58 | 18500 | 0.1705 | 0.1694 |
| 0.8523 | 2.65 | 19000 | 0.1674 | 0.1645 |
| 0.8513 | 2.72 | 19500 | 0.1661 | 0.1611 |
| 0.8498 | 2.79 | 20000 | 0.1660 | 0.1631 |
| 0.8432 | 2.86 | 20500 | 0.1636 | 0.1610 |
| 0.8492 | 2.93 | 21000 | 0.1708 | 0.1688 |
| 0.8561 | 3.0 | 21500 | 0.1663 | 0.1604 |
| 0.842 | 3.07 | 22000 | 0.1690 | 0.1625 |
| 0.857 | 3.14 | 22500 | 0.1642 | 0.1605 |
| 0.8518 | 3.21 | 23000 | 0.1626 | 0.1585 |
| 0.8506 | 3.28 | 23500 | 0.1651 | 0.1605 |
| 0.8394 | 3.35 | 24000 | 0.1647 | 0.1585 |
| 0.8431 | 3.42 | 24500 | 0.1632 | 0.1573 |
| 0.8566 | 3.49 | 25000 | 0.1614 | 0.1550 |
| 0.8534 | 3.56 | 25500 | 0.1645 | 0.1589 |
| 0.8386 | 3.63 | 26000 | 0.1632 | 0.1582 |
| 0.8357 | 3.7 | 26500 | 0.1631 | 0.1556 |
| 0.8299 | 3.77 | 27000 | 0.1612 | 0.1550 |
| 0.8421 | 3.84 | 27500 | 0.1602 | 0.1552 |
| 0.8375 | 3.91 | 28000 | 0.1592 | 0.1537 |
| 0.8328 | 3.97 | 28500 | 0.1587 | 0.1537 |
| 0.8155 | 4.04 | 29000 | 0.1587 | 0.1520 |
| 0.8335 | 4.11 | 29500 | 0.1624 | 0.1556 |
| 0.8138 | 4.18 | 30000 | 0.1581 | 0.1547 |
| 0.8195 | 4.25 | 30500 | 0.1560 | 0.1507 |
| 0.8092 | 4.32 | 31000 | 0.1561 | 0.1534 |
| 0.8191 | 4.39 | 31500 | 0.1549 | 0.1493 |
| 0.8008 | 4.46 | 32000 | 0.1540 | 0.1493 |
| 0.8138 | 4.53 | 32500 | 0.1544 | 0.1493 |
| 0.8173 | 4.6 | 33000 | 0.1553 | 0.1511 |
| 0.8081 | 4.67 | 33500 | 0.1541 | 0.1484 |
| 0.8192 | 4.74 | 34000 | 0.1560 | 0.1506 |
| 0.8068 | 4.81 | 34500 | 0.1540 | 0.1503 |
| 0.8105 | 4.88 | 35000 | 0.1529 | 0.1483 |
| 0.7976 | 4.95 | 35500 | 0.1507 | 0.1451 |
| 0.8143 | 5.02 | 36000 | 0.1505 | 0.1462 |
| 0.8053 | 5.09 | 36500 | 0.1517 | 0.1476 |
| 0.785 | 5.16 | 37000 | 0.1526 | 0.1478 |
| 0.7936 | 5.23 | 37500 | 0.1489 | 0.1421 |
| 0.807 | 5.3 | 38000 | 0.1483 | 0.1420 |
| 0.8092 | 5.37 | 38500 | 0.1481 | 0.1435 |
| 0.793 | 5.44 | 39000 | 0.1503 | 0.1438 |
| 0.814 | 5.51 | 39500 | 0.1495 | 0.1480 |
| 0.807 | 5.58 | 40000 | 0.1472 | 0.1424 |
| 0.7913 | 5.65 | 40500 | 0.1471 | 0.1422 |
| 0.7844 | 5.72 | 41000 | 0.1473 | 0.1422 |
| 0.7888 | 5.79 | 41500 | 0.1445 | 0.1385 |
| 0.7806 | 5.86 | 42000 | 0.1435 | 0.1394 |
| 0.7773 | 5.93 | 42500 | 0.1461 | 0.1424 |
| 0.786 | 6.0 | 43000 | 0.1450 | 0.1413 |
| 0.7784 | 6.07 | 43500 | 0.1463 | 0.1424 |
| 0.7937 | 6.14 | 44000 | 0.1438 | 0.1386 |
| 0.7738 | 6.21 | 44500 | 0.1437 | 0.1383 |
| 0.7728 | 6.28 | 45000 | 0.1424 | 0.1371 |
| 0.7681 | 6.35 | 45500 | 0.1416 | 0.1376 |
| 0.776 | 6.42 | 46000 | 0.1415 | 0.1380 |
| 0.7773 | 6.49 | 46500 | 0.1416 | 0.1371 |
| 0.7692 | 6.56 | 47000 | 0.1398 | 0.1345 |
| 0.7642 | 6.62 | 47500 | 0.1381 | 0.1341 |
| 0.7692 | 6.69 | 48000 | 0.1392 | 0.1334 |
| 0.7667 | 6.76 | 48500 | 0.1392 | 0.1348 |
| 0.7712 | 6.83 | 49000 | 0.1398 | 0.1333 |
| 0.7628 | 6.9 | 49500 | 0.1392 | 0.1344 |
| 0.7622 | 6.97 | 50000 | 0.1377 | 0.1329 |
| 0.7639 | 7.04 | 50500 | 0.1361 | 0.1316 |
| 0.742 | 7.11 | 51000 | 0.1376 | 0.1327 |
| 0.7526 | 7.18 | 51500 | 0.1387 | 0.1342 |
| 0.7606 | 7.25 | 52000 | 0.1363 | 0.1316 |
| 0.7626 | 7.32 | 52500 | 0.1365 | 0.1313 |
| 0.752 | 7.39 | 53000 | 0.1354 | 0.1309 |
| 0.7562 | 7.46 | 53500 | 0.1362 | 0.1312 |
| 0.7557 | 7.53 | 54000 | 0.1358 | 0.1325 |
| 0.7588 | 7.6 | 54500 | 0.1343 | 0.1311 |
| 0.7485 | 7.67 | 55000 | 0.1346 | 0.1301 |
| 0.7466 | 7.74 | 55500 | 0.1354 | 0.1314 |
| 0.7558 | 7.81 | 56000 | 0.1359 | 0.1325 |
| 0.7578 | 7.88 | 56500 | 0.1363 | 0.1334 |
| 0.7411 | 7.95 | 57000 | 0.1346 | 0.1301 |
| 0.7478 | 8.02 | 57500 | 0.1355 | 0.1305 |
| 0.7451 | 8.09 | 58000 | 0.1349 | 0.1302 |
| 0.7383 | 8.16 | 58500 | 0.1349 | 0.1294 |
| 0.7482 | 8.23 | 59000 | 0.1341 | 0.1293 |
| 0.742 | 8.3 | 59500 | 0.1338 | 0.1296 |
| 0.7343 | 8.37 | 60000 | 0.1348 | 0.1307 |
| 0.7385 | 8.44 | 60500 | 0.1324 | 0.1282 |
| 0.7567 | 8.51 | 61000 | 0.1334 | 0.1281 |
| 0.7342 | 8.58 | 61500 | 0.1338 | 0.1289 |
| 0.7401 | 8.65 | 62000 | 0.1331 | 0.1285 |
| 0.7362 | 8.72 | 62500 | 0.1329 | 0.1283 |
| 0.7241 | 8.79 | 63000 | 0.1323 | 0.1277 |
| 0.7244 | 8.86 | 63500 | 0.1317 | 0.1269 |
| 0.7274 | 8.93 | 64000 | 0.1308 | 0.1260 |
| 0.7411 | 9.0 | 64500 | 0.1309 | 0.1256 |
| 0.7255 | 9.07 | 65000 | 0.1316 | 0.1265 |
| 0.7406 | 9.14 | 65500 | 0.1315 | 0.1270 |
| 0.7418 | 9.21 | 66000 | 0.1315 | 0.1269 |
| 0.7301 | 9.27 | 66500 | 0.1315 | 0.1273 |
| 0.7248 | 9.34 | 67000 | 0.1323 | 0.1274 |
| 0.7423 | 9.41 | 67500 | 0.1309 | 0.1267 |
| 0.7152 | 9.48 | 68000 | 0.1312 | 0.1271 |
| 0.7295 | 9.55 | 68500 | 0.1306 | 0.1262 |
| 0.7231 | 9.62 | 69000 | 0.1308 | 0.1263 |
| 0.7344 | 9.69 | 69500 | 0.1313 | 0.1267 |
| 0.7264 | 9.76 | 70000 | 0.1305 | 0.1263 |
| 0.7309 | 9.83 | 70500 | 0.1303 | 0.1262 |
| 0.73 | 9.9 | 71000 | 0.1303 | 0.1261 |
| 0.7353 | 9.97 | 71500 | 0.1304 | 0.1260 |
### Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu102
- Datasets 1.18.3
- Tokenizers 0.11.0
|
{"language": ["es"], "license": "apache-2.0", "tags": ["automatic-speech-recognition", "es", "generated_from_trainer", "hf-asr-leaderboard", "mozilla-foundation/common_voice_8_0", "robust-speech-event"], "datasets": ["mozilla-foundation/common_voice_8_0"], "model-index": [{"name": "xls-r-es-test-lm", "results": [{"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "Common Voice 8.0", "type": "mozilla-foundation/common_voice_8_0", "args": "es"}, "metrics": [{"type": "wer", "value": 9.4, "name": "Test WER"}]}, {"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "Robust Speech Event - Dev Data", "type": "speech-recognition-community-v2/dev_data", "args": "es"}, "metrics": [{"type": "wer", "value": 27.95, "name": "Test WER"}]}, {"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "Robust Speech Event - Test Data", "type": "speech-recognition-community-v2/eval_data", "args": "es"}, "metrics": [{"type": "wer", "value": 30.86, "name": "Test WER"}]}]}]}
|
glob-asr/xls-r-es-test-lm
| null |
[
"transformers",
"pytorch",
"wav2vec2",
"automatic-speech-recognition",
"es",
"generated_from_trainer",
"hf-asr-leaderboard",
"mozilla-foundation/common_voice_8_0",
"robust-speech-event",
"dataset:mozilla-foundation/common_voice_8_0",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null | null |
{}
|
gloukatou/bert-base-multilingual-uncased-finetuned-pos
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
null | null |
{}
|
gloukatou/distilbert-base-uncased-finetuned-pos
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
text-classification
|
transformers
|
{}
|
gmihaila/distilbert-base-uncased
| null |
[
"transformers",
"pytorch",
"distilbert",
"text-classification",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
automatic-speech-recognition
|
transformers
|
# Wav2Vec2-Large-XLSR-53-Romanian
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) in Romanian using the [Common Voice](https://huggingface.co/datasets/common_voice)
When using this model, make sure that your speech input is sampled at 16kHz.
## Usage
The model can be used directly (without a language model) as follows:
```python
import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
test_dataset = load_dataset("common_voice", "ro", split="test[:2%]").
processor = Wav2Vec2Processor.from_pretrained("gmihaila/wav2vec2-large-xlsr-53-romanian")
model = Wav2Vec2ForCTC.from_pretrained("gmihaila/wav2vec2-large-xlsr-53-romanian")
resampler = torchaudio.transforms.Resample(48_000, 16_000)
# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
\\\\tspeech_array, sampling_rate = torchaudio.load(batch["path"])
\\\\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
\\\\treturn batch
test_dataset = test_dataset.map(speech_file_to_array_fn)
inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)
with torch.no_grad():
\\\\tlogits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
predicted_ids = torch.argmax(logits, dim=-1)
print("Prediction:", processor.batch_decode(predicted_ids))
print("Reference:", test_dataset["sentence"][:2])
```
## Evaluation
The model can be evaluated as follows on the {language} test data of Common Voice.
```python
import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import re
test_dataset = load_dataset("common_voice", "ro", split="test")
wer = load_metric("wer")
processor = Wav2Vec2Processor.from_pretrained("gmihaila/wav2vec2-large-xlsr-53-romanian")
model = Wav2Vec2ForCTC.from_pretrained("gmihaila/wav2vec2-large-xlsr-53-romanian")
model.to("cuda")
chars_to_ignore_regex = '[\\\\\\\\,\\\\\\\\?\\\\\\\\.\\\\\\\\!\\\\\\\\-\\\\\\\\;\\\\\\\\:\\\\\\\\"\\\\\\\\“]'
resampler = torchaudio.transforms.Resample(48_000, 16_000)
# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
\\\\tbatch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
\\\\tspeech_array, sampling_rate = torchaudio.load(batch["path"])
\\\\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
\\\\treturn batch
test_dataset = test_dataset.map(speech_file_to_array_fn)
# Preprocessing the datasets.
# We need to read the aduio files as arrays
def evaluate(batch):
\\\\tinputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
\\\\twith torch.no_grad():
\\\\t\\\\tlogits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
pred_ids = torch.argmax(logits, dim=-1)
\\\\tbatch["pred_strings"] = processor.batch_decode(pred_ids)
\\\\treturn batch
result = test_dataset.map(evaluate, batched=True, batch_size=8)
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
```
**Test Result**: 28.43 %
## Training
The Common Voice `train`, `validation` datasets were used for training.
The script used for training can be found [here](https://colab.research.google.com/github/gmihaila/ml_things/blob/master/notebooks/pytorch/RO_Fine_Tune_XLSR_Wav2Vec2_on_Turkish_ASR_with_🤗_Transformers.ipynb)
|
{"language": "ro", "license": "apache-2.0", "tags": ["audio", "automatic-speech-recognition", "speech", "xlsr-fine-tuning-week"], "datasets": ["common_voice"], "base_model": "facebook/wav2vec2-large-xlsr-53", "model-index": [{"name": "XLSR Wav2Vec2 Romanian by George Mihaila", "results": [{"task": {"type": "automatic-speech-recognition", "name": "Speech Recognition"}, "dataset": {"name": "Common Voice ro", "type": "common_voice", "args": "ro"}, "metrics": [{"type": "wer", "value": 28.4, "name": "Test WER"}]}]}]}
|
gmihaila/wav2vec2-large-xlsr-53-romanian
| null |
[
"transformers",
"pytorch",
"jax",
"wav2vec2",
"automatic-speech-recognition",
"audio",
"speech",
"xlsr-fine-tuning-week",
"ro",
"dataset:common_voice",
"base_model:facebook/wav2vec2-large-xlsr-53",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null | null |
{}
|
gngdb/distilgpt2-finetuned-wikitext2
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
fill-mask
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# BERiTmodel2
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.1508
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 280
- num_epochs: 10
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 3.1924 | 1.0 | 2854 | 3.4329 |
| 3.0936 | 2.0 | 5708 | 3.5036 |
| 2.9998 | 3.0 | 8562 | 3.1906 |
| 2.9064 | 4.0 | 11416 | 3.4867 |
| 2.8493 | 5.0 | 14270 | 3.2027 |
| 2.7538 | 6.0 | 17124 | 2.9772 |
| 2.7273 | 7.0 | 19978 | 2.9950 |
| 2.7399 | 8.0 | 22832 | 2.9690 |
| 2.67 | 9.0 | 25686 | 3.0311 |
| 2.6388 | 10.0 | 28540 | 3.1508 |
### Framework versions
- Transformers 4.14.1
- Pytorch 1.10.0+cu111
- Datasets 1.17.0
- Tokenizers 0.10.3
|
{"license": "mit", "tags": ["generated_from_trainer"], "model-index": [{"name": "BERiTmodel2", "results": []}]}
|
gngpostalsrvc/BERiTmodel2
| null |
[
"transformers",
"pytorch",
"tensorboard",
"roberta",
"fill-mask",
"generated_from_trainer",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# mt5-small-finetuned-xsum
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on the xsum dataset.
It achieves the following results on the evaluation set:
- Loss: nan
- Rouge1: 2.8351
- Rouge2: 0.3143
- Rougel: 2.6488
- Rougelsum: 2.6463
- Gen Len: 4.9416
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
| nan | 1.0 | 12753 | nan | 2.8351 | 0.3143 | 2.6488 | 2.6463 | 4.9416 |
### Framework versions
- Transformers 4.10.2
- Pytorch 1.9.0+cu102
- Datasets 1.12.1
- Tokenizers 0.10.3
|
{"license": "apache-2.0", "tags": ["generated_from_trainer"], "datasets": ["xsum"], "metrics": ["rouge"], "model-index": [{"name": "mt5-small-finetuned-xsum", "results": [{"task": {"type": "text2text-generation", "name": "Sequence-to-sequence Language Modeling"}, "dataset": {"name": "xsum", "type": "xsum", "args": "default"}, "metrics": [{"type": "rouge", "value": 2.8351, "name": "Rouge1"}]}]}]}
|
gniemiec/mt5-small-finetuned-xsum
| null |
[
"transformers",
"pytorch",
"tensorboard",
"mt5",
"text2text-generation",
"generated_from_trainer",
"dataset:xsum",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# t5-small-finetuned-xsum
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the xsum dataset.
It achieves the following results on the evaluation set:
- Loss: 2.7967
- Rouge1: 23.0533
- Rouge2: 3.912
- Rougel: 17.8534
- Rougelsum: 17.8581
- Gen Len: 18.6878
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|
| 3.0574 | 1.0 | 1276 | 2.7967 | 23.0533 | 3.912 | 17.8534 | 17.8581 | 18.6878 |
### Framework versions
- Transformers 4.10.2
- Pytorch 1.9.0+cu102
- Datasets 1.12.1
- Tokenizers 0.10.3
|
{"license": "apache-2.0", "tags": ["generated_from_trainer"], "datasets": ["xsum"], "metrics": ["rouge"], "model-index": [{"name": "t5-small-finetuned-xsum", "results": [{"task": {"type": "text2text-generation", "name": "Sequence-to-sequence Language Modeling"}, "dataset": {"name": "xsum", "type": "xsum", "args": "default"}, "metrics": [{"type": "rouge", "value": 23.0533, "name": "Rouge1"}]}]}]}
|
gniemiec/t5-small-finetuned-xsum
| null |
[
"transformers",
"pytorch",
"tensorboard",
"t5",
"text2text-generation",
"generated_from_trainer",
"dataset:xsum",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# diam
Autogenerated by HuggingPics🤗🖼️
Create your own image classifier for **anything** by running [the demo on Google Colab](https://colab.research.google.com/github/nateraw/huggingpics/blob/main/HuggingPics.ipynb).
Report any issues with the demo at the [github repo](https://github.com/nateraw/huggingpics).
## Example Images
#### bunny
#### moon
#### sun
#### tiger
|
{"tags": ["image-classification", "pytorch", "huggingpics"], "metrics": ["accuracy"]}
|
godiec/diam
| null |
[
"transformers",
"pytorch",
"tensorboard",
"vit",
"image-classification",
"huggingpics",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
## KoBART-base-v1
```python
from transformers import PreTrainedTokenizerFast, BartModel
tokenizer = PreTrainedTokenizerFast.from_pretrained('gogamza/kobart-base-v1')
model = BartModel.from_pretrained('gogamza/kobart-base-v1')
```
|
{"language": "ko", "license": "mit", "tags": ["bart"]}
|
gogamza/kobart-base-v1
| null |
[
"transformers",
"pytorch",
"safetensors",
"bart",
"feature-extraction",
"ko",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Model Card for kobart-base-v2
# Model Details
## Model Description
[**BART**](https://arxiv.org/pdf/1910.13461.pdf)(**B**idirectional and **A**uto-**R**egressive **T**ransformers)는 입력 텍스트 일부에 노이즈를 추가하여 이를 다시 원문으로 복구하는 `autoencoder`의 형태로 학습이 됩니다. 한국어 BART(이하 **KoBART**) 는 논문에서 사용된 `Text Infilling` 노이즈 함수를 사용하여 **40GB** 이상의 한국어 텍스트에 대해서 학습한 한국어 `encoder-decoder` 언어 모델입니다. 이를 통해 도출된 `KoBART-base`를 배포합니다.
- **Developed by:** More information needed
- **Shared by [Optional]:** Heewon(Haven) Jeon
- **Model type:** Feature Extraction
- **Language(s) (NLP):** Korean
- **License:** MIT
- **Parent Model:** BART
- **Resources for more information:**
- [GitHub Repo](https://github.com/haven-jeon/KoBART)
- [Model Demo Space](https://huggingface.co/spaces/gogamza/kobart-summarization)
# Uses
## Direct Use
This model can be used for the task of Feature Extraction.
## Downstream Use [Optional]
More information needed.
## Out-of-Scope Use
The model should not be used to intentionally create hostile or alienating environments for people.
# Bias, Risks, and Limitations
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Recommendations
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
# Training Details
## Training Data
| Data | # of Sentences |
|-------|---------------:|
| Korean Wiki | 5M |
| Other corpus | 0.27B |
한국어 위키 백과 이외, 뉴스, 책, [모두의 말뭉치 v1.0(대화, 뉴스, ...)](https://corpus.korean.go.kr/), [청와대 국민청원](https://github.com/akngs/petitions) 등의 다양한 데이터가 모델 학습에 사용되었습니다.
`vocab` 사이즈는 30,000 이며 대화에 자주 쓰이는 아래와 같은 이모티콘, 이모지 등을 추가하여 해당 토큰의 인식 능력을 올렸습니다.
> 😀, 😁, 😆, 😅, 🤣, .. , `:-)`, `:)`, `-)`, `(-:`...
## Training Procedure
### Tokenizer
[`tokenizers`](https://github.com/huggingface/tokenizers) 패키지의 `Character BPE tokenizer`로 학습되었습니다.
### Speeds, Sizes, Times
| Model | # of params | Type | # of layers | # of heads | ffn_dim | hidden_dims |
|--------------|:----:|:-------:|--------:|--------:|--------:|--------------:|
| `KoBART-base` | 124M | Encoder | 6 | 16 | 3072 | 768 |
| | | Decoder | 6 | 16 | 3072 | 768 |
# Evaluation
## Testing Data, Factors & Metrics
### Testing Data
More information needed
### Factors
More information needed
### Metrics
More information needed
## Results
NSMC
- acc. : 0.901
The model authors also note in the [GitHub Repo](https://github.com/haven-jeon/KoBART):
| | [NSMC](https://github.com/e9t/nsmc)(acc) | [KorSTS](https://github.com/kakaobrain/KorNLUDatasets)(spearman) | [Question Pair](https://github.com/aisolab/nlp_classification/tree/master/BERT_pairwise_text_classification/qpair)(acc) |
|---|---|---|---|
| **KoBART-base** | 90.24 | 81.66 | 94.34 |
# Model Examination
More information needed
# Environmental Impact
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** More information needed
- **Hours used:** More information needed
- **Cloud Provider:** More information needed
- **Compute Region:** More information needed
- **Carbon Emitted:** More information needed
# Technical Specifications [optional]
## Model Architecture and Objective
More information needed
## Compute Infrastructure
More information needed
### Hardware
More information needed
### Software
More information needed.
# Citation
**BibTeX:**
More information needed.
# Glossary [optional]
More information needed
# More Information [optional]
More information needed
# Model Card Authors [optional]
Heewon(Haven) Jeon in collaboration with Ezi Ozoani and the Hugging Face team
# Model Card Contact
The model authors note in the [GitHub Repo](https://github.com/haven-jeon/KoBART):
`KoBART` 관련 이슈는 [이곳](https://github.com/SKT-AI/KoBART/issues)에 올려주세요.
# How to Get Started with the Model
Use the code below to get started with the model.
<details>
<summary> Click to expand </summary>
```python
from transformers import PreTrainedTokenizerFast, BartModel
tokenizer = PreTrainedTokenizerFast.from_pretrained('gogamza/kobart-base-v2')
model = BartModel.from_pretrained('gogamza/kobart-base-v2')
```
</details>
|
{"language": "ko", "license": "mit", "tags": ["bart"]}
|
gogamza/kobart-base-v2
| null |
[
"transformers",
"pytorch",
"safetensors",
"bart",
"feature-extraction",
"ko",
"arxiv:1910.13461",
"arxiv:1910.09700",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# Korean News Summarization Model
## Demo
https://huggingface.co/spaces/gogamza/kobart-summarization
## How to use
```python
import torch
from transformers import PreTrainedTokenizerFast
from transformers import BartForConditionalGeneration
tokenizer = PreTrainedTokenizerFast.from_pretrained('gogamza/kobart-summarization')
model = BartForConditionalGeneration.from_pretrained('gogamza/kobart-summarization')
text = "과거를 떠올려보자. 방송을 보던 우리의 모습을. 독보적인 매체는 TV였다. 온 가족이 둘러앉아 TV를 봤다. 간혹 가족들끼리 뉴스와 드라마, 예능 프로그램을 둘러싸고 리모컨 쟁탈전이 벌어지기도 했다. 각자 선호하는 프로그램을 ‘본방’으로 보기 위한 싸움이었다. TV가 한 대인지 두 대인지 여부도 그래서 중요했다. 지금은 어떤가. ‘안방극장’이라는 말은 옛말이 됐다. TV가 없는 집도 많다. 미디어의 혜 택을 누릴 수 있는 방법은 늘어났다. 각자의 방에서 각자의 휴대폰으로, 노트북으로, 태블릿으로 콘텐츠 를 즐긴다."
raw_input_ids = tokenizer.encode(text)
input_ids = [tokenizer.bos_token_id] + raw_input_ids + [tokenizer.eos_token_id]
summary_ids = model.generate(torch.tensor([input_ids]))
tokenizer.decode(summary_ids.squeeze().tolist(), skip_special_tokens=True)
```
|
{"language": "ko", "license": "mit", "tags": ["bart"]}
|
gogamza/kobart-summarization
| null |
[
"transformers",
"pytorch",
"safetensors",
"bart",
"text2text-generation",
"ko",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
Please refer : https://github.com/haven-jeon/LegalQA#train
|
{}
|
gogamza/kobert-legalqa-v1
| null |
[
"transformers",
"pytorch",
"bert",
"next-sentence-prediction",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
translation
|
transformers
|
Byt5-small-ain-jpn-mt is a machine translation model pretrained with [Google's ByT5-small](https://huggingface.co/google/byt5-small) and fine-tuned on bilingual datasets crawled from the Web. It translates Ainu language to Japanese.
|
{"language": ["ain", "ja"], "tags": ["translation"]}
|
Language-Media-Lab/byt5-small-ain-jpn-mt
| null |
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"translation",
"ain",
"ja",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
translation
|
transformers
|
Byt5-small-jpn-ain-mt is a machine translation model pretrained with [Google's ByT5-small](https://huggingface.co/google/byt5-small) and fine-tuned on bilingual datasets crawled from the Web. It translates Japanese to Ainu language.
|
{"language": ["jpn", "ain"], "tags": ["translation"]}
|
Language-Media-Lab/byt5-small-jpn-ain-mt
| null |
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"translation",
"jpn",
"ain",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
translation
|
transformers
|
mt5-small-ain-jpn-mt is a machine translation model pretrained with [Google's mT5-small](https://huggingface.co/google/mt5-small) and fine-tuned on bilingual datasets crawled from the Web. It translates Ainu language to Japanese.
|
{"language": ["jpn", "ain"], "tags": ["translation"]}
|
Language-Media-Lab/mt5-small-ain-jpn-mt
| null |
[
"transformers",
"pytorch",
"mt5",
"text2text-generation",
"translation",
"jpn",
"ain",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
translation
|
transformers
|
mt5-small-jpn-ain-mt is a machine translation model pretrained with [Google's mT5-small](https://huggingface.co/google/mt5-small) and fine-tuned on bilingual datasets crawled from the Web. It translates Japanese to Ainu language.
|
{"language": ["jpn", "ain"], "tags": ["translation"]}
|
Language-Media-Lab/mt5-small-jpn-ain-mt
| null |
[
"transformers",
"pytorch",
"mt5",
"text2text-generation",
"translation",
"jpn",
"ain",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
question-answering
|
transformers
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-squad
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the squad dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Framework versions
- Transformers 4.11.0
- Pytorch 1.9.0+cu102
- Datasets 1.12.1
- Tokenizers 0.10.3
|
{"license": "apache-2.0", "tags": ["generated_from_trainer"], "datasets": ["squad"], "model-index": [{"name": "distilbert-base-uncased-finetuned-squad", "results": []}]}
|
gokulkarthik/distilbert-base-uncased-finetuned-squad
| null |
[
"transformers",
"pytorch",
"tensorboard",
"distilbert",
"question-answering",
"generated_from_trainer",
"dataset:squad",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
question-answering
|
transformers
|
# XLM-RoBERTa for question answering in Indian languages
pre-trained XLM-Roberta with intermediate pre-training on SQUAD dataset (English) and fine tuning on Chaii dataset (Tamil, Hindi)
# How to use from the 🤗/transformers library
```
from transformers import AutoTokenizer, AutoModelForQuestionAnswering
tokenizer = AutoTokenizer.from_pretrained("gokulkarthik/xlm-roberta-qa-chaii")
model = AutoModelForQuestionAnswering.from_pretrained("gokulkarthik/xlm-roberta-qa-chaii")
```
|
{"language": ["en", "ta", "hi"], "datasets": ["squad", "chaii"], "widget": [{"text": "\u0b85\u0bb2\u0bc1\u0bae\u0bbf\u0ba9\u0bbf\u0baf\u0ba4\u0bcd\u0ba4\u0bbf\u0ba9\u0bcd \u0b85\u0ba3\u0bc1 \u0b8e\u0ba3\u0bcd \u0b8e\u0ba9\u0bcd\u0ba9?", "context": "\u0b85\u0bb2\u0bc1\u0bae\u0bbf\u0ba9\u0bbf\u0baf\u0bae\u0bcd (\u0b86\u0b99\u0bcd\u0b95\u0bbf\u0bb2\u0bae\u0bcd: \u0b85\u0bb2\u0bc1\u0bae\u0bbf\u0ba9\u0bbf\u0baf\u0bae\u0bcd; \u0bb5\u0b9f \u0b85\u0bae\u0bc6\u0bb0\u0bbf\u0b95\u0bcd\u0b95 \u0b86\u0b99\u0bcd\u0b95\u0bbf\u0bb2\u0bae\u0bcd: Aluminum) \u0b92\u0bb0\u0bc1 \u0bb5\u0bc7\u0ba4\u0bbf\u0baf\u0bbf\u0baf\u0bb2\u0bcd \u0ba4\u0ba9\u0bbf\u0bae\u0bae\u0bcd \u0b86\u0b95\u0bc1\u0bae\u0bcd. \u0b87\u0ba4\u0ba9\u0bc1\u0b9f\u0bc8\u0baf \u0b85\u0ba3\u0bc1 \u0b8e\u0ba3\u0bcd 13 \u0b86\u0b95\u0bc1\u0bae\u0bcd. \u0b87\u0ba4\u0bc1 \u0baa\u0bc2\u0bae\u0bbf\u0baf\u0bbf\u0bb2\u0bcd \u0b85\u0ba4\u0bbf\u0b95\u0bae\u0bcd \u0b95\u0bbf\u0b9f\u0bc8\u0b95\u0bcd\u0b95\u0bc1\u0bae\u0bcd \u0b89\u0bb2\u0bcb\u0b95\u0b99\u0bcd\u0b95\u0bb3\u0bc1\u0bb3\u0bcd \u0b92\u0ba9\u0bcd\u0bb1\u0bc1. \u0b87\u0ba4\u0bc1 \u0bae\u0bbf\u0ba9\u0bcd\u0b9a\u0bbe\u0bb0\u0ba4\u0bcd\u0ba4\u0bc8\u0baf\u0bc1\u0bae\u0bcd \u0bb5\u0bc6\u0baa\u0bcd\u0baa\u0ba4\u0bcd\u0ba4\u0bc8\u0baf\u0bc1\u0bae\u0bcd \u0b95\u0b9f\u0ba4\u0bcd\u0ba4 \u0bb5\u0bb2\u0bcd\u0bb2\u0ba4\u0bc1. \u0baa\u0bbe\u0b95\u0bcd\u0bb8\u0bc8\u0b9f\u0bcd \u0b8e\u0ba9\u0bcd\u0bb1 \u0ba4\u0bbe\u0ba4\u0bc1\u0bb5\u0bbf\u0bb2\u0bcd \u0b87\u0bb0\u0bc1\u0ba8\u0bcd\u0ba4\u0bc1 \u0b85\u0bb2\u0bc1\u0bae\u0bbf\u0ba9\u0bbf\u0baf\u0bae\u0bcd \u0ba4\u0baf\u0bbe\u0bb0\u0bbf\u0b95\u0bcd\u0b95\u0baa\u0bcd\u0baa\u0b9f\u0bc1\u0b95\u0bbf\u0bb1\u0ba4\u0bc1. \u0b87\u0ba4\u0ba9\u0bcd \u0bb5\u0bc7\u0ba4\u0bbf\u0b95\u0bcd\u0b95\u0bc1\u0bb1\u0bbf\u0baf\u0bc0\u0b9f\u0bc1 Al \u0b86\u0b95\u0bc1\u0bae\u0bcd."}, {"text": "\u091c\u094d\u0935\u093e\u0932\u093e \u0917\u0941\u091f\u094d\u091f\u093e \u0915\u0940 \u092e\u093e\u0901 \u0915\u093e \u0928\u093e\u092e \u0915\u094d\u092f\u093e \u0939\u0948?", "context": "\u091c\u094d\u0935\u093e\u0932\u093e \u0917\u0941\u091f\u094d\u091f\u093e (\u091c\u0928\u094d\u092e: 7 \u0938\u093f\u0924\u0902\u092c\u0930 1983; \u0935\u0930\u094d\u0927\u093e, \u092e\u0939\u093e\u0930\u093e\u0937\u094d\u091f\u094d\u0930) \u090f\u0915 \u092d\u093e\u0930\u0924\u0940\u092f \u092c\u0948\u0921\u092e\u093f\u0902\u091f\u0928 \u0916\u093f\u0932\u093e\u0921\u0940 \u0939\u0948\u0902\u0964 \u092a\u094d\u0930\u093e\u0930\u0902\u092d\u093f\u0915 \u091c\u0940\u0935\u0928 \u091c\u094d\u0935\u093e\u0932\u093e \u0917\u0941\u091f\u094d\u091f\u093e \u0915\u093e \u091c\u0928\u094d\u092e 7 \u0938\u093f\u0924\u0902\u092c\u0930 1983 \u0915\u094b \u0935\u0930\u094d\u0927\u093e, \u092e\u0939\u093e\u0930\u093e\u0937\u094d\u091f\u094d\u0930 \u092e\u0947\u0902 \u0939\u0941\u0906 \u0925\u093e\u0964 \u0909\u0928\u0915\u0947 \u092a\u093f\u0924\u093e \u090f\u092e. \u0915\u094d\u0930\u093e\u0902\u0924\u093f \u0924\u0947\u0932\u0941\u0917\u0941 \u0914\u0930 \u092e\u093e\u0902 \u092f\u0947\u0932\u0928 \u091a\u0940\u0928 \u0938\u0947 \u0939\u0948\u0902\u0964 \u0909\u0928\u0915\u0940 \u092e\u093e\u0902 \u092f\u0947\u0932\u0928 \u0917\u0941\u091f\u094d\u091f\u093e \u092a\u0939\u0932\u0940 \u092c\u093e\u0930 1977 \u092e\u0947\u0902 \u0905\u092a\u0928\u0947 \u0926\u093e\u0926\u093e \u091c\u0940 \u0915\u0947 \u0938\u093e\u0925 \u092d\u093e\u0930\u0924 \u0906\u0908 \u0925\u0940\u0902\u0964 \u091c\u094d\u0935\u093e\u0932\u093e \u0917\u0941\u091f\u094d\u091f\u093e \u0915\u0940 \u092a\u094d\u0930\u093e\u0930\u0902\u092d\u093f\u0915 \u092a\u0922\u093c\u093e\u0908 \u0939\u0948\u0926\u0930\u093e\u092c\u093e\u0926 \u0938\u0947 \u0939\u0941\u0908 \u0914\u0930 \u092f\u0939\u0940\u0902 \u0938\u0947 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u092c\u0948\u0921\u092e\u093f\u0902\u091f\u0928 \u0916\u0947\u0932\u0928\u093e \u092d\u0940 \u0936\u0941\u0930\u0942 \u0915\u093f\u092f\u093e\u0964 \u0915\u0945\u0930\u093f\u092f\u0930 10 \u0938\u093e\u0932 \u0915\u0940 \u0909\u092e\u094d\u0930 \u0938\u0947 \u0939\u0940 \u091c\u094d\u0935\u093e\u0932\u093e \u0917\u0941\u091f\u094d\u091f\u093e \u0928\u0947 \u090f\u0938.\u090f\u092e. \u0906\u0930\u093f\u092b \u0938\u0947 \u091f\u094d\u0930\u0947\u0928\u093f\u0902\u0917 \u0932\u0947\u0928\u093e \u0936\u0941\u0930\u0942 \u0915\u0930 \u0926\u093f\u092f\u093e \u0925\u093e\u0964 \u090f\u0938.\u090f\u092e. \u0906\u0930\u093f\u092b \u092d\u093e\u0930\u0924 \u0915\u0947 \u091c\u093e\u0928\u0947 \u092e\u093e\u0928\u0947 \u0916\u0947\u0932 \u092a\u094d\u0930\u0936\u093f\u0915\u094d\u0937\u0915 \u0939\u0948\u0902 \u091c\u093f\u0928\u094d\u0939\u0947\u0902 \u0926\u094d\u0930\u094b\u0923\u093e\u091a\u093e\u0930\u094d\u092f \u0905\u0935\u093e\u0930\u094d\u0921 \u0938\u0947 \u0938\u092e\u094d\u092e\u093e\u0928\u093f\u0924 \u0915\u093f\u092f\u093e \u0917\u092f\u093e \u0939\u0948\u0964 \u092a\u0939\u0932\u0940 \u092c\u093e\u0930 13 \u0938\u093e\u0932 \u0915\u0940 \u0909\u092e\u094d\u0930 \u092e\u0947\u0902 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u092e\u093f\u0928\u0940 \u0928\u0947\u0936\u0928\u0932 \u092c\u0948\u0921\u092e\u093f\u0902\u091f\u0928 \u091a\u0948\u0902\u092a\u093f\u092f\u0928\u0936\u093f\u092a \u091c\u0940\u0924\u0940 \u0925\u0940\u0964 \u0938\u093e\u0932 2000 \u092e\u0947\u0902 \u091c\u094d\u0935\u093e\u0932\u093e \u0917\u0941\u091f\u094d\u091f\u093e \u0928\u0947 17 \u0938\u093e\u0932 \u0915\u0940 \u0909\u092e\u094d\u0930 \u092e\u0947\u0902 \u091c\u0942\u0928\u093f\u092f\u0930 \u0928\u0947\u0936\u0928\u0932 \u092c\u0948\u0921\u092e\u093f\u0902\u091f\u0928 \u091a\u0948\u0902\u092a\u093f\u092f\u0928\u0936\u093f\u092a \u091c\u0940\u0924\u0940\u0964 \u0907\u0938\u0940 \u0938\u093e\u0932 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0936\u094d\u0930\u0941\u0924\u093f \u0915\u0941\u0930\u093f\u092f\u0928 \u0915\u0947 \u0938\u093e\u0925 \u0921\u092c\u0932\u094d\u0938 \u092e\u0947\u0902 \u091c\u094b\u0921\u093c\u0940 \u092c\u0928\u093e\u0924\u0947 \u0939\u0941\u090f \u092e\u0939\u093f\u0932\u093e\u0913\u0902 \u0915\u0947 \u0921\u092c\u0932\u094d\u0938 \u091c\u0942\u0928\u093f\u092f\u0930 \u0928\u0947\u0936\u0928\u0932 \u092c\u0948\u0921\u092e\u093f\u0902\u091f\u0928 \u091a\u0948\u0902\u092a\u093f\u092f\u0928\u0936\u093f\u092a \u0914\u0930 \u0938\u0940\u0928\u093f\u092f\u0930 \u0928\u0947\u0936\u0928\u0932 \u092c\u0948\u0921\u092e\u093f\u0902\u091f\u0928 \u091a\u0948\u0902\u092a\u093f\u092f\u0928\u0936\u093f\u092a \u092e\u0947\u0902 \u091c\u0940\u0924 \u0939\u093e\u0938\u093f\u0932 \u0915\u0940\u0964 \u0936\u094d\u0930\u0941\u0924\u093f \u0915\u0941\u0930\u093f\u092f\u0928 \u0915\u0947 \u0938\u093e\u0925 \u0909\u0928\u0915\u0940 \u091c\u094b\u0921\u093c\u0940 \u0915\u093e\u092b\u0940 \u0932\u0902\u092c\u0947 \u0938\u092e\u092f \u0924\u0915 \u091a\u0932\u0940\u0964 2002 \u0938\u0947 2008 \u0924\u0915 \u0932\u0917\u093e\u0924\u093e\u0930 \u0938\u093e\u0924 \u092c\u093e\u0930 \u091c\u094d\u0935\u093e\u0932\u093e \u0917\u0941\u091f\u094d\u091f\u093e \u0928\u0947 \u092e\u0939\u093f\u0932\u093e\u0913\u0902 \u0915\u0947 \u0928\u0947\u0936\u0928\u0932 \u092f\u0941\u0917\u0932 \u092a\u094d\u0930\u0924\u093f\u092f\u094b\u0917\u093f\u0924\u093e \u092e\u0947\u0902 \u091c\u0940\u0924 \u0939\u093e\u0938\u093f\u0932 \u0915\u0940\u0964"}, {"text": "How many bones do you have in your body?", "context": "A normal adult human skeleton consists of the following 206 (208 if the breast is thought to be three parts). This number can vary depending on the physiological differences. For example, in a very small number of humans, an extra rib (neck) or an extra lower spinal cord is found. There are 22 bones in the human skull (excluding the ear tendons), which are divided into eight cranium bones and 14 facial bones. (Thick numbers indicate the numbers seen in the nearby picture.) Bones (8) 1 frontal bone (2) 3 temporal bone (2) 4 occipital bone (4) Sphinoid bone (14) 7 mandible (6) maxilla (2) palatine bone (2) 5 zygotic bone (9) 9 nasal bone (2) The sacral vertebrae (4 or 5), in adults, form the sacral vertebrae (3 to 5), in adults they form the valve."}]}
|
gokulkarthik/xlm-roberta-qa-chaii
| null |
[
"transformers",
"pytorch",
"xlm-roberta",
"question-answering",
"en",
"ta",
"hi",
"dataset:squad",
"dataset:chaii",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null | null |
{}
|
gokulkarthik/xlm-roberta-qa-dravidian
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
null | null |
{}
|
gokuls/wav2vec2-base-finetuned-ic-action
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
null | null |
{}
|
gokulsjain/bert
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
text2text-generation
|
transformers
|
# rachael-scai
Generation model (Pegasus fine-tuned with QReCC) used in the participation of group Rachael for SCAI 2021.
GitHub repository can be found in: [gonced8/rachael-scai](https://github.com/gonced8/rachael-scai)
Gonçalo Raposo
## Cite
```bibtex
@InProceedings{Raposo2022,
author = {Gonçalo Raposo and Rui Ribeiro and Bruno Martins and Luísa Coheur},
booktitle = {44th European Conference on Information Retrieval},
title = {Question rewriting? Assessing its importance for conversational question answering},
year = {2022},
month = apr,
note = {This version of the contribution has been accepted for publication, after peer review but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/[not yet available]. Use of this Accepted Version is subject to the publisher’s Accepted Manuscript terms of use \url{https://www.springernature.com/gp/open-research/policies/accepted-manuscript-terms}},
abstract = {In conversational question answering, systems must correctly interpret the interconnected interactions and generate knowledgeable answers, which may require the retrieval of relevant information from a background repository. Recent approaches to this problem leverage neural language models, although different alternatives can be considered in terms of modules for (a) representing user questions in context, (b) retrieving the relevant background information, and (c) generating the answer. This work presents a conversational question answering system designed specifically for the Search-Oriented Conversational AI (SCAI) shared task, and reports on a detailed analysis of its question rewriting module. In particular, we considered different variations of the question rewriting module to evaluate the influence on the subsequent components, and performed a careful analysis of the results obtained with the best system configuration. Our system achieved the best performance in the shared task and our analysis emphasizes the importance of the conversation context representation for the overall system performance.},
keywords = {conversational question answering, conversational search, question rewriting, transformer-based neural language models},
}
```
|
{"license": "gpl-3.0"}
|
gonced8/pegasus-conversational-qa
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"pegasus",
"text2text-generation",
"license:gpl-3.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
{}
|
gonced8/pegasus-qa
| null |
[
"transformers",
"pytorch",
"pegasus",
"text2text-generation",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
null | null |
{}
|
gone2808/distilbert-base-uncased-finetuned-squad
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
null | null |
{}
|
gone2808/test
| null |
[
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
text-generation
|
transformers
|
{}
|
goodjw/gpt-trinity-poem
| null |
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
fill-mask
|
transformers
|
{}
|
goodjw/klue-bert-mlm
| null |
[
"transformers",
"pytorch",
"bert",
"fill-mask",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
fill-mask
|
transformers
|
{}
|
goodjw/klue-roberta-large-tapt
| null |
[
"transformers",
"pytorch",
"roberta",
"fill-mask",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
fill-mask
|
transformers
|
{}
|
goodjw/klue-roberta-mlm
| null |
[
"transformers",
"pytorch",
"roberta",
"fill-mask",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
fill-mask
|
transformers
|
{}
|
goodjw/koelectra-mlm
| null |
[
"transformers",
"pytorch",
"electra",
"fill-mask",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
translation
|
transformers
|
# bert2bert_L-24_wmt_de_en EncoderDecoder model
The model was introduced in
[this paper](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn and first released in [this repository](https://tfhub.dev/google/bertseq2seq/bert24_de_en/1).
The model is an encoder-decoder model that was initialized on the `bert-large` checkpoints for both the encoder
and decoder and fine-tuned on German to English translation on the WMT dataset, which is linked above.
Disclaimer: The model card has been written by the Hugging Face team.
## How to use
You can use this model for translation, *e.g.*
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("google/bert2bert_L-24_wmt_de_en", pad_token="<pad>", eos_token="</s>", bos_token="<s>")
model = AutoModelForSeq2SeqLM.from_pretrained("google/bert2bert_L-24_wmt_de_en")
sentence = "Willst du einen Kaffee trinken gehen mit mir?"
input_ids = tokenizer(sentence, return_tensors="pt", add_special_tokens=False).input_ids
output_ids = model.generate(input_ids)[0]
print(tokenizer.decode(output_ids, skip_special_tokens=True))
# should output
# Want to drink a kaffee go with me? .
```
|
{"language": ["en", "de"], "license": "apache-2.0", "tags": ["translation"], "datasets": ["wmt14"]}
|
google/bert2bert_L-24_wmt_de_en
| null |
[
"transformers",
"pytorch",
"encoder-decoder",
"text2text-generation",
"translation",
"en",
"de",
"dataset:wmt14",
"arxiv:1907.12461",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
translation
|
transformers
|
# bert2bert_L-24_wmt_en_de EncoderDecoder model
The model was introduced in
[this paper](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn and first released in [this repository](https://tfhub.dev/google/bertseq2seq/bert24_en_de/1).
The model is an encoder-decoder model that was initialized on the `bert-large` checkpoints for both the encoder
and decoder and fine-tuned on English to German translation on the WMT dataset, which is linked above.
Disclaimer: The model card has been written by the Hugging Face team.
## How to use
You can use this model for translation, *e.g.*
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("google/bert2bert_L-24_wmt_en_de", pad_token="<pad>", eos_token="</s>", bos_token="<s>")
model = AutoModelForSeq2SeqLM.from_pretrained("google/bert2bert_L-24_wmt_en_de")
sentence = "Would you like to grab a coffee with me this week?"
input_ids = tokenizer(sentence, return_tensors="pt", add_special_tokens=False).input_ids
output_ids = model.generate(input_ids)[0]
print(tokenizer.decode(output_ids, skip_special_tokens=True))
# should output
# Möchten Sie diese Woche einen Kaffee mit mir schnappen?
|
{"language": ["en", "de"], "license": "apache-2.0", "tags": ["translation"], "datasets": ["wmt14"]}
|
google/bert2bert_L-24_wmt_en_de
| null |
[
"transformers",
"pytorch",
"encoder-decoder",
"text2text-generation",
"translation",
"en",
"de",
"dataset:wmt14",
"arxiv:1907.12461",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
{}
|
google/bert_for_seq_generation_L-24_bbc_encoder
| null |
[
"transformers",
"pytorch",
"bert-generation",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-10_H-128_A-2
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-10_H-256_A-4
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-10_H-512_A-8
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-10_H-768_A-12
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-12_H-128_A-2
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-12_H-256_A-4
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-12_H-512_A-8
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-12_H-768_A-12
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-2_H-128_A-2
| null |
[
"transformers",
"pytorch",
"jax",
"safetensors",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-2_H-256_A-4
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-2_H-512_A-8
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-2_H-768_A-12
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-4_H-128_A-2
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-4_H-256_A-4
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-4_H-512_A-8
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-4_H-768_A-12
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-6_H-128_A-2
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-6_H-256_A-4
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-6_H-512_A-8
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-6_H-768_A-12
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-8_H-128_A-2
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-8_H-256_A-4
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-8_H-512_A-8
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
BERT Miniatures
===
This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking).
We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.
Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.
You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below:
| |H=128|H=256|H=512|H=768|
|---|:---:|:---:|:---:|:---:|
| **L=2** |[**2/128 (BERT-Tiny)**][2_128]|[2/256][2_256]|[2/512][2_512]|[2/768][2_768]|
| **L=4** |[4/128][4_128]|[**4/256 (BERT-Mini)**][4_256]|[**4/512 (BERT-Small)**][4_512]|[4/768][4_768]|
| **L=6** |[6/128][6_128]|[6/256][6_256]|[6/512][6_512]|[6/768][6_768]|
| **L=8** |[8/128][8_128]|[8/256][8_256]|[**8/512 (BERT-Medium)**][8_512]|[8/768][8_768]|
| **L=10** |[10/128][10_128]|[10/256][10_256]|[10/512][10_512]|[10/768][10_768]|
| **L=12** |[12/128][12_128]|[12/256][12_256]|[12/512][12_512]|[**12/768 (BERT-Base)**][12_768]|
Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.
Here are the corresponding GLUE scores on the test set:
|Model|Score|CoLA|SST-2|MRPC|STS-B|QQP|MNLI-m|MNLI-mm|QNLI(v2)|RTE|WNLI|AX|
|---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|BERT-Tiny|64.2|0.0|83.2|81.1/71.1|74.3/73.6|62.2/83.4|70.2|70.3|81.5|57.2|62.3|21.0|
|BERT-Mini|65.8|0.0|85.9|81.1/71.8|75.4/73.3|66.4/86.2|74.8|74.3|84.1|57.9|62.3|26.1|
|BERT-Small|71.2|27.8|89.7|83.4/76.2|78.8/77.0|68.1/87.0|77.6|77.0|86.4|61.8|62.3|28.6|
|BERT-Medium|73.5|38.0|89.6|86.6/81.6|80.4/78.4|69.6/87.9|80.0|79.1|87.7|62.2|62.3|30.5|
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:
- batch sizes: 8, 16, 32, 64, 128
- learning rates: 3e-4, 1e-4, 5e-5, 3e-5
If you use these models, please cite the following paper:
```
@article{turc2019,
title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
journal={arXiv preprint arXiv:1908.08962v2 },
year={2019}
}
```
[2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2
[2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4
[2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8
[2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12
[4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2
[4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4
[4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8
[4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12
[6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2
[6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4
[6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8
[6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12
[8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2
[8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4
[8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8
[8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12
[10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2
[10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4
[10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8
[10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12
[12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2
[12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4
[12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8
[12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/bert_uncased_L-8_H-768_A-12
| null |
[
"transformers",
"pytorch",
"jax",
"bert",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
question-answering
|
transformers
|
# BigBird base trivia-itc
This model is a fine-tune checkpoint of `bigbird-roberta-base`, fine-tuned on `trivia_qa` with `BigBirdForQuestionAnsweringHead` on its top.
Check out [this](https://colab.research.google.com/drive/1DVOm1VHjW0eKCayFq1N2GpY6GR9M4tJP?usp=sharing) to see how well `google/bigbird-base-trivia-itc` performs on question answering.
## How to use
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import BigBirdForQuestionAnswering
# by default its in `block_sparse` mode with num_random_blocks=3, block_size=64
model = BigBirdForQuestionAnswering.from_pretrained("google/bigbird-base-trivia-itc")
# you can change `attention_type` to full attention like this:
model = BigBirdForQuestionAnswering.from_pretrained("google/bigbird-base-trivia-itc", attention_type="original_full")
# you can change `block_size` & `num_random_blocks` like this:
model = BigBirdForQuestionAnswering.from_pretrained("google/bigbird-base-trivia-itc", block_size=16, num_random_blocks=2)
question = "Replace me by any text you'd like."
context = "Put some context for answering"
encoded_input = tokenizer(question, context, return_tensors='pt')
output = model(**encoded_input)
```
# Fine-tuning config & hyper-parameters
- No. of global token = 128
- Window length = 192
- No. of random token = 192
- Max. sequence length = 4096
- No. of heads = 12
- No. of hidden layers = 12
- Hidden layer size = 768
- Batch size = 32
- Loss = cross-entropy noisy spans
## BibTeX entry and citation info
```tex
@misc{zaheer2021big,
title={Big Bird: Transformers for Longer Sequences},
author={Manzil Zaheer and Guru Guruganesh and Avinava Dubey and Joshua Ainslie and Chris Alberti and Santiago Ontanon and Philip Pham and Anirudh Ravula and Qifan Wang and Li Yang and Amr Ahmed},
year={2021},
eprint={2007.14062},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
|
{"language": "en", "license": "apache-2.0", "datasets": ["trivia_qa"]}
|
google/bigbird-base-trivia-itc
| null |
[
"transformers",
"pytorch",
"jax",
"big_bird",
"question-answering",
"en",
"dataset:trivia_qa",
"arxiv:2007.14062",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
summarization
|
transformers
|
# BigBirdPegasus model (large)
BigBird, is a sparse-attention based transformer which extends Transformer based models, such as BERT to much longer sequences. Moreover, BigBird comes along with a theoretical understanding of the capabilities of a complete transformer that the sparse model can handle.
BigBird was introduced in this [paper](https://arxiv.org/abs/2007.14062) and first released in this [repository](https://github.com/google-research/bigbird).
Disclaimer: The team releasing BigBird did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
BigBird relies on **block sparse attention** instead of normal attention (i.e. BERT's attention) and can handle sequences up to a length of 4096 at a much lower compute cost compared to BERT. It has achieved SOTA on various tasks involving very long sequences such as long documents summarization, question-answering with long contexts.
## How to use
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import BigBirdPegasusForConditionalGeneration, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("google/bigbird-pegasus-large-arxiv")
# by default encoder-attention is `block_sparse` with num_random_blocks=3, block_size=64
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-arxiv")
# decoder attention type can't be changed & will be "original_full"
# you can change `attention_type` (encoder only) to full attention like this:
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-arxiv", attention_type="original_full")
# you can change `block_size` & `num_random_blocks` like this:
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-arxiv", block_size=16, num_random_blocks=2)
text = "Replace me by any text you'd like."
inputs = tokenizer(text, return_tensors='pt')
prediction = model.generate(**inputs)
prediction = tokenizer.batch_decode(prediction)
```
## Training Procedure
This checkpoint is obtained after fine-tuning `BigBirdPegasusForConditionalGeneration` for **summarization** on **arxiv dataset** from [scientific_papers](https://huggingface.co/datasets/scientific_papers).
## BibTeX entry and citation info
```tex
@misc{zaheer2021big,
title={Big Bird: Transformers for Longer Sequences},
author={Manzil Zaheer and Guru Guruganesh and Avinava Dubey and Joshua Ainslie and Chris Alberti and Santiago Ontanon and Philip Pham and Anirudh Ravula and Qifan Wang and Li Yang and Amr Ahmed},
year={2021},
eprint={2007.14062},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["summarization"], "datasets": ["scientific_papers"], "model-index": [{"name": "google/bigbird-pegasus-large-arxiv", "results": [{"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "scientific_papers", "type": "scientific_papers", "config": "pubmed", "split": "test"}, "metrics": [{"type": "rouge", "value": 36.0276, "name": "ROUGE-1", "verified": true}, {"type": "rouge", "value": 13.4166, "name": "ROUGE-2", "verified": true}, {"type": "rouge", "value": 21.9612, "name": "ROUGE-L", "verified": true}, {"type": "rouge", "value": 29.648, "name": "ROUGE-LSUM", "verified": true}, {"type": "loss", "value": 2.774355173110962, "name": "loss", "verified": true}, {"type": "meteor", "value": 0.2824, "name": "meteor", "verified": true}, {"type": "gen_len", "value": 209.2537, "name": "gen_len", "verified": true}]}, {"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "cnn_dailymail", "type": "cnn_dailymail", "config": "3.0.0", "split": "test"}, "metrics": [{"type": "rouge", "value": 9.0885, "name": "ROUGE-1", "verified": true}, {"type": "rouge", "value": 1.0325, "name": "ROUGE-2", "verified": true}, {"type": "rouge", "value": 7.3182, "name": "ROUGE-L", "verified": true}, {"type": "rouge", "value": 8.1455, "name": "ROUGE-LSUM", "verified": true}, {"type": "loss", "value": NaN, "name": "loss", "verified": true}, {"type": "gen_len", "value": 210.4762, "name": "gen_len", "verified": true}]}, {"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "xsum", "type": "xsum", "config": "default", "split": "test"}, "metrics": [{"type": "rouge", "value": 4.9787, "name": "ROUGE-1", "verified": true}, {"type": "rouge", "value": 0.3527, "name": "ROUGE-2", "verified": true}, {"type": "rouge", "value": 4.3679, "name": "ROUGE-L", "verified": true}, {"type": "rouge", "value": 4.1723, "name": "ROUGE-LSUM", "verified": true}, {"type": "loss", "value": NaN, "name": "loss", "verified": true}, {"type": "gen_len", "value": 230.4886, "name": "gen_len", "verified": true}]}, {"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "scientific_papers", "type": "scientific_papers", "config": "arxiv", "split": "test"}, "metrics": [{"type": "rouge", "value": 43.4702, "name": "ROUGE-1", "verified": true}, {"type": "rouge", "value": 17.4297, "name": "ROUGE-2", "verified": true}, {"type": "rouge", "value": 26.2587, "name": "ROUGE-L", "verified": true}, {"type": "rouge", "value": 35.5587, "name": "ROUGE-LSUM", "verified": true}, {"type": "loss", "value": 2.1113228797912598, "name": "loss", "verified": true}, {"type": "gen_len", "value": 183.3702, "name": "gen_len", "verified": true}]}, {"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "samsum", "type": "samsum", "config": "samsum", "split": "test"}, "metrics": [{"type": "rouge", "value": 3.621, "name": "ROUGE-1", "verified": true}, {"type": "rouge", "value": 0.1699, "name": "ROUGE-2", "verified": true}, {"type": "rouge", "value": 3.2016, "name": "ROUGE-L", "verified": true}, {"type": "rouge", "value": 3.3269, "name": "ROUGE-LSUM", "verified": true}, {"type": "loss", "value": 7.664482116699219, "name": "loss", "verified": true}, {"type": "gen_len", "value": 233.8107, "name": "gen_len", "verified": true}]}]}]}
|
google/bigbird-pegasus-large-arxiv
| null |
[
"transformers",
"pytorch",
"bigbird_pegasus",
"text2text-generation",
"summarization",
"en",
"dataset:scientific_papers",
"arxiv:2007.14062",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
summarization
|
transformers
|
# BigBirdPegasus model (large)
BigBird, is a sparse-attention based transformer which extends Transformer based models, such as BERT to much longer sequences. Moreover, BigBird comes along with a theoretical understanding of the capabilities of a complete transformer that the sparse model can handle.
BigBird was introduced in this [paper](https://arxiv.org/abs/2007.14062) and first released in this [repository](https://github.com/google-research/bigbird).
Disclaimer: The team releasing BigBird did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
BigBird relies on **block sparse attention** instead of normal attention (i.e. BERT's attention) and can handle sequences up to a length of 4096 at a much lower compute cost compared to BERT. It has achieved SOTA on various tasks involving very long sequences such as long documents summarization, question-answering with long contexts.
## How to use
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import BigBirdPegasusForConditionalGeneration, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("google/bigbird-pegasus-large-bigpatent")
# by default encoder-attention is `block_sparse` with num_random_blocks=3, block_size=64
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-bigpatent")
# decoder attention type can't be changed & will be "original_full"
# you can change `attention_type` (encoder only) to full attention like this:
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-bigpatent", attention_type="original_full")
# you can change `block_size` & `num_random_blocks` like this:
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-bigpatent", block_size=16, num_random_blocks=2)
text = "Replace me by any text you'd like."
inputs = tokenizer(text, return_tensors='pt')
prediction = model.generate(**inputs)
prediction = tokenizer.batch_decode(prediction)
```
## Training Procedure
This checkpoint is obtained after fine-tuning `BigBirdPegasusForConditionalGeneration` for **summarization** on [big_patent](https://huggingface.co/datasets/big_patent) dataset.
## BibTeX entry and citation info
```tex
@misc{zaheer2021big,
title={Big Bird: Transformers for Longer Sequences},
author={Manzil Zaheer and Guru Guruganesh and Avinava Dubey and Joshua Ainslie and Chris Alberti and Santiago Ontanon and Philip Pham and Anirudh Ravula and Qifan Wang and Li Yang and Amr Ahmed},
year={2021},
eprint={2007.14062},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["summarization"], "datasets": ["big_patent"]}
|
google/bigbird-pegasus-large-bigpatent
| null |
[
"transformers",
"pytorch",
"bigbird_pegasus",
"text2text-generation",
"summarization",
"en",
"dataset:big_patent",
"arxiv:2007.14062",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
summarization
|
transformers
|
# BigBirdPegasus model (large)
BigBird, is a sparse-attention based transformer which extends Transformer based models, such as BERT to much longer sequences. Moreover, BigBird comes along with a theoretical understanding of the capabilities of a complete transformer that the sparse model can handle.
BigBird was introduced in this [paper](https://arxiv.org/abs/2007.14062) and first released in this [repository](https://github.com/google-research/bigbird).
Disclaimer: The team releasing BigBird did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
BigBird relies on **block sparse attention** instead of normal attention (i.e. BERT's attention) and can handle sequences up to a length of 4096 at a much lower compute cost compared to BERT. It has achieved SOTA on various tasks involving very long sequences such as long documents summarization, question-answering with long contexts.
## How to use
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import BigBirdPegasusForConditionalGeneration, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("google/bigbird-pegasus-large-pubmed")
# by default encoder-attention is `block_sparse` with num_random_blocks=3, block_size=64
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-pubmed")
# decoder attention type can't be changed & will be "original_full"
# you can change `attention_type` (encoder only) to full attention like this:
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-pubmed", attention_type="original_full")
# you can change `block_size` & `num_random_blocks` like this:
model = BigBirdPegasusForConditionalGeneration.from_pretrained("google/bigbird-pegasus-large-pubmed", block_size=16, num_random_blocks=2)
text = "Replace me by any text you'd like."
inputs = tokenizer(text, return_tensors='pt')
prediction = model.generate(**inputs)
prediction = tokenizer.batch_decode(prediction)
```
## Training Procedure
This checkpoint is obtained after fine-tuning `BigBirdPegasusForConditionalGeneration` for **summarization** on **pubmed dataset** from [scientific_papers](https://huggingface.co/datasets/scientific_papers).
## BibTeX entry and citation info
```tex
@misc{zaheer2021big,
title={Big Bird: Transformers for Longer Sequences},
author={Manzil Zaheer and Guru Guruganesh and Avinava Dubey and Joshua Ainslie and Chris Alberti and Santiago Ontanon and Philip Pham and Anirudh Ravula and Qifan Wang and Li Yang and Amr Ahmed},
year={2021},
eprint={2007.14062},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["summarization"], "datasets": ["scientific_papers"], "model-index": [{"name": "google/bigbird-pegasus-large-pubmed", "results": [{"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "scientific_papers", "type": "scientific_papers", "config": "pubmed", "split": "test"}, "metrics": [{"type": "rouge", "value": 40.8966, "name": "ROUGE-1", "verified": true}, {"type": "rouge", "value": 18.1161, "name": "ROUGE-2", "verified": true}, {"type": "rouge", "value": 26.1743, "name": "ROUGE-L", "verified": true}, {"type": "rouge", "value": 34.2773, "name": "ROUGE-LSUM", "verified": true}, {"type": "loss", "value": 2.1707184314727783, "name": "loss", "verified": true}, {"type": "meteor", "value": 0.3513, "name": "meteor", "verified": true}, {"type": "gen_len", "value": 221.2531, "name": "gen_len", "verified": true}]}, {"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "scientific_papers", "type": "scientific_papers", "config": "arxiv", "split": "test"}, "metrics": [{"type": "rouge", "value": 40.3815, "name": "ROUGE-1", "verified": true}, {"type": "rouge", "value": 14.374, "name": "ROUGE-2", "verified": true}, {"type": "rouge", "value": 23.4773, "name": "ROUGE-L", "verified": true}, {"type": "rouge", "value": 33.772, "name": "ROUGE-LSUM", "verified": true}, {"type": "loss", "value": 3.235051393508911, "name": "loss", "verified": true}, {"type": "gen_len", "value": 186.2003, "name": "gen_len", "verified": true}]}]}]}
|
google/bigbird-pegasus-large-pubmed
| null |
[
"transformers",
"pytorch",
"bigbird_pegasus",
"text2text-generation",
"summarization",
"en",
"dataset:scientific_papers",
"arxiv:2007.14062",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# BigBird base model
BigBird, is a sparse-attention based transformer which extends Transformer based models, such as BERT to much longer sequences. Moreover, BigBird comes along with a theoretical understanding of the capabilities of a complete transformer that the sparse model can handle.
It is a pretrained model on English language using a masked language modeling (MLM) objective. It was introduced in this [paper](https://arxiv.org/abs/2007.14062) and first released in this [repository](https://github.com/google-research/bigbird).
Disclaimer: The team releasing BigBird did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
BigBird relies on **block sparse attention** instead of normal attention (i.e. BERT's attention) and can handle sequences up to a length of 4096 at a much lower compute cost compared to BERT. It has achieved SOTA on various tasks involving very long sequences such as long documents summarization, question-answering with long contexts.
## How to use
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import BigBirdModel
# by default its in `block_sparse` mode with num_random_blocks=3, block_size=64
model = BigBirdModel.from_pretrained("google/bigbird-roberta-base")
# you can change `attention_type` to full attention like this:
model = BigBirdModel.from_pretrained("google/bigbird-roberta-base", attention_type="original_full")
# you can change `block_size` & `num_random_blocks` like this:
model = BigBirdModel.from_pretrained("google/bigbird-roberta-base", block_size=16, num_random_blocks=2)
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Training Data
This model is pre-trained on four publicly available datasets: **Books**, **CC-News**, **Stories** and **Wikipedia**. It used same sentencepiece vocabulary as RoBERTa (which is in turn borrowed from GPT2).
## Training Procedure
Document longer than 4096 were split into multiple documents and documents that were much smaller than 4096 were joined. Following the original BERT training, 15% of tokens were masked and model is trained to predict the mask.
Model is warm started from RoBERTa’s checkpoint.
## BibTeX entry and citation info
```tex
@misc{zaheer2021big,
title={Big Bird: Transformers for Longer Sequences},
author={Manzil Zaheer and Guru Guruganesh and Avinava Dubey and Joshua Ainslie and Chris Alberti and Santiago Ontanon and Philip Pham and Anirudh Ravula and Qifan Wang and Li Yang and Amr Ahmed},
year={2021},
eprint={2007.14062},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
|
{"language": "en", "license": "apache-2.0", "datasets": ["bookcorpus", "wikipedia", "cc_news"]}
|
google/bigbird-roberta-base
| null |
[
"transformers",
"pytorch",
"jax",
"big_bird",
"pretraining",
"en",
"dataset:bookcorpus",
"dataset:wikipedia",
"dataset:cc_news",
"arxiv:2007.14062",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
fill-mask
|
transformers
|
# BigBird large model
BigBird, is a sparse-attention based transformer which extends Transformer based models, such as BERT to much longer sequences. Moreover, BigBird comes along with a theoretical understanding of the capabilities of a complete transformer that the sparse model can handle.
It is a pretrained model on English language using a masked language modeling (MLM) objective. It was introduced in this [paper](https://arxiv.org/abs/2007.14062) and first released in this [repository](https://github.com/google-research/bigbird).
Disclaimer: The team releasing BigBird did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
BigBird relies on **block sparse attention** instead of normal attention (i.e. BERT's attention) and can handle sequences up to a length of 4096 at a much lower compute cost compared to BERT. It has achieved SOTA on various tasks involving very long sequences such as long documents summarization, question-answering with long contexts.
## How to use
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import BigBirdModel
# by default its in `block_sparse` mode with num_random_blocks=3, block_size=64
model = BigBirdModel.from_pretrained("google/bigbird-roberta-large")
# you can change `attention_type` to full attention like this:
model = BigBirdModel.from_pretrained("google/bigbird-roberta-large", attention_type="original_full")
# you can change `block_size` & `num_random_blocks` like this:
model = BigBirdModel.from_pretrained("google/bigbird-roberta-large", block_size=16, num_random_blocks=2)
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Training Data
This model is pre-trained on four publicly available datasets: **Books**, **CC-News**, **Stories** and **Wikipedia**. It used same sentencepiece vocabulary as RoBERTa (which is in turn borrowed from GPT2).
## Training Procedure
Document longer than 4096 were split into multiple documents and documents that were much smaller than 4096 were joined. Following the original BERT training, 15% of tokens were masked and model is trained to predict the mask.
Model is warm started from RoBERTa’s checkpoint.
## BibTeX entry and citation info
```tex
@misc{zaheer2021big,
title={Big Bird: Transformers for Longer Sequences},
author={Manzil Zaheer and Guru Guruganesh and Avinava Dubey and Joshua Ainslie and Chris Alberti and Santiago Ontanon and Philip Pham and Anirudh Ravula and Qifan Wang and Li Yang and Amr Ahmed},
year={2021},
eprint={2007.14062},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
|
{"language": "en", "license": "apache-2.0", "datasets": ["bookcorpus", "wikipedia", "cc_news"]}
|
google/bigbird-roberta-large
| null |
[
"transformers",
"pytorch",
"jax",
"big_bird",
"fill-mask",
"en",
"dataset:bookcorpus",
"dataset:wikipedia",
"dataset:cc_news",
"arxiv:2007.14062",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# ByT5 - Base
ByT5 is a tokenizer-free version of [Google's T5](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) and generally follows the architecture of [MT5](https://huggingface.co/google/mt5-base).
ByT5 was only pre-trained on [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) excluding any supervised training with an average span-mask of 20 UTF-8 characters. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
ByT5 works especially well on noisy text data,*e.g.*, `google/byt5-base` significantly outperforms [mt5-base](https://huggingface.co/google/mt5-base) on [TweetQA](https://arxiv.org/abs/1907.06292).
Paper: [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626)
Authors: *Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel*
## Example Inference
ByT5 works on raw UTF-8 bytes and can be used without a tokenizer:
```python
from transformers import T5ForConditionalGeneration
import torch
model = T5ForConditionalGeneration.from_pretrained('google/byt5-base')
input_ids = torch.tensor([list("Life is like a box of chocolates.".encode("utf-8"))]) + 3 # add 3 for special tokens
labels = torch.tensor([list("La vie est comme une boîte de chocolat.".encode("utf-8"))]) + 3 # add 3 for special tokens
loss = model(input_ids, labels=labels).loss # forward pass
```
For batched inference & training it is however recommended using a tokenizer class for padding:
```python
from transformers import T5ForConditionalGeneration, AutoTokenizer
model = T5ForConditionalGeneration.from_pretrained('google/byt5-base')
tokenizer = AutoTokenizer.from_pretrained('google/byt5-base')
model_inputs = tokenizer(["Life is like a box of chocolates.", "Today is Monday."], padding="longest", return_tensors="pt")
labels = tokenizer(["La vie est comme une boîte de chocolat.", "Aujourd'hui c'est lundi."], padding="longest", return_tensors="pt").input_ids
loss = model(**model_inputs, labels=labels).loss # forward pass
```
## Abstract
Most widely-used pre-trained language models operate on sequences of tokens corresponding to word or subword units. Encoding text as a sequence of tokens requires a tokenizer, which is typically created as an independent artifact from the model. Token-free models that instead operate directly on raw text (bytes or characters) have many benefits: they can process text in any language out of the box, they are more robust to noise, and they minimize technical debt by removing complex and error-prone text preprocessing pipelines. Since byte or character sequences are longer than token sequences, past work on token-free models has often introduced new model architectures designed to amortize the cost of operating directly on raw text. In this paper, we show that a standard Transformer architecture can be used with minimal modifications to process byte sequences. We carefully characterize the trade-offs in terms of parameter count, training FLOPs, and inference speed, and show that byte-level models are competitive with their token-level counterparts. We also demonstrate that byte-level models are significantly more robust to noise and perform better on tasks that are sensitive to spelling and pronunciation. As part of our contribution, we release a new set of pre-trained byte-level Transformer models based on the T5 architecture, as well as all code and data used in our experiments.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/byt5-base
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"t5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:1907.06292",
"arxiv:2105.13626",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# ByT5 - large
ByT5 is a tokenizer-free version of [Google's T5](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) and generally follows the architecture of [MT5](https://huggingface.co/google/mt5-large).
ByT5 was only pre-trained on [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) excluding any supervised training with an average span-mask of 20 UTF-8 characters. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
ByT5 works especially well on noisy text data,*e.g.*, `google/byt5-large` significantly outperforms [mt5-large](https://huggingface.co/google/mt5-large) on [TweetQA](https://arxiv.org/abs/1907.06292).
Paper: [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626)
Authors: *Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel*
## Example Inference
ByT5 works on raw UTF-8 bytes and can be used without a tokenizer:
```python
from transformers import T5ForConditionalGeneration
import torch
model = T5ForConditionalGeneration.from_pretrained('google/byt5-large')
input_ids = torch.tensor([list("Life is like a box of chocolates.".encode("utf-8"))]) + 3 # add 3 for special tokens
labels = torch.tensor([list("La vie est comme une boîte de chocolat.".encode("utf-8"))]) + 3 # add 3 for special tokens
loss = model(input_ids, labels=labels).loss # forward pass
```
For batched inference & training it is however recommended using a tokenizer class for padding:
```python
from transformers import T5ForConditionalGeneration, AutoTokenizer
model = T5ForConditionalGeneration.from_pretrained('google/byt5-large')
tokenizer = AutoTokenizer.from_pretrained('google/byt5-large')
model_inputs = tokenizer(["Life is like a box of chocolates.", "Today is Monday."], padding="longest", return_tensors="pt")
labels = tokenizer(["La vie est comme une boîte de chocolat.", "Aujourd'hui c'est lundi."], padding="longest", return_tensors="pt").input_ids
loss = model(**model_inputs, labels=labels).loss # forward pass
```
## Abstract
Most widely-used pre-trained language models operate on sequences of tokens corresponding to word or subword units. Encoding text as a sequence of tokens requires a tokenizer, which is typically created as an independent artifact from the model. Token-free models that instead operate directly on raw text (bytes or characters) have many benefits: they can process text in any language out of the box, they are more robust to noise, and they minimize technical debt by removing complex and error-prone text preprocessing pipelines. Since byte or character sequences are longer than token sequences, past work on token-free models has often introduced new model architectures designed to amortize the cost of operating directly on raw text. In this paper, we show that a standard Transformer architecture can be used with minimal modifications to process byte sequences. We carefully characterize the trade-offs in terms of parameter count, training FLOPs, and inference speed, and show that byte-level models are competitive with their token-level counterparts. We also demonstrate that byte-level models are significantly more robust to noise and perform better on tasks that are sensitive to spelling and pronunciation. As part of our contribution, we release a new set of pre-trained byte-level Transformer models based on the T5 architecture, as well as all code and data used in our experiments.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/byt5-large
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"t5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:1907.06292",
"arxiv:2105.13626",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# ByT5 - Small
ByT5 is a tokenizer-free version of [Google's T5](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) and generally follows the architecture of [MT5](https://huggingface.co/google/mt5-small).
ByT5 was only pre-trained on [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) excluding any supervised training with an average span-mask of 20 UTF-8 characters. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
ByT5 works especially well on noisy text data,*e.g.*, `google/byt5-small` significantly outperforms [mt5-small](https://huggingface.co/google/mt5-small) on [TweetQA](https://arxiv.org/abs/1907.06292).
Paper: [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626)
Authors: *Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel*
## Example Inference
ByT5 works on raw UTF-8 bytes and can be used without a tokenizer:
```python
from transformers import T5ForConditionalGeneration
import torch
model = T5ForConditionalGeneration.from_pretrained('google/byt5-small')
input_ids = torch.tensor([list("Life is like a box of chocolates.".encode("utf-8"))]) + 3 # add 3 for special tokens
labels = torch.tensor([list("La vie est comme une boîte de chocolat.".encode("utf-8"))]) + 3 # add 3 for special tokens
loss = model(input_ids, labels=labels).loss # forward pass
```
For batched inference & training it is however recommended using a tokenizer class for padding:
```python
from transformers import T5ForConditionalGeneration, AutoTokenizer
model = T5ForConditionalGeneration.from_pretrained('google/byt5-small')
tokenizer = AutoTokenizer.from_pretrained('google/byt5-small')
model_inputs = tokenizer(["Life is like a box of chocolates.", "Today is Monday."], padding="longest", return_tensors="pt")
labels = tokenizer(["La vie est comme une boîte de chocolat.", "Aujourd'hui c'est lundi."], padding="longest", return_tensors="pt").input_ids
loss = model(**model_inputs, labels=labels).loss # forward pass
```
## Abstract
Most widely-used pre-trained language models operate on sequences of tokens corresponding to word or subword units. Encoding text as a sequence of tokens requires a tokenizer, which is typically created as an independent artifact from the model. Token-free models that instead operate directly on raw text (bytes or characters) have many benefits: they can process text in any language out of the box, they are more robust to noise, and they minimize technical debt by removing complex and error-prone text preprocessing pipelines. Since byte or character sequences are longer than token sequences, past work on token-free models has often introduced new model architectures designed to amortize the cost of operating directly on raw text. In this paper, we show that a standard Transformer architecture can be used with minimal modifications to process byte sequences. We carefully characterize the trade-offs in terms of parameter count, training FLOPs, and inference speed, and show that byte-level models are competitive with their token-level counterparts. We also demonstrate that byte-level models are significantly more robust to noise and perform better on tasks that are sensitive to spelling and pronunciation. As part of our contribution, we release a new set of pre-trained byte-level Transformer models based on the T5 architecture, as well as all code and data used in our experiments.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/byt5-small
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"t5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:1907.06292",
"arxiv:2105.13626",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# ByT5 - xl
ByT5 is a tokenizer-free version of [Google's T5](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) and generally follows the architecture of [MT5](https://huggingface.co/google/mt5-xl).
ByT5 was only pre-trained on [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) excluding any supervised training with an average span-mask of 20 UTF-8 characters. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
ByT5 works especially well on noisy text data,*e.g.*, `google/byt5-xl` significantly outperforms [mt5-xl](https://huggingface.co/google/mt5-xl) on [TweetQA](https://arxiv.org/abs/1907.06292).
Paper: [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626)
Authors: *Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel*
## Example Inference
ByT5 works on raw UTF-8 bytes and can be used without a tokenizer:
```python
from transformers import T5ForConditionalGeneration
import torch
model = T5ForConditionalGeneration.from_pretrained('google/byt5-xl')
input_ids = torch.tensor([list("Life is like a box of chocolates.".encode("utf-8"))]) + 3 # add 3 for special tokens
labels = torch.tensor([list("La vie est comme une boîte de chocolat.".encode("utf-8"))]) + 3 # add 3 for special tokens
loss = model(input_ids, labels=labels).loss # forward pass
```
For batched inference & training it is however recommended using a tokenizer class for padding:
```python
from transformers import T5ForConditionalGeneration, AutoTokenizer
model = T5ForConditionalGeneration.from_pretrained('google/byt5-xl')
tokenizer = AutoTokenizer.from_pretrained('google/byt5-xl')
model_inputs = tokenizer(["Life is like a box of chocolates.", "Today is Monday."], padding="longest", return_tensors="pt")
labels = tokenizer(["La vie est comme une boîte de chocolat.", "Aujourd'hui c'est lundi."], padding="longest", return_tensors="pt").input_ids
loss = model(**model_inputs, labels=labels).loss # forward pass
```
## Abstract
Most widely-used pre-trained language models operate on sequences of tokens corresponding to word or subword units. Encoding text as a sequence of tokens requires a tokenizer, which is typically created as an independent artifact from the model. Token-free models that instead operate directly on raw text (bytes or characters) have many benefits: they can process text in any language out of the box, they are more robust to noise, and they minimize technical debt by removing complex and error-prone text preprocessing pipelines. Since byte or character sequences are longer than token sequences, past work on token-free models has often introduced new model architectures designed to amortize the cost of operating directly on raw text. In this paper, we show that a standard Transformer architecture can be used with minimal modifications to process byte sequences. We carefully characterize the trade-offs in terms of parameter count, training FLOPs, and inference speed, and show that byte-level models are competitive with their token-level counterparts. We also demonstrate that byte-level models are significantly more robust to noise and perform better on tasks that are sensitive to spelling and pronunciation. As part of our contribution, we release a new set of pre-trained byte-level Transformer models based on the T5 architecture, as well as all code and data used in our experiments.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/byt5-xl
| null |
[
"transformers",
"pytorch",
"tf",
"t5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:1907.06292",
"arxiv:2105.13626",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# ByT5 - xxl
ByT5 is a tokenizer-free version of [Google's T5](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) and generally follows the architecture of [MT5](https://huggingface.co/google/mt5-xxl).
ByT5 was only pre-trained on [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) excluding any supervised training with an average span-mask of 20 UTF-8 characters. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
ByT5 works especially well on noisy text data,*e.g.*, `google/byt5-xxl` significantly outperforms [mt5-xxl](https://huggingface.co/google/mt5-xxl) on [TweetQA](https://arxiv.org/abs/1907.06292).
Paper: [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626)
Authors: *Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel*
## Example Inference
ByT5 works on raw UTF-8 bytes and can be used without a tokenizer:
```python
from transformers import T5ForConditionalGeneration
import torch
model = T5ForConditionalGeneration.from_pretrained('google/byt5-xxl')
input_ids = torch.tensor([list("Life is like a box of chocolates.".encode("utf-8"))]) + 3 # add 3 for special tokens
labels = torch.tensor([list("La vie est comme une boîte de chocolat.".encode("utf-8"))]) + 3 # add 3 for special tokens
loss = model(input_ids, labels=labels).loss # forward pass
```
For batched inference & training it is however recommended using a tokenizer class for padding:
```python
from transformers import T5ForConditionalGeneration, AutoTokenizer
model = T5ForConditionalGeneration.from_pretrained('google/byt5-xxl')
tokenizer = AutoTokenizer.from_pretrained('google/byt5-xxl')
model_inputs = tokenizer(["Life is like a box of chocolates.", "Today is Monday."], padding="longest", return_tensors="pt")
labels = tokenizer(["La vie est comme une boîte de chocolat.", "Aujourd'hui c'est lundi."], padding="longest", return_tensors="pt").input_ids
loss = model(**model_inputs, labels=labels).loss # forward pass
```
## Abstract
Most widely-used pre-trained language models operate on sequences of tokens corresponding to word or subword units. Encoding text as a sequence of tokens requires a tokenizer, which is typically created as an independent artifact from the model. Token-free models that instead operate directly on raw text (bytes or characters) have many benefits: they can process text in any language out of the box, they are more robust to noise, and they minimize technical debt by removing complex and error-prone text preprocessing pipelines. Since byte or character sequences are longer than token sequences, past work on token-free models has often introduced new model architectures designed to amortize the cost of operating directly on raw text. In this paper, we show that a standard Transformer architecture can be used with minimal modifications to process byte sequences. We carefully characterize the trade-offs in terms of parameter count, training FLOPs, and inference speed, and show that byte-level models are competitive with their token-level counterparts. We also demonstrate that byte-level models are significantly more robust to noise and perform better on tasks that are sensitive to spelling and pronunciation. As part of our contribution, we release a new set of pre-trained byte-level Transformer models based on the T5 architecture, as well as all code and data used in our experiments.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/byt5-xxl
| null |
[
"transformers",
"pytorch",
"tf",
"t5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:1907.06292",
"arxiv:2105.13626",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# CANINE-c (CANINE pre-trained with autoregressive character loss)
Pretrained CANINE model on 104 languages using a masked language modeling (MLM) objective. It was introduced in the paper [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) and first released in [this repository](https://github.com/google-research/language/tree/master/language/canine).
What's special about CANINE is that it doesn't require an explicit tokenizer (such as WordPiece or SentencePiece) as other models like BERT and RoBERTa. Instead, it directly operates at a character level: each character is turned into its [Unicode code point](https://en.wikipedia.org/wiki/Code_point#:~:text=For%20Unicode%2C%20the%20particular%20sequence,forming%20a%20self%2Dsynchronizing%20code.).
This means that input processing is trivial and can typically be accomplished as:
```
input_ids = [ord(char) for char in text]
```
The ord() function is part of Python, and turns each character into its Unicode code point.
Disclaimer: The team releasing CANINE did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
CANINE is a transformers model pretrained on a large corpus of multilingual data in a self-supervised fashion, similar to BERT. This means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely, it was pretrained with two objectives:
* Masked language modeling (MLM): one randomly masks part of the inputs, which the model needs to predict. This model (CANINE-c) is trained with an autoregressive character loss. One masks several character spans within each sequence, which the model then autoregressively predicts.
* Next sentence prediction (NSP): the model concatenates two sentences as inputs during pretraining. Sometimes they correspond to sentences that were next to each other in the original text, sometimes not. The model then has to predict if the two sentences were following each other or not.
This way, the model learns an inner representation of multiple languages that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard classifier using the features produced by the CANINE model as inputs.
## Intended uses & limitations
You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to be fine-tuned on a downstream task. See the [model hub](https://huggingface.co/models?filter=canine) to look for fine-tuned versions on a task that interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked) to make decisions, such as sequence classification, token classification or question answering. For tasks such as text generation you should look at models like GPT2.
### How to use
Here is how to use this model:
```python
from transformers import CanineTokenizer, CanineModel
model = CanineModel.from_pretrained('google/canine-c')
tokenizer = CanineTokenizer.from_pretrained('google/canine-c')
inputs = ["Life is like a box of chocolates.", "You never know what you gonna get."]
encoding = tokenizer(inputs, padding="longest", truncation=True, return_tensors="pt")
outputs = model(**encoding) # forward pass
pooled_output = outputs.pooler_output
sequence_output = outputs.last_hidden_state
```
## Training data
The CANINE model was pretrained on on the multilingual Wikipedia data of [mBERT](https://github.com/google-research/bert/blob/master/multilingual.md), which includes 104 languages.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2103-06874,
author = {Jonathan H. Clark and
Dan Garrette and
Iulia Turc and
John Wieting},
title = {{CANINE:} Pre-training an Efficient Tokenization-Free Encoder for
Language Representation},
journal = {CoRR},
volume = {abs/2103.06874},
year = {2021},
url = {https://arxiv.org/abs/2103.06874},
archivePrefix = {arXiv},
eprint = {2103.06874},
timestamp = {Tue, 16 Mar 2021 11:26:59 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2103-06874.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"language": ["multilingual", "af", "sq", "ar", "an", "hy", "ast", "az", "ba", "eu", "bar", "be", "bn", "inc", "bs", "br", "bg", "my", "ca", "ceb", "ce", "zh", "cv", "hr", "cs", "da", "nl", "en", "et", "fi", "fr", "gl", "ka", "de", "el", "gu", "ht", "he", "hi", "hu", "is", "io", "id", "ga", "it", "ja", "jv", "kn", "kk", "ky", "ko", "la", "lv", "lt", "roa", "nds", "lm", "mk", "mg", "ms", "ml", "mr", "mn", "min", "ne", "new", "nb", "nn", "oc", "fa", "pms", "pl", "pt", "pa", "ro", "ru", "sco", "sr", "hr", "scn", "sk", "sl", "aze", "es", "su", "sw", "sv", "tl", "tg", "th", "ta", "tt", "te", "tr", "uk", "ud", "uz", "vi", "vo", "war", "cy", "fry", "pnb", "yo"], "license": "apache-2.0", "datasets": ["bookcorpus", "wikipedia"]}
|
google/canine-c
| null |
[
"transformers",
"pytorch",
"canine",
"feature-extraction",
"multilingual",
"af",
"sq",
"ar",
"an",
"hy",
"ast",
"az",
"ba",
"eu",
"bar",
"be",
"bn",
"inc",
"bs",
"br",
"bg",
"my",
"ca",
"ceb",
"ce",
"zh",
"cv",
"hr",
"cs",
"da",
"nl",
"en",
"et",
"fi",
"fr",
"gl",
"ka",
"de",
"el",
"gu",
"ht",
"he",
"hi",
"hu",
"is",
"io",
"id",
"ga",
"it",
"ja",
"jv",
"kn",
"kk",
"ky",
"ko",
"la",
"lv",
"lt",
"roa",
"nds",
"lm",
"mk",
"mg",
"ms",
"ml",
"mr",
"mn",
"min",
"ne",
"new",
"nb",
"nn",
"oc",
"fa",
"pms",
"pl",
"pt",
"pa",
"ro",
"ru",
"sco",
"sr",
"scn",
"sk",
"sl",
"aze",
"es",
"su",
"sw",
"sv",
"tl",
"tg",
"th",
"ta",
"tt",
"te",
"tr",
"uk",
"ud",
"uz",
"vi",
"vo",
"war",
"cy",
"fry",
"pnb",
"yo",
"dataset:bookcorpus",
"dataset:wikipedia",
"arxiv:2103.06874",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# CANINE-s (CANINE pre-trained with subword loss)
Pretrained CANINE model on 104 languages using a masked language modeling (MLM) objective. It was introduced in the paper [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) and first released in [this repository](https://github.com/google-research/language/tree/master/language/canine).
What's special about CANINE is that it doesn't require an explicit tokenizer (such as WordPiece or SentencePiece) as other models like BERT and RoBERTa. Instead, it directly operates at a character level: each character is turned into its [Unicode code point](https://en.wikipedia.org/wiki/Code_point#:~:text=For%20Unicode%2C%20the%20particular%20sequence,forming%20a%20self%2Dsynchronizing%20code.).
This means that input processing is trivial and can typically be accomplished as:
```
input_ids = [ord(char) for char in text]
```
The ord() function is part of Python, and turns each character into its Unicode code point.
Disclaimer: The team releasing CANINE did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
CANINE is a transformers model pretrained on a large corpus of multilingual data in a self-supervised fashion, similar to BERT. This means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely, it was pretrained with two objectives:
* Masked language modeling (MLM): one randomly masks part of the inputs, which the model needs to predict. This model (CANINE-s) is trained with a subword loss, meaning that the model needs to predict the identities of subword tokens, while taking characters as input. By reading characters yet predicting subword tokens, the hard token boundary constraint found in other models such as BERT is turned into a soft inductive bias in CANINE.
* Next sentence prediction (NSP): the model concatenates two sentences as inputs during pretraining. Sometimes they correspond to sentences that were next to each other in the original text, sometimes not. The model then has to predict if the two sentences were following each other or not.
This way, the model learns an inner representation of multiple languages that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard classifier using the features produced by the CANINE model as inputs.
## Intended uses & limitations
You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to be fine-tuned on a downstream task. See the [model hub](https://huggingface.co/models?filter=canine) to look for fine-tuned versions on a task that interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked) to make decisions, such as sequence classification, token classification or question answering. For tasks such as text generation you should look at models like GPT2.
### How to use
Here is how to use this model:
```python
from transformers import CanineTokenizer, CanineModel
model = CanineModel.from_pretrained('google/canine-s')
tokenizer = CanineTokenizer.from_pretrained('google/canine-s')
inputs = ["Life is like a box of chocolates.", "You never know what you gonna get."]
encoding = tokenizer(inputs, padding="longest", truncation=True, return_tensors="pt")
outputs = model(**encoding) # forward pass
pooled_output = outputs.pooler_output
sequence_output = outputs.last_hidden_state
```
## Training data
The CANINE model was pretrained on on the multilingual Wikipedia data of [mBERT](https://github.com/google-research/bert/blob/master/multilingual.md), which includes 104 languages.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2103-06874,
author = {Jonathan H. Clark and
Dan Garrette and
Iulia Turc and
John Wieting},
title = {{CANINE:} Pre-training an Efficient Tokenization-Free Encoder for
Language Representation},
journal = {CoRR},
volume = {abs/2103.06874},
year = {2021},
url = {https://arxiv.org/abs/2103.06874},
archivePrefix = {arXiv},
eprint = {2103.06874},
timestamp = {Tue, 16 Mar 2021 11:26:59 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2103-06874.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"language": ["multilingual", "af", "sq", "ar", "an", "hy", "ast", "az", "ba", "eu", "bar", "be", "bn", "inc", "bs", "br", "bg", "my", "ca", "ceb", "ce", "zh", "cv", "hr", "cs", "da", "nl", "en", "et", "fi", "fr", "gl", "ka", "de", "el", "gu", "ht", "he", "hi", "hu", "is", "io", "id", "ga", "it", "ja", "jv", "kn", "kk", "ky", "ko", "la", "lv", "lt", "roa", "nds", "lm", "mk", "mg", "ms", "ml", "mr", "mn", "min", "ne", "new", "nb", "nn", "oc", "fa", "pms", "pl", "pt", "pa", "ro", "ru", "sco", "sr", "hr", "scn", "sk", "sl", "aze", "es", "su", "sw", "sv", "tl", "tg", "th", "ta", "tt", "te", "tr", "uk", "ud", "uz", "vi", "vo", "war", "cy", "fry", "pnb", "yo"], "license": "apache-2.0", "datasets": ["bookcorpus", "wikipedia"]}
|
google/canine-s
| null |
[
"transformers",
"pytorch",
"canine",
"feature-extraction",
"multilingual",
"af",
"sq",
"ar",
"an",
"hy",
"ast",
"az",
"ba",
"eu",
"bar",
"be",
"bn",
"inc",
"bs",
"br",
"bg",
"my",
"ca",
"ceb",
"ce",
"zh",
"cv",
"hr",
"cs",
"da",
"nl",
"en",
"et",
"fi",
"fr",
"gl",
"ka",
"de",
"el",
"gu",
"ht",
"he",
"hi",
"hu",
"is",
"io",
"id",
"ga",
"it",
"ja",
"jv",
"kn",
"kk",
"ky",
"ko",
"la",
"lv",
"lt",
"roa",
"nds",
"lm",
"mk",
"mg",
"ms",
"ml",
"mr",
"mn",
"min",
"ne",
"new",
"nb",
"nn",
"oc",
"fa",
"pms",
"pl",
"pt",
"pa",
"ro",
"ru",
"sco",
"sr",
"scn",
"sk",
"sl",
"aze",
"es",
"su",
"sw",
"sv",
"tl",
"tg",
"th",
"ta",
"tt",
"te",
"tr",
"uk",
"ud",
"uz",
"vi",
"vo",
"war",
"cy",
"fry",
"pnb",
"yo",
"dataset:bookcorpus",
"dataset:wikipedia",
"arxiv:2103.06874",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
## ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
**ELECTRA** is a new method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. At large scale, ELECTRA achieves state-of-the-art results on the [SQuAD 2.0](https://rajpurkar.github.io/SQuAD-explorer/) dataset.
For a detailed description and experimental results, please refer to our paper [ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators](https://openreview.net/pdf?id=r1xMH1BtvB).
This repository contains code to pre-train ELECTRA, including small ELECTRA models on a single GPU. It also supports fine-tuning ELECTRA on downstream tasks including classification tasks (e.g,. [GLUE](https://gluebenchmark.com/)), QA tasks (e.g., [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/)), and sequence tagging tasks (e.g., [text chunking](https://www.clips.uantwerpen.be/conll2000/chunking/)).
## How to use the discriminator in `transformers`
```python
from transformers import ElectraForPreTraining, ElectraTokenizerFast
import torch
discriminator = ElectraForPreTraining.from_pretrained("google/electra-base-discriminator")
tokenizer = ElectraTokenizerFast.from_pretrained("google/electra-base-discriminator")
sentence = "The quick brown fox jumps over the lazy dog"
fake_sentence = "The quick brown fox fake over the lazy dog"
fake_tokens = tokenizer.tokenize(fake_sentence)
fake_inputs = tokenizer.encode(fake_sentence, return_tensors="pt")
discriminator_outputs = discriminator(fake_inputs)
predictions = torch.round((torch.sign(discriminator_outputs[0]) + 1) / 2)
[print("%7s" % token, end="") for token in fake_tokens]
[print("%7s" % int(prediction), end="") for prediction in predictions.tolist()]
```
|
{"language": "en", "license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/electra-base-discriminator
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"rust",
"electra",
"pretraining",
"en",
"arxiv:1406.2661",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
fill-mask
|
transformers
|
## ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
**ELECTRA** is a new method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. At large scale, ELECTRA achieves state-of-the-art results on the [SQuAD 2.0](https://rajpurkar.github.io/SQuAD-explorer/) dataset.
For a detailed description and experimental results, please refer to our paper [ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators](https://openreview.net/pdf?id=r1xMH1BtvB).
This repository contains code to pre-train ELECTRA, including small ELECTRA models on a single GPU. It also supports fine-tuning ELECTRA on downstream tasks including classification tasks (e.g,. [GLUE](https://gluebenchmark.com/)), QA tasks (e.g., [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/)), and sequence tagging tasks (e.g., [text chunking](https://www.clips.uantwerpen.be/conll2000/chunking/)).
## How to use the generator in `transformers`
```python
from transformers import pipeline
fill_mask = pipeline(
"fill-mask",
model="google/electra-base-generator",
tokenizer="google/electra-base-generator"
)
print(
fill_mask(f"HuggingFace is creating a {fill_mask.tokenizer.mask_token} that the community uses to solve NLP tasks.")
)
```
|
{"language": "en", "license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/electra-base-generator
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"rust",
"electra",
"fill-mask",
"en",
"arxiv:1406.2661",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
## ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
**ELECTRA** is a new method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. At large scale, ELECTRA achieves state-of-the-art results on the [SQuAD 2.0](https://rajpurkar.github.io/SQuAD-explorer/) dataset.
For a detailed description and experimental results, please refer to our paper [ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators](https://openreview.net/pdf?id=r1xMH1BtvB).
This repository contains code to pre-train ELECTRA, including small ELECTRA models on a single GPU. It also supports fine-tuning ELECTRA on downstream tasks including classification tasks (e.g,. [GLUE](https://gluebenchmark.com/)), QA tasks (e.g., [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/)), and sequence tagging tasks (e.g., [text chunking](https://www.clips.uantwerpen.be/conll2000/chunking/)).
## How to use the discriminator in `transformers`
```python
from transformers import ElectraForPreTraining, ElectraTokenizerFast
import torch
discriminator = ElectraForPreTraining.from_pretrained("google/electra-large-discriminator")
tokenizer = ElectraTokenizerFast.from_pretrained("google/electra-large-discriminator")
sentence = "The quick brown fox jumps over the lazy dog"
fake_sentence = "The quick brown fox fake over the lazy dog"
fake_tokens = tokenizer.tokenize(fake_sentence)
fake_inputs = tokenizer.encode(fake_sentence, return_tensors="pt")
discriminator_outputs = discriminator(fake_inputs)
predictions = torch.round((torch.sign(discriminator_outputs[0]) + 1) / 2)
[print("%7s" % token, end="") for token in fake_tokens]
[print("%7s" % int(prediction), end="") for prediction in predictions.tolist()]
```
|
{"language": "en", "license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/electra-large-discriminator
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"electra",
"pretraining",
"en",
"arxiv:1406.2661",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
fill-mask
|
transformers
|
## ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
**ELECTRA** is a new method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. At large scale, ELECTRA achieves state-of-the-art results on the [SQuAD 2.0](https://rajpurkar.github.io/SQuAD-explorer/) dataset.
For a detailed description and experimental results, please refer to our paper [ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators](https://openreview.net/pdf?id=r1xMH1BtvB).
This repository contains code to pre-train ELECTRA, including small ELECTRA models on a single GPU. It also supports fine-tuning ELECTRA on downstream tasks including classification tasks (e.g,. [GLUE](https://gluebenchmark.com/)), QA tasks (e.g., [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/)), and sequence tagging tasks (e.g., [text chunking](https://www.clips.uantwerpen.be/conll2000/chunking/)).
## How to use the generator in `transformers`
```python
from transformers import pipeline
fill_mask = pipeline(
"fill-mask",
model="google/electra-large-generator",
tokenizer="google/electra-large-generator"
)
print(
fill_mask(f"HuggingFace is creating a {nlp.tokenizer.mask_token} that the community uses to solve NLP tasks.")
)
```
|
{"language": "en", "license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/electra-large-generator
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"electra",
"fill-mask",
"en",
"arxiv:1406.2661",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
## ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
**ELECTRA** is a new method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. At large scale, ELECTRA achieves state-of-the-art results on the [SQuAD 2.0](https://rajpurkar.github.io/SQuAD-explorer/) dataset.
For a detailed description and experimental results, please refer to our paper [ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators](https://openreview.net/pdf?id=r1xMH1BtvB).
This repository contains code to pre-train ELECTRA, including small ELECTRA models on a single GPU. It also supports fine-tuning ELECTRA on downstream tasks including classification tasks (e.g,. [GLUE](https://gluebenchmark.com/)), QA tasks (e.g., [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/)), and sequence tagging tasks (e.g., [text chunking](https://www.clips.uantwerpen.be/conll2000/chunking/)).
## How to use the discriminator in `transformers`
```python
from transformers import ElectraForPreTraining, ElectraTokenizerFast
import torch
discriminator = ElectraForPreTraining.from_pretrained("google/electra-small-discriminator")
tokenizer = ElectraTokenizerFast.from_pretrained("google/electra-small-discriminator")
sentence = "The quick brown fox jumps over the lazy dog"
fake_sentence = "The quick brown fox fake over the lazy dog"
fake_tokens = tokenizer.tokenize(fake_sentence)
fake_inputs = tokenizer.encode(fake_sentence, return_tensors="pt")
discriminator_outputs = discriminator(fake_inputs)
predictions = torch.round((torch.sign(discriminator_outputs[0]) + 1) / 2)
[print("%7s" % token, end="") for token in fake_tokens]
[print("%7s" % int(prediction), end="") for prediction in predictions.squeeze().tolist()]
```
|
{"language": "en", "license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/electra-small-discriminator
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"electra",
"pretraining",
"en",
"arxiv:1406.2661",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
fill-mask
|
transformers
|
**WARNING**: This is the official generator checkpoint as in the [ELECTRA original codebase](https://github.com/google-research/electra). However, this model is not scaled properly for pre-training with [google/electra-small-discriminator](https://huggingface.co/google/electra-small-discriminator). The paper recommends a hyperparameter multiplier of 1/4 between the discriminator and generator for this given model to avoid training instabilities. This would not be the case when using `google/electra-small-generator` and `google/electra-small-discriminator`, which are similar in size.
## ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
**ELECTRA** is a new method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. At large scale, ELECTRA achieves state-of-the-art results on the [SQuAD 2.0](https://rajpurkar.github.io/SQuAD-explorer/) dataset.
For a detailed description and experimental results, please refer to our paper [ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators](https://openreview.net/pdf?id=r1xMH1BtvB).
This repository contains code to pre-train ELECTRA, including small ELECTRA models on a single GPU. It also supports fine-tuning ELECTRA on downstream tasks including classification tasks (e.g,. [GLUE](https://gluebenchmark.com/)), QA tasks (e.g., [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/)), and sequence tagging tasks (e.g., [text chunking](https://www.clips.uantwerpen.be/conll2000/chunking/)).
## How to use the generator in `transformers`
```python
from transformers import pipeline
fill_mask = pipeline(
"fill-mask",
model="google/electra-small-generator",
tokenizer="google/electra-small-generator"
)
print(
fill_mask(f"HuggingFace is creating a {nlp.tokenizer.mask_token} that the community uses to solve NLP tasks.")
)
```
|
{"language": "en", "license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/electra-small-generator
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"electra",
"fill-mask",
"en",
"arxiv:1406.2661",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# FNet base model
Pretrained model on English language using a masked language modeling (MLM) and next sentence prediction (NSP) objective. It was
introduced in [this paper](https://arxiv.org/abs/2105.03824) and first released in [this repository](https://github.com/google-research/google-research/tree/master/f_net).
This model is cased: it makes a difference between english and English. The model achieves 0.58 accuracy on MLM objective and 0.80 on NSP objective.
Disclaimer: This model card has been written by [gchhablani](https://huggingface.co/gchhablani).
## Model description
FNet is a transformers model with attention replaced with fourier transforms. Hence, the inputs do not contain an `attention_mask`. It is pretrained on a large corpus of
English data in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labelling
them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and
labels from those texts. More precisely, it was pretrained with two objectives:
- Masked language modeling (MLM): taking a sentence, the model randomly masks 15% of the words in the input then run
the entire masked sentence through the model and has to predict the masked words. This is different from traditional
recurrent neural networks (RNNs) that usually see the words one after the other, or from autoregressive models like
GPT which internally mask the future tokens. It allows the model to learn a bidirectional representation of the
sentence.
- Next sentence prediction (NSP): the models concatenates two masked sentences as inputs during pretraining. Sometimes
they correspond to sentences that were next to each other in the original text, sometimes not. The model then has to
predict if the two sentences were following each other or not.
This way, the model learns an inner representation of the English language that can then be used to extract features
useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard
classifier using the features produced by the FNet model as inputs.
## Intended uses & limitations
You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to
be fine-tuned on a downstream task. See the [model hub](https://huggingface.co/models?filter=fnet) to look for
fine-tuned versions on a task that interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked)
to make decisions, such as sequence classification, token classification or question answering. For tasks such as text
generation you should look at model like GPT2.
## Training data
The FNet model was pretrained on [C4](https://huggingface.co/datasets/c4), a cleaned version of the Common Crawl dataset.
## Training procedure
### Preprocessing
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 32,000. The inputs of the model are
then of the form:
```
[CLS] Sentence A [SEP] Sentence B [SEP]
```
With probability 0.5, sentence A and sentence B correspond to two consecutive sentences in the original corpus and in
the other cases, it's another random sentence in the corpus. Note that what is considered a sentence here is a
consecutive span of text usually longer than a single sentence. The only constrain is that the result with the two
"sentences" has a combined length of less than 512 tokens.
The details of the masking procedure for each sentence are the following:
- 15% of the tokens are masked.
- In 80% of the cases, the masked tokens are replaced by `[MASK]`.
- In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace.
- In the 10% remaining cases, the masked tokens are left as is.
### Pretraining
FNet-base was trained on 4 cloud TPUs in Pod configuration (16 TPU chips total) for one million steps with a batch size
of 256. The sequence length was limited to 512 tokens. The optimizer
used is Adam with a learning rate of 1e-4, \\(\beta_{1} = 0.9\\) and \\(\beta_{2} = 0.999\\), a weight decay of 0.01,
learning rate warmup for 10,000 steps and linear decay of the learning rate after.
## Evaluation results
FNet-base was fine-tuned and evaluated on the validation data of the [GLUE benchamrk](https://huggingface.co/datasets/glue). The results of the official model (written in Flax) can be seen in Table 1 on page 7 of [the official paper](https://arxiv.org/abs/2105.03824).
For comparison, this model (ported to PyTorch) was fine-tuned and evaluated using the [official Hugging Face GLUE evaluation scripts](https://github.com/huggingface/transformers/tree/master/examples/pytorch/text-classification#glue-tasks) alongside [bert-base-cased](https://hf.co/models/bert-base-cased) for comparison.
The training was done on a single 16GB NVIDIA Tesla V100 GPU. For MRPC/WNLI, the models were trained for 5 epochs, while for other tasks, the models were trained for 3 epochs. A sequence length of 512 was used with batch size 16 and learning rate 2e-5.
The following table summarizes the results for [fnet-base](https://huggingface.co/google/fnet-base) (called *FNet (PyTorch) - Reproduced*) and [bert-base-cased](https://hf.co/models/bert-base-cased) (called *Bert (PyTorch) - Reproduced*) in terms of **fine-tuning** speed. The format is *hour:min:seconds*. **Note** that the authors compared **pre-traning** speed in [the official paper](https://arxiv.org/abs/2105.03824) instead.
| Task/Model | FNet-base (PyTorch) |Bert-base (PyTorch)|
|:----:|:-----------:|:----:|
| MNLI-(m/mm) | [06:40:55](https://huggingface.co/gchhablani/fnet-base-finetuned-mnli) | [09:52:33](https://huggingface.co/gchhablani/bert-base-cased-finetuned-mnli)|
| QQP | [06:21:16](https://huggingface.co/gchhablani/fnet-base-finetuned-qqp) | [09:25:01](https://huggingface.co/gchhablani/bert-base-cased-finetuned-qqp) |
| QNLI | [01:48:22](https://huggingface.co/gchhablani/fnet-base-finetuned-qnli) | [02:40:22](https://huggingface.co/gchhablani/bert-base-cased-finetuned-qnli)|
| SST-2 | [01:09:27](https://huggingface.co/gchhablani/fnet-base-finetuned-sst2) | [01:42:17](https://huggingface.co/gchhablani/bert-base-cased-finetuned-sst2)|
| CoLA | [00:09:47](https://huggingface.co/gchhablani/fnet-base-finetuned-cola) | [00:14:20](https://huggingface.co/gchhablani/bert-base-cased-finetuned-cola)|
| STS-B | [00:07:09](https://huggingface.co/gchhablani/fnet-base-finetuned-stsb) | [00:10:24](https://huggingface.co/gchhablani/bert-base-cased-finetuned-stsb)|
| MRPC | [00:07:48](https://huggingface.co/gchhablani/fnet-base-finetuned-mrpc) | [00:11:12](https://huggingface.co/gchhablani/bert-base-cased-finetuned-mrpc)|
| RTE | [00:03:24](https://huggingface.co/gchhablani/fnet-base-finetuned-rte) | [00:04:51](https://huggingface.co/gchhablani/bert-base-cased-finetuned-rte)|
| WNLI | [00:02:37](https://huggingface.co/gchhablani/fnet-base-finetuned-wnli) | [00:03:23](https://huggingface.co/gchhablani/bert-base-cased-finetuned-wnli)|
| SUM | 16:30:45 | 24:23:56 |
On average the PyTorch version of FNet-base requires *ca.* 32% less time for GLUE fine-tuning on GPU.
The following table summarizes the results for [fnet-base](https://huggingface.co/google/fnet-base) (called *FNet (PyTorch) - Reproduced*) and [bert-base-cased](https://hf.co/models/bert-base-cased) (called *Bert (PyTorch) - Reproduced*) in terms of performance and compares it to the reported performance of the official FNet-base model (called *FNet (Flax) - Official*). Note that the training hyperparameters of the reproduced models were not the same as the official model, so the performance may differ significantly for some tasks (for example: CoLA).
| Task/Model | Metric | FNet-base (PyTorch) | Bert-base (PyTorch) | FNet-Base (Flax - official) |
|:----:|:-----------:|:----:|:-----------:|:----:|
| MNLI-(m/mm) | Accuracy or Match/Mismatch | [76.75](https://huggingface.co/gchhablani/fnet-base-finetuned-mnli) | [84.10](https://huggingface.co/gchhablani/bert-base-cased-finetuned-mnli) | 72/73 |
| QQP | mean(Accuracy,F1) | [86.5](https://huggingface.co/gchhablani/fnet-base-finetuned-qqp) | [89.26](https://huggingface.co/gchhablani/bert-base-cased-finetuned-qqp) | 83 |
| QNLI | Accuracy | [84.39](https://huggingface.co/gchhablani/fnet-base-finetuned-qnli) | [90.99](https://huggingface.co/gchhablani/bert-base-cased-finetuned-qnli) | 80 |
| SST-2 | Accuracy | [89.45](https://huggingface.co/gchhablani/fnet-base-finetuned-sst2) | [92.32](https://huggingface.co/gchhablani/bert-base-cased-finetuned-sst2) | 95 |
| CoLA | Matthews corr or Accuracy | [35.94](https://huggingface.co/gchhablani/fnet-base-finetuned-cola) | [59.57](https://huggingface.co/gchhablani/bert-base-cased-finetuned-cola) | 69 |
| STS-B | Spearman corr. | [82.19](https://huggingface.co/gchhablani/fnet-base-finetuned-stsb) | [88.98](https://huggingface.co/gchhablani/bert-base-cased-finetuned-stsb) | 79 |
| MRPC | mean(F1/Accuracy) | [81.15](https://huggingface.co/gchhablani/fnet-base-finetuned-mrpc) | [88.15](https://huggingface.co/gchhablani/bert-base-cased-finetuned-mrpc) | 76 |
| RTE | Accuracy | [62.82](https://huggingface.co/gchhablani/fnet-base-finetuned-rte) | [67.15](https://huggingface.co/gchhablani/bert-base-cased-finetuned-rte) | 63 |
| WNLI | Accuracy | [54.93](https://huggingface.co/gchhablani/fnet-base-finetuned-wnli) | [46.48](https://huggingface.co/gchhablani/bert-base-cased-finetuned-wnli) | - |
| Avg | - | 72.7 | 78.6 | 76.7 |
We can see that FNet-base achieves around 93% of BERT-base's performance on average.
For more details, please refer to the checkpoints linked with the scores. On overview of all fine-tuned checkpoints of the following table can be accessed [here](https://huggingface.co/models?other=fnet-bert-base-comparison).
### How to use
You can use this model directly with a pipeline for masked language modeling:
**Note: The mask filling pipeline doesn't work exactly as the original model performs masking after converting to tokens. In masking pipeline an additional space is added after the [MASK].**
```python
>>> from transformers import FNetForMaskedLM, FNetTokenizer, pipeline
>>> tokenizer = FNetTokenizer.from_pretrained("google/fnet-base")
>>> model = FNetForMaskedLM.from_pretrained("google/fnet-base")
>>> unmasker = pipeline('fill-mask', model=model, tokenizer=tokenizer)
>>> unmasker("Hello I'm a [MASK] model.")
[
{"sequence": "hello i'm a new model.", "score": 0.12073223292827606, "token": 351, "token_str": "new"},
{"sequence": "hello i'm a first model.", "score": 0.08501081168651581, "token": 478, "token_str": "first"},
{"sequence": "hello i'm a next model.", "score": 0.060546260327100754, "token": 1037, "token_str": "next"},
{"sequence": "hello i'm a last model.", "score": 0.038265593349933624, "token": 813, "token_str": "last"},
{"sequence": "hello i'm a sister model.", "score": 0.033868927508592606, "token": 6232, "token_str": "sister"},
]
```
Here is how to use this model to get the features of a given text in PyTorch:
**Note: You must specify the maximum sequence length to be 512 and truncate/pad to the same length because the original model has no attention mask and considers all the hidden states during forward pass.**
```python
from transformers import FNetTokenizer, FNetModel
tokenizer = FNetTokenizer.from_pretrained("google/fnet-base")
model = FNetModel.from_pretrained("google/fnet-base")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt', padding='max_length', truncation=True, max_length=512)
output = model(**encoded_input)
```
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2105-03824,
author = {James Lee{-}Thorp and
Joshua Ainslie and
Ilya Eckstein and
Santiago Onta{\~{n}}{\'{o}}n},
title = {FNet: Mixing Tokens with Fourier Transforms},
journal = {CoRR},
volume = {abs/2105.03824},
year = {2021},
url = {https://arxiv.org/abs/2105.03824},
archivePrefix = {arXiv},
eprint = {2105.03824},
timestamp = {Fri, 14 May 2021 12:13:30 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2105-03824.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
## Contributions
Thanks to [@gchhablani](https://huggingface.co/gchhablani) for adding this model.
|
{"language": "en", "license": "apache-2.0", "tags": ["fnet"], "datasets": ["c4"]}
|
google/fnet-base
| null |
[
"transformers",
"pytorch",
"rust",
"fnet",
"pretraining",
"en",
"dataset:c4",
"arxiv:2105.03824",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# FNet large model
Pretrained model on English language using a masked language modeling (MLM) and next sentence prediction (NSP) objective. It was
introduced in [this paper](https://arxiv.org/abs/2105.03824) and first released in [this repository](https://github.com/google-research/google-research/tree/master/f_net).
This model is cased: it makes a difference between english and English. The model achieves 0.58 accuracy on MLM objective and 0.80 on NSP objective.
Disclaimer: This model card has been written by [gchhablani](https://huggingface.co/gchhablani).
## Model description
FNet is a transformers model with attention replaced with fourier transforms. Hence, the inputs do not contain an `attention_mask`. It is pretrained on a large corpus of
English data in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labelling
them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and
labels from those texts. More precisely, it was pretrained with two objectives:
- Masked language modeling (MLM): taking a sentence, the model randomly masks 15% of the words in the input then run
the entire masked sentence through the model and has to predict the masked words. This is different from traditional
recurrent neural networks (RNNs) that usually see the words one after the other, or from autoregressive models like
GPT which internally mask the future tokens. It allows the model to learn a bidirectional representation of the
sentence.
- Next sentence prediction (NSP): the models concatenates two masked sentences as inputs during pretraining. Sometimes
they correspond to sentences that were next to each other in the original text, sometimes not. The model then has to
predict if the two sentences were following each other or not.
This way, the model learns an inner representation of the English language that can then be used to extract features
useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard
classifier using the features produced by the FNet model as inputs.
This model has the following configuration:
- 24-layer
- 1024 hidden dimension
## Intended uses & limitations
You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to
be fine-tuned on a downstream task. See the [model hub](https://huggingface.co/models?filter=fnet) to look for
fine-tuned versions on a task that interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked)
to make decisions, such as sequence classification, token classification or question answering. For tasks such as text
generation you should look at model like GPT2.
### How to use
You can use this model directly with a pipeline for masked language modeling:
**Note: The mask filling pipeline doesn't work exactly as the original model performs masking after converting to tokens. In masking pipeline an additional space is added after the [MASK].**
```python
>>> from transformers import FNetForMaskedLM, FNetTokenizer, pipeline
>>> tokenizer = FNetTokenizer.from_pretrained("google/fnet-large")
>>> model = FNetForMaskedLM.from_pretrained("google/fnet-large")
>>> unmasker = pipeline('fill-mask', model=model, tokenizer=tokenizer)
>>> unmasker("Hello I'm a [MASK] model.")
[
{"sequence": "hello i'm a. model.", "score": 0.12840192019939423, "token": 16678, "token_str": "."},
{"sequence": "hello i'm a a model.", "score": 0.07460460811853409, "token": 8, "token_str": "a"},
{"sequence": "hello i'm a, model.", "score": 0.05011311173439026, "token": 16680, "token_str": ","},
{"sequence": "hello i'm a and model.", "score": 0.047409165650606155, "token": 36, "token_str": "and"},
{"sequence": "hello i'm a the model.", "score": 0.0269990973174572, "token": 13, "token_str": "the"},
]
```
Here is how to use this model to get the features of a given text in PyTorch:
**Note: You must specify the maximum sequence length to be 512 and truncate/pad to the same length because the original model has no attention mask and considers all the hidden states during forward pass.**
```python
from transformers import FNetTokenizer, FNetModel
tokenizer = FNetTokenizer.from_pretrained("google/fnet-large")
model = FNetModel.from_pretrained("google/fnet-large")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt', padding='max_length', truncation=True, max_length=512)
output = model(**encoded_input)
```
### Limitations and bias
Even if the training data used for this model could be characterized as fairly neutral, this model can have biased predictions. However, the model's MLM accuracy may also affect answers. Given below are some example where gender-bias could be expected:
```python
>>> from transformers import FNetForMaskedLM, FNetTokenizer, pipeline
>>> tokenizer = FNetTokenizer.from_pretrained("google/fnet-large")
>>> model = FNetForMaskedLM.from_pretrained("google/fnet-large")
>>> unmasker = pipeline('fill-mask', model=model, tokenizer=tokenizer)
>>> unmasker("The man worked as a [MASK].")
[
{"sequence": "the man worked as a a.", "score": 0.39862048625946045, "token": 8, "token_str": "a"},
{"sequence": "the man worked as a the.", "score": 0.20786496996879578, "token": 13, "token_str": "the"},
{"sequence": "the man worked as a as.", "score": 0.012523212470114231, "token": 106, "token_str": "as"},
{"sequence": "the man worked as a an.", "score": 0.010838045738637447, "token": 102, "token_str": "an"},
{"sequence": "the man worked as a and.", "score": 0.006571347825229168, "token": 36, "token_str": "and"},
]
>>> unmasker("The woman worked as a [MASK].")
[
{"sequence": "the woman worked as a the.", "score": 0.3320266902446747, "token": 13, "token_str": "the"},
{"sequence": "the woman worked as a a.", "score": 0.2591220438480377, "token": 8, "token_str": "a"},
{"sequence": "the woman worked as a as.", "score": 0.011250585317611694, "token": 106, "token_str": "as"},
{"sequence": "the woman worked as a an.", "score": 0.010153685696423054, "token": 102, "token_str": "an"},
{"sequence": "the woman worked as a and.", "score": 0.010126154869794846, "token": 36, "token_str": "and"},
]
```
This bias will also affect all fine-tuned versions of this model.
## Training data
The FNet model was pretrained on [C4](https://huggingface.co/datasets/c4), a cleaned version of the Common Crawl dataset.
## Training procedure
### Preprocessing
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 32,000. The inputs of the model are
then of the form:
```
[CLS] Sentence A [SEP] Sentence B [SEP]
```
With probability 0.5, sentence A and sentence B correspond to two consecutive sentences in the original corpus and in
the other cases, it's another random sentence in the corpus. Note that what is considered a sentence here is a
consecutive span of text usually longer than a single sentence. The only constrain is that the result with the two
"sentences" has a combined length of less than 512 tokens.
The details of the masking procedure for each sentence are the following:
- 15% of the tokens are masked.
- In 80% of the cases, the masked tokens are replaced by `[MASK]`.
- In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace.
- In the 10% remaining cases, the masked tokens are left as is.
### Pretraining
The model was trained on 4 cloud TPUs in Pod configuration (16 TPU chips total) for one million steps with a batch size
of 256. The sequence length was limited to 512 tokens. The optimizer
used is Adam with a learning rate of 1e-4, \\(\beta_{1} = 0.9\\) and \\(\beta_{2} = 0.999\\), a weight decay of 0.01,
learning rate warmup for 10,000 steps and linear decay of the learning rate after.
## Evaluation results
When fine-tuned on downstream tasks, this model achieves the following results:
Glue test results:
| Task | MNLI-(m/mm) | QQP | QNLI | SST-2 | CoLA | STS-B | MRPC | RTE | Average |
|:----:|:-----------:|:----:|:----:|:-----:|:----:|:-----:|:----:|:----:|:-------:|
| | 78/76 | 85 | 85 | 94 | 78 | 84 | 88 | 69| 81.9 |
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2105-03824,
author = {James Lee{-}Thorp and
Joshua Ainslie and
Ilya Eckstein and
Santiago Onta{\~{n}}{\'{o}}n},
title = {FNet: Mixing Tokens with Fourier Transforms},
journal = {CoRR},
volume = {abs/2105.03824},
year = {2021},
url = {https://arxiv.org/abs/2105.03824},
archivePrefix = {arXiv},
eprint = {2105.03824},
timestamp = {Fri, 14 May 2021 12:13:30 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2105-03824.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
## Contributions
Thanks to [@gchhablani](https://huggingface.co/gchhablani) for adding this model.
|
{"language": "en", "license": "apache-2.0", "tags": ["fnet"], "datasets": ["c4"]}
|
google/fnet-large
| null |
[
"transformers",
"pytorch",
"fnet",
"pretraining",
"en",
"dataset:c4",
"arxiv:2105.03824",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
## MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices
MobileBERT is a thin version of BERT_LARGE, while equipped with bottleneck structures and a carefully designed balance
between self-attentions and feed-forward networks.
This checkpoint is the original MobileBert Optimized Uncased English:
[uncased_L-24_H-128_B-512_A-4_F-4_OPT](https://storage.googleapis.com/cloud-tpu-checkpoints/mobilebert/uncased_L-24_H-128_B-512_A-4_F-4_OPT.tar.gz)
checkpoint.
## How to use MobileBERT in `transformers`
```python
from transformers import pipeline
fill_mask = pipeline(
"fill-mask",
model="google/mobilebert-uncased",
tokenizer="google/mobilebert-uncased"
)
print(
fill_mask(f"HuggingFace is creating a {fill_mask.tokenizer.mask_token} that the community uses to solve NLP tasks.")
)
```
|
{"language": "en", "license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/google.png"}
|
google/mobilebert-uncased
| null |
[
"transformers",
"pytorch",
"tf",
"rust",
"mobilebert",
"pretraining",
"en",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
[Google's mT5](https://github.com/google-research/multilingual-t5)
mT5 is pretrained on the [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) corpus, covering 101 languages:
Afrikaans, Albanian, Amharic, Arabic, Armenian, Azerbaijani, Basque, Belarusian, Bengali, Bulgarian, Burmese, Catalan, Cebuano, Chichewa, Chinese, Corsican, Czech, Danish, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Korean, Kurdish, Kyrgyz, Lao, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Nepali, Norwegian, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scottish Gaelic, Serbian, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Sotho, Spanish, Sundanese, Swahili, Swedish, Tajik, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Uzbek, Vietnamese, Welsh, West Frisian, Xhosa, Yiddish, Yoruba, Zulu.
**Note**: mT5 was only pre-trained on mC4 excluding any supervised training. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
Pretraining Dataset: [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual)
Other Community Checkpoints: [here](https://huggingface.co/models?search=mt5)
Paper: [mT5: A massively multilingual pre-trained text-to-text transformer](https://arxiv.org/abs/2010.11934)
Authors: *Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, Colin Raffel*
## Abstract
The recent "Text-to-Text Transfer Transformer" (T5) leveraged a unified text-to-text format and scale to attain state-of-the-art results on a wide variety of English-language NLP tasks. In this paper, we introduce mT5, a multilingual variant of T5 that was pre-trained on a new Common Crawl-based dataset covering 101 languages. We describe the design and modified training of mT5 and demonstrate its state-of-the-art performance on many multilingual benchmarks. All of the code and model checkpoints used in this work are publicly available.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/mt5-base
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"mt5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:2010.11934",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
[Google's mT5](https://github.com/google-research/multilingual-t5)
mT5 is pretrained on the [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) corpus, covering 101 languages:
Afrikaans, Albanian, Amharic, Arabic, Armenian, Azerbaijani, Basque, Belarusian, Bengali, Bulgarian, Burmese, Catalan, Cebuano, Chichewa, Chinese, Corsican, Czech, Danish, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Korean, Kurdish, Kyrgyz, Lao, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Nepali, Norwegian, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scottish Gaelic, Serbian, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Sotho, Spanish, Sundanese, Swahili, Swedish, Tajik, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Uzbek, Vietnamese, Welsh, West Frisian, Xhosa, Yiddish, Yoruba, Zulu.
**Note**: mT5 was only pre-trained on mC4 excluding any supervised training. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
Pretraining Dataset: [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual)
Other Community Checkpoints: [here](https://huggingface.co/models?search=mt5)
Paper: [mT5: A massively multilingual pre-trained text-to-text transformer](https://arxiv.org/abs/2010.11934)
Authors: *Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, Colin Raffel*
## Abstract
The recent "Text-to-Text Transfer Transformer" (T5) leveraged a unified text-to-text format and scale to attain state-of-the-art results on a wide variety of English-language NLP tasks. In this paper, we introduce mT5, a multilingual variant of T5 that was pre-trained on a new Common Crawl-based dataset covering 101 languages. We describe the design and modified training of mT5 and demonstrate its state-of-the-art performance on many multilingual benchmarks. All of the code and model checkpoints used in this work are publicly available.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/mt5-large
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"mt5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:2010.11934",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
[Google's mT5](https://github.com/google-research/multilingual-t5)
mT5 is pretrained on the [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) corpus, covering 101 languages:
Afrikaans, Albanian, Amharic, Arabic, Armenian, Azerbaijani, Basque, Belarusian, Bengali, Bulgarian, Burmese, Catalan, Cebuano, Chichewa, Chinese, Corsican, Czech, Danish, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Korean, Kurdish, Kyrgyz, Lao, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Nepali, Norwegian, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scottish Gaelic, Serbian, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Sotho, Spanish, Sundanese, Swahili, Swedish, Tajik, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Uzbek, Vietnamese, Welsh, West Frisian, Xhosa, Yiddish, Yoruba, Zulu.
**Note**: mT5 was only pre-trained on mC4 excluding any supervised training. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
Pretraining Dataset: [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual)
Other Community Checkpoints: [here](https://huggingface.co/models?search=mt5)
Paper: [mT5: A massively multilingual pre-trained text-to-text transformer](https://arxiv.org/abs/2010.11934)
Authors: *Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, Colin Raffel*
## Abstract
The recent "Text-to-Text Transfer Transformer" (T5) leveraged a unified text-to-text format and scale to attain state-of-the-art results on a wide variety of English-language NLP tasks. In this paper, we introduce mT5, a multilingual variant of T5 that was pre-trained on a new Common Crawl-based dataset covering 101 languages. We describe the design and modified training of mT5 and demonstrate its state-of-the-art performance on many multilingual benchmarks. All of the code and model checkpoints used in this work are publicly available.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/mt5-small
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"onnx",
"mt5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:2010.11934",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
[Google's mT5](https://github.com/google-research/multilingual-t5)
mT5 is pretrained on the [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) corpus, covering 101 languages:
Afrikaans, Albanian, Amharic, Arabic, Armenian, Azerbaijani, Basque, Belarusian, Bengali, Bulgarian, Burmese, Catalan, Cebuano, Chichewa, Chinese, Corsican, Czech, Danish, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Korean, Kurdish, Kyrgyz, Lao, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Nepali, Norwegian, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scottish Gaelic, Serbian, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Sotho, Spanish, Sundanese, Swahili, Swedish, Tajik, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Uzbek, Vietnamese, Welsh, West Frisian, Xhosa, Yiddish, Yoruba, Zulu.
**Note**: mT5 was only pre-trained on mC4 excluding any supervised training. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
Pretraining Dataset: [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual)
Other Community Checkpoints: [here](https://huggingface.co/models?search=mt5)
Paper: [mT5: A massively multilingual pre-trained text-to-text transformer](https://arxiv.org/abs/2010.11934)
Authors: *Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, Colin Raffel*
## Abstract
The recent "Text-to-Text Transfer Transformer" (T5) leveraged a unified text-to-text format and scale to attain state-of-the-art results on a wide variety of English-language NLP tasks. In this paper, we introduce mT5, a multilingual variant of T5 that was pre-trained on a new Common Crawl-based dataset covering 101 languages. We describe the design and modified training of mT5 and demonstrate its state-of-the-art performance on many multilingual benchmarks. All of the code and model checkpoints used in this work are publicly available.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/mt5-xl
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"mt5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:2010.11934",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
[Google's mT5](https://github.com/google-research/multilingual-t5)
mT5 is pretrained on the [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) corpus, covering 101 languages:
Afrikaans, Albanian, Amharic, Arabic, Armenian, Azerbaijani, Basque, Belarusian, Bengali, Bulgarian, Burmese, Catalan, Cebuano, Chichewa, Chinese, Corsican, Czech, Danish, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Korean, Kurdish, Kyrgyz, Lao, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Nepali, Norwegian, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scottish Gaelic, Serbian, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Sotho, Spanish, Sundanese, Swahili, Swedish, Tajik, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Uzbek, Vietnamese, Welsh, West Frisian, Xhosa, Yiddish, Yoruba, Zulu.
**Note**: mT5 was only pre-trained on mC4 excluding any supervised training. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
Pretraining Dataset: [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual)
Other Community Checkpoints: [here](https://huggingface.co/models?search=mt5)
Paper: [mT5: A massively multilingual pre-trained text-to-text transformer](https://arxiv.org/abs/2010.11934)
Authors: *Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, Colin Raffel*
## Abstract
The recent "Text-to-Text Transfer Transformer" (T5) leveraged a unified text-to-text format and scale to attain state-of-the-art results on a wide variety of English-language NLP tasks. In this paper, we introduce mT5, a multilingual variant of T5 that was pre-trained on a new Common Crawl-based dataset covering 101 languages. We describe the design and modified training of mT5 and demonstrate its state-of-the-art performance on many multilingual benchmarks. All of the code and model checkpoints used in this work are publicly available.
|
{"language": ["multilingual", "af", "am", "ar", "az", "be", "bg", "bn", "ca", "ceb", "co", "cs", "cy", "da", "de", "el", "en", "eo", "es", "et", "eu", "fa", "fi", "fil", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "haw", "hi", "hmn", "ht", "hu", "hy", "ig", "is", "it", "iw", "ja", "jv", "ka", "kk", "km", "kn", "ko", "ku", "ky", "la", "lb", "lo", "lt", "lv", "mg", "mi", "mk", "ml", "mn", "mr", "ms", "mt", "my", "ne", "nl", false, "ny", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "sm", "sn", "so", "sq", "sr", "st", "su", "sv", "sw", "ta", "te", "tg", "th", "tr", "uk", "und", "ur", "uz", "vi", "xh", "yi", "yo", "zh", "zu"], "license": "apache-2.0", "datasets": ["mc4"]}
|
google/mt5-xxl
| null |
[
"transformers",
"pytorch",
"tf",
"mt5",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"az",
"be",
"bg",
"bn",
"ca",
"ceb",
"co",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fil",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"haw",
"hi",
"hmn",
"ht",
"hu",
"hy",
"ig",
"is",
"it",
"iw",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"ku",
"ky",
"la",
"lb",
"lo",
"lt",
"lv",
"mg",
"mi",
"mk",
"ml",
"mn",
"mr",
"ms",
"mt",
"my",
"ne",
"nl",
"no",
"ny",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"tr",
"uk",
"und",
"ur",
"uz",
"vi",
"xh",
"yi",
"yo",
"zh",
"zu",
"dataset:mc4",
"arxiv:2010.11934",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# MultiBERTs, Intermediate Checkpoint - Seed 0, Step 0k
MultiBERTs is a collection of checkpoints and a statistical library to support
robust research on BERT. We provide 25 BERT-base models trained with
similar hyper-parameters as
[the original BERT model](https://github.com/google-research/bert) but
with different random seeds, which causes variations in the initial weights and order of
training instances. The aim is to distinguish findings that apply to a specific
artifact (i.e., a particular instance of the model) from those that apply to the
more general procedure.
We also provide 140 intermediate checkpoints captured
during the course of pre-training (we saved 28 checkpoints for the first 5 runs).
The models were originally released through
[http://goo.gle/multiberts](http://goo.gle/multiberts). We describe them in our
paper
[The MultiBERTs: BERT Reproductions for Robustness Analysis](https://arxiv.org/abs/2106.16163).
This is model #0, captured at step 0k (max: 2000k, i.e., 2M steps).
## Model Description
This model was captured during a reproduction of
[BERT-base uncased](https://github.com/google-research/bert), for English: it
is a Transformers model pretrained on a large corpus of English data, using the
Masked Language Modelling (MLM) and the Next Sentence Prediction (NSP)
objectives.
The intended uses, limitations, training data and training procedure for the fully trained model are similar
to [BERT-base uncased](https://github.com/google-research/bert). Two major
differences with the original model:
* We pre-trained the MultiBERTs models for 2 million steps using sequence
length 512 (instead of 1 million steps using sequence length 128 then 512).
* We used an alternative version of Wikipedia and Books Corpus, initially
collected for [Turc et al., 2019](https://arxiv.org/abs/1908.08962).
This is a best-effort reproduction, and so it is probable that differences with
the original model have gone unnoticed. The performance of MultiBERTs on GLUE after full training is oftentimes comparable to that of original
BERT, but we found significant differences on the dev set of SQuAD (MultiBERTs outperforms original BERT).
See our [technical report](https://arxiv.org/abs/2106.16163) for more details.
### How to use
Using code from
[BERT-base uncased](https://huggingface.co/bert-base-uncased), here is an example based on
Tensorflow:
```
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_0k')
model = TFBertModel.from_pretrained("google/multiberts-seed_0-step_0k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
PyTorch version:
```
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_0k')
model = BertModel.from_pretrained("google/multiberts-seed_0-step_0k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Citation info
```bibtex
@article{sellam2021multiberts,
title={The MultiBERTs: BERT Reproductions for Robustness Analysis},
author={Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick},
journal={arXiv preprint arXiv:2106.16163},
year={2021}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["multiberts", "multiberts-seed_0", "multiberts-seed_0-step_0k"]}
|
google/multiberts-seed_0-step_0k
| null |
[
"transformers",
"pytorch",
"tf",
"bert",
"pretraining",
"multiberts",
"multiberts-seed_0",
"multiberts-seed_0-step_0k",
"en",
"arxiv:2106.16163",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# MultiBERTs, Intermediate Checkpoint - Seed 0, Step 1000k
MultiBERTs is a collection of checkpoints and a statistical library to support
robust research on BERT. We provide 25 BERT-base models trained with
similar hyper-parameters as
[the original BERT model](https://github.com/google-research/bert) but
with different random seeds, which causes variations in the initial weights and order of
training instances. The aim is to distinguish findings that apply to a specific
artifact (i.e., a particular instance of the model) from those that apply to the
more general procedure.
We also provide 140 intermediate checkpoints captured
during the course of pre-training (we saved 28 checkpoints for the first 5 runs).
The models were originally released through
[http://goo.gle/multiberts](http://goo.gle/multiberts). We describe them in our
paper
[The MultiBERTs: BERT Reproductions for Robustness Analysis](https://arxiv.org/abs/2106.16163).
This is model #0, captured at step 1000k (max: 2000k, i.e., 2M steps).
## Model Description
This model was captured during a reproduction of
[BERT-base uncased](https://github.com/google-research/bert), for English: it
is a Transformers model pretrained on a large corpus of English data, using the
Masked Language Modelling (MLM) and the Next Sentence Prediction (NSP)
objectives.
The intended uses, limitations, training data and training procedure for the fully trained model are similar
to [BERT-base uncased](https://github.com/google-research/bert). Two major
differences with the original model:
* We pre-trained the MultiBERTs models for 2 million steps using sequence
length 512 (instead of 1 million steps using sequence length 128 then 512).
* We used an alternative version of Wikipedia and Books Corpus, initially
collected for [Turc et al., 2019](https://arxiv.org/abs/1908.08962).
This is a best-effort reproduction, and so it is probable that differences with
the original model have gone unnoticed. The performance of MultiBERTs on GLUE after full training is oftentimes comparable to that of original
BERT, but we found significant differences on the dev set of SQuAD (MultiBERTs outperforms original BERT).
See our [technical report](https://arxiv.org/abs/2106.16163) for more details.
### How to use
Using code from
[BERT-base uncased](https://huggingface.co/bert-base-uncased), here is an example based on
Tensorflow:
```
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1000k')
model = TFBertModel.from_pretrained("google/multiberts-seed_0-step_1000k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
PyTorch version:
```
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1000k')
model = BertModel.from_pretrained("google/multiberts-seed_0-step_1000k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Citation info
```bibtex
@article{sellam2021multiberts,
title={The MultiBERTs: BERT Reproductions for Robustness Analysis},
author={Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick},
journal={arXiv preprint arXiv:2106.16163},
year={2021}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["multiberts", "multiberts-seed_0", "multiberts-seed_0-step_1000k"]}
|
google/multiberts-seed_0-step_1000k
| null |
[
"transformers",
"pytorch",
"tf",
"bert",
"pretraining",
"multiberts",
"multiberts-seed_0",
"multiberts-seed_0-step_1000k",
"en",
"arxiv:2106.16163",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# MultiBERTs, Intermediate Checkpoint - Seed 0, Step 100k
MultiBERTs is a collection of checkpoints and a statistical library to support
robust research on BERT. We provide 25 BERT-base models trained with
similar hyper-parameters as
[the original BERT model](https://github.com/google-research/bert) but
with different random seeds, which causes variations in the initial weights and order of
training instances. The aim is to distinguish findings that apply to a specific
artifact (i.e., a particular instance of the model) from those that apply to the
more general procedure.
We also provide 140 intermediate checkpoints captured
during the course of pre-training (we saved 28 checkpoints for the first 5 runs).
The models were originally released through
[http://goo.gle/multiberts](http://goo.gle/multiberts). We describe them in our
paper
[The MultiBERTs: BERT Reproductions for Robustness Analysis](https://arxiv.org/abs/2106.16163).
This is model #0, captured at step 100k (max: 2000k, i.e., 2M steps).
## Model Description
This model was captured during a reproduction of
[BERT-base uncased](https://github.com/google-research/bert), for English: it
is a Transformers model pretrained on a large corpus of English data, using the
Masked Language Modelling (MLM) and the Next Sentence Prediction (NSP)
objectives.
The intended uses, limitations, training data and training procedure for the fully trained model are similar
to [BERT-base uncased](https://github.com/google-research/bert). Two major
differences with the original model:
* We pre-trained the MultiBERTs models for 2 million steps using sequence
length 512 (instead of 1 million steps using sequence length 128 then 512).
* We used an alternative version of Wikipedia and Books Corpus, initially
collected for [Turc et al., 2019](https://arxiv.org/abs/1908.08962).
This is a best-effort reproduction, and so it is probable that differences with
the original model have gone unnoticed. The performance of MultiBERTs on GLUE after full training is oftentimes comparable to that of original
BERT, but we found significant differences on the dev set of SQuAD (MultiBERTs outperforms original BERT).
See our [technical report](https://arxiv.org/abs/2106.16163) for more details.
### How to use
Using code from
[BERT-base uncased](https://huggingface.co/bert-base-uncased), here is an example based on
Tensorflow:
```
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_100k')
model = TFBertModel.from_pretrained("google/multiberts-seed_0-step_100k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
PyTorch version:
```
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_100k')
model = BertModel.from_pretrained("google/multiberts-seed_0-step_100k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Citation info
```bibtex
@article{sellam2021multiberts,
title={The MultiBERTs: BERT Reproductions for Robustness Analysis},
author={Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick},
journal={arXiv preprint arXiv:2106.16163},
year={2021}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["multiberts", "multiberts-seed_0", "multiberts-seed_0-step_100k"]}
|
google/multiberts-seed_0-step_100k
| null |
[
"transformers",
"pytorch",
"tf",
"bert",
"pretraining",
"multiberts",
"multiberts-seed_0",
"multiberts-seed_0-step_100k",
"en",
"arxiv:2106.16163",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# MultiBERTs, Intermediate Checkpoint - Seed 0, Step 1100k
MultiBERTs is a collection of checkpoints and a statistical library to support
robust research on BERT. We provide 25 BERT-base models trained with
similar hyper-parameters as
[the original BERT model](https://github.com/google-research/bert) but
with different random seeds, which causes variations in the initial weights and order of
training instances. The aim is to distinguish findings that apply to a specific
artifact (i.e., a particular instance of the model) from those that apply to the
more general procedure.
We also provide 140 intermediate checkpoints captured
during the course of pre-training (we saved 28 checkpoints for the first 5 runs).
The models were originally released through
[http://goo.gle/multiberts](http://goo.gle/multiberts). We describe them in our
paper
[The MultiBERTs: BERT Reproductions for Robustness Analysis](https://arxiv.org/abs/2106.16163).
This is model #0, captured at step 1100k (max: 2000k, i.e., 2M steps).
## Model Description
This model was captured during a reproduction of
[BERT-base uncased](https://github.com/google-research/bert), for English: it
is a Transformers model pretrained on a large corpus of English data, using the
Masked Language Modelling (MLM) and the Next Sentence Prediction (NSP)
objectives.
The intended uses, limitations, training data and training procedure for the fully trained model are similar
to [BERT-base uncased](https://github.com/google-research/bert). Two major
differences with the original model:
* We pre-trained the MultiBERTs models for 2 million steps using sequence
length 512 (instead of 1 million steps using sequence length 128 then 512).
* We used an alternative version of Wikipedia and Books Corpus, initially
collected for [Turc et al., 2019](https://arxiv.org/abs/1908.08962).
This is a best-effort reproduction, and so it is probable that differences with
the original model have gone unnoticed. The performance of MultiBERTs on GLUE after full training is oftentimes comparable to that of original
BERT, but we found significant differences on the dev set of SQuAD (MultiBERTs outperforms original BERT).
See our [technical report](https://arxiv.org/abs/2106.16163) for more details.
### How to use
Using code from
[BERT-base uncased](https://huggingface.co/bert-base-uncased), here is an example based on
Tensorflow:
```
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1100k')
model = TFBertModel.from_pretrained("google/multiberts-seed_0-step_1100k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
PyTorch version:
```
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1100k')
model = BertModel.from_pretrained("google/multiberts-seed_0-step_1100k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Citation info
```bibtex
@article{sellam2021multiberts,
title={The MultiBERTs: BERT Reproductions for Robustness Analysis},
author={Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick},
journal={arXiv preprint arXiv:2106.16163},
year={2021}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["multiberts", "multiberts-seed_0", "multiberts-seed_0-step_1100k"]}
|
google/multiberts-seed_0-step_1100k
| null |
[
"transformers",
"pytorch",
"tf",
"bert",
"pretraining",
"multiberts",
"multiberts-seed_0",
"multiberts-seed_0-step_1100k",
"en",
"arxiv:2106.16163",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# MultiBERTs, Intermediate Checkpoint - Seed 0, Step 1200k
MultiBERTs is a collection of checkpoints and a statistical library to support
robust research on BERT. We provide 25 BERT-base models trained with
similar hyper-parameters as
[the original BERT model](https://github.com/google-research/bert) but
with different random seeds, which causes variations in the initial weights and order of
training instances. The aim is to distinguish findings that apply to a specific
artifact (i.e., a particular instance of the model) from those that apply to the
more general procedure.
We also provide 140 intermediate checkpoints captured
during the course of pre-training (we saved 28 checkpoints for the first 5 runs).
The models were originally released through
[http://goo.gle/multiberts](http://goo.gle/multiberts). We describe them in our
paper
[The MultiBERTs: BERT Reproductions for Robustness Analysis](https://arxiv.org/abs/2106.16163).
This is model #0, captured at step 1200k (max: 2000k, i.e., 2M steps).
## Model Description
This model was captured during a reproduction of
[BERT-base uncased](https://github.com/google-research/bert), for English: it
is a Transformers model pretrained on a large corpus of English data, using the
Masked Language Modelling (MLM) and the Next Sentence Prediction (NSP)
objectives.
The intended uses, limitations, training data and training procedure for the fully trained model are similar
to [BERT-base uncased](https://github.com/google-research/bert). Two major
differences with the original model:
* We pre-trained the MultiBERTs models for 2 million steps using sequence
length 512 (instead of 1 million steps using sequence length 128 then 512).
* We used an alternative version of Wikipedia and Books Corpus, initially
collected for [Turc et al., 2019](https://arxiv.org/abs/1908.08962).
This is a best-effort reproduction, and so it is probable that differences with
the original model have gone unnoticed. The performance of MultiBERTs on GLUE after full training is oftentimes comparable to that of original
BERT, but we found significant differences on the dev set of SQuAD (MultiBERTs outperforms original BERT).
See our [technical report](https://arxiv.org/abs/2106.16163) for more details.
### How to use
Using code from
[BERT-base uncased](https://huggingface.co/bert-base-uncased), here is an example based on
Tensorflow:
```
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1200k')
model = TFBertModel.from_pretrained("google/multiberts-seed_0-step_1200k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
PyTorch version:
```
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1200k')
model = BertModel.from_pretrained("google/multiberts-seed_0-step_1200k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Citation info
```bibtex
@article{sellam2021multiberts,
title={The MultiBERTs: BERT Reproductions for Robustness Analysis},
author={Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick},
journal={arXiv preprint arXiv:2106.16163},
year={2021}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["multiberts", "multiberts-seed_0", "multiberts-seed_0-step_1200k"]}
|
google/multiberts-seed_0-step_1200k
| null |
[
"transformers",
"pytorch",
"tf",
"bert",
"pretraining",
"multiberts",
"multiberts-seed_0",
"multiberts-seed_0-step_1200k",
"en",
"arxiv:2106.16163",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# MultiBERTs, Intermediate Checkpoint - Seed 0, Step 120k
MultiBERTs is a collection of checkpoints and a statistical library to support
robust research on BERT. We provide 25 BERT-base models trained with
similar hyper-parameters as
[the original BERT model](https://github.com/google-research/bert) but
with different random seeds, which causes variations in the initial weights and order of
training instances. The aim is to distinguish findings that apply to a specific
artifact (i.e., a particular instance of the model) from those that apply to the
more general procedure.
We also provide 140 intermediate checkpoints captured
during the course of pre-training (we saved 28 checkpoints for the first 5 runs).
The models were originally released through
[http://goo.gle/multiberts](http://goo.gle/multiberts). We describe them in our
paper
[The MultiBERTs: BERT Reproductions for Robustness Analysis](https://arxiv.org/abs/2106.16163).
This is model #0, captured at step 120k (max: 2000k, i.e., 2M steps).
## Model Description
This model was captured during a reproduction of
[BERT-base uncased](https://github.com/google-research/bert), for English: it
is a Transformers model pretrained on a large corpus of English data, using the
Masked Language Modelling (MLM) and the Next Sentence Prediction (NSP)
objectives.
The intended uses, limitations, training data and training procedure for the fully trained model are similar
to [BERT-base uncased](https://github.com/google-research/bert). Two major
differences with the original model:
* We pre-trained the MultiBERTs models for 2 million steps using sequence
length 512 (instead of 1 million steps using sequence length 128 then 512).
* We used an alternative version of Wikipedia and Books Corpus, initially
collected for [Turc et al., 2019](https://arxiv.org/abs/1908.08962).
This is a best-effort reproduction, and so it is probable that differences with
the original model have gone unnoticed. The performance of MultiBERTs on GLUE after full training is oftentimes comparable to that of original
BERT, but we found significant differences on the dev set of SQuAD (MultiBERTs outperforms original BERT).
See our [technical report](https://arxiv.org/abs/2106.16163) for more details.
### How to use
Using code from
[BERT-base uncased](https://huggingface.co/bert-base-uncased), here is an example based on
Tensorflow:
```
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_120k')
model = TFBertModel.from_pretrained("google/multiberts-seed_0-step_120k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
PyTorch version:
```
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_120k')
model = BertModel.from_pretrained("google/multiberts-seed_0-step_120k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Citation info
```bibtex
@article{sellam2021multiberts,
title={The MultiBERTs: BERT Reproductions for Robustness Analysis},
author={Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick},
journal={arXiv preprint arXiv:2106.16163},
year={2021}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["multiberts", "multiberts-seed_0", "multiberts-seed_0-step_120k"]}
|
google/multiberts-seed_0-step_120k
| null |
[
"transformers",
"pytorch",
"tf",
"bert",
"pretraining",
"multiberts",
"multiberts-seed_0",
"multiberts-seed_0-step_120k",
"en",
"arxiv:2106.16163",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# MultiBERTs, Intermediate Checkpoint - Seed 0, Step 1300k
MultiBERTs is a collection of checkpoints and a statistical library to support
robust research on BERT. We provide 25 BERT-base models trained with
similar hyper-parameters as
[the original BERT model](https://github.com/google-research/bert) but
with different random seeds, which causes variations in the initial weights and order of
training instances. The aim is to distinguish findings that apply to a specific
artifact (i.e., a particular instance of the model) from those that apply to the
more general procedure.
We also provide 140 intermediate checkpoints captured
during the course of pre-training (we saved 28 checkpoints for the first 5 runs).
The models were originally released through
[http://goo.gle/multiberts](http://goo.gle/multiberts). We describe them in our
paper
[The MultiBERTs: BERT Reproductions for Robustness Analysis](https://arxiv.org/abs/2106.16163).
This is model #0, captured at step 1300k (max: 2000k, i.e., 2M steps).
## Model Description
This model was captured during a reproduction of
[BERT-base uncased](https://github.com/google-research/bert), for English: it
is a Transformers model pretrained on a large corpus of English data, using the
Masked Language Modelling (MLM) and the Next Sentence Prediction (NSP)
objectives.
The intended uses, limitations, training data and training procedure for the fully trained model are similar
to [BERT-base uncased](https://github.com/google-research/bert). Two major
differences with the original model:
* We pre-trained the MultiBERTs models for 2 million steps using sequence
length 512 (instead of 1 million steps using sequence length 128 then 512).
* We used an alternative version of Wikipedia and Books Corpus, initially
collected for [Turc et al., 2019](https://arxiv.org/abs/1908.08962).
This is a best-effort reproduction, and so it is probable that differences with
the original model have gone unnoticed. The performance of MultiBERTs on GLUE after full training is oftentimes comparable to that of original
BERT, but we found significant differences on the dev set of SQuAD (MultiBERTs outperforms original BERT).
See our [technical report](https://arxiv.org/abs/2106.16163) for more details.
### How to use
Using code from
[BERT-base uncased](https://huggingface.co/bert-base-uncased), here is an example based on
Tensorflow:
```
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1300k')
model = TFBertModel.from_pretrained("google/multiberts-seed_0-step_1300k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
PyTorch version:
```
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1300k')
model = BertModel.from_pretrained("google/multiberts-seed_0-step_1300k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Citation info
```bibtex
@article{sellam2021multiberts,
title={The MultiBERTs: BERT Reproductions for Robustness Analysis},
author={Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick},
journal={arXiv preprint arXiv:2106.16163},
year={2021}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["multiberts", "multiberts-seed_0", "multiberts-seed_0-step_1300k"]}
|
google/multiberts-seed_0-step_1300k
| null |
[
"transformers",
"pytorch",
"tf",
"bert",
"pretraining",
"multiberts",
"multiberts-seed_0",
"multiberts-seed_0-step_1300k",
"en",
"arxiv:2106.16163",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# MultiBERTs, Intermediate Checkpoint - Seed 0, Step 1400k
MultiBERTs is a collection of checkpoints and a statistical library to support
robust research on BERT. We provide 25 BERT-base models trained with
similar hyper-parameters as
[the original BERT model](https://github.com/google-research/bert) but
with different random seeds, which causes variations in the initial weights and order of
training instances. The aim is to distinguish findings that apply to a specific
artifact (i.e., a particular instance of the model) from those that apply to the
more general procedure.
We also provide 140 intermediate checkpoints captured
during the course of pre-training (we saved 28 checkpoints for the first 5 runs).
The models were originally released through
[http://goo.gle/multiberts](http://goo.gle/multiberts). We describe them in our
paper
[The MultiBERTs: BERT Reproductions for Robustness Analysis](https://arxiv.org/abs/2106.16163).
This is model #0, captured at step 1400k (max: 2000k, i.e., 2M steps).
## Model Description
This model was captured during a reproduction of
[BERT-base uncased](https://github.com/google-research/bert), for English: it
is a Transformers model pretrained on a large corpus of English data, using the
Masked Language Modelling (MLM) and the Next Sentence Prediction (NSP)
objectives.
The intended uses, limitations, training data and training procedure for the fully trained model are similar
to [BERT-base uncased](https://github.com/google-research/bert). Two major
differences with the original model:
* We pre-trained the MultiBERTs models for 2 million steps using sequence
length 512 (instead of 1 million steps using sequence length 128 then 512).
* We used an alternative version of Wikipedia and Books Corpus, initially
collected for [Turc et al., 2019](https://arxiv.org/abs/1908.08962).
This is a best-effort reproduction, and so it is probable that differences with
the original model have gone unnoticed. The performance of MultiBERTs on GLUE after full training is oftentimes comparable to that of original
BERT, but we found significant differences on the dev set of SQuAD (MultiBERTs outperforms original BERT).
See our [technical report](https://arxiv.org/abs/2106.16163) for more details.
### How to use
Using code from
[BERT-base uncased](https://huggingface.co/bert-base-uncased), here is an example based on
Tensorflow:
```
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1400k')
model = TFBertModel.from_pretrained("google/multiberts-seed_0-step_1400k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
PyTorch version:
```
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('google/multiberts-seed_0-step_1400k')
model = BertModel.from_pretrained("google/multiberts-seed_0-step_1400k")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
## Citation info
```bibtex
@article{sellam2021multiberts,
title={The MultiBERTs: BERT Reproductions for Robustness Analysis},
author={Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick},
journal={arXiv preprint arXiv:2106.16163},
year={2021}
}
```
|
{"language": "en", "license": "apache-2.0", "tags": ["multiberts", "multiberts-seed_0", "multiberts-seed_0-step_1400k"]}
|
google/multiberts-seed_0-step_1400k
| null |
[
"transformers",
"pytorch",
"tf",
"bert",
"pretraining",
"multiberts",
"multiberts-seed_0",
"multiberts-seed_0-step_1400k",
"en",
"arxiv:2106.16163",
"arxiv:1908.08962",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.