metadata
license: apache-2.0
base_model: asapp/sew-d-tiny-100k-ft-ls100h
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: sewd_classifier
results: []
sewd_classifier
This model is a fine-tuned version of asapp/sew-d-tiny-100k-ft-ls100h on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.8841
- Accuracy: 0.2518
- Precision: 0.1735
- Recall: 0.2518
- F1: 0.1722
- Binary: 0.4673
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: 1e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Binary |
|---|---|---|---|---|---|---|---|---|
| No log | 0.17 | 50 | 4.4308 | 0.0097 | 0.0003 | 0.0097 | 0.0005 | 0.1507 |
| No log | 0.35 | 100 | 4.4108 | 0.0170 | 0.0018 | 0.0170 | 0.0032 | 0.1881 |
| No log | 0.52 | 150 | 4.3449 | 0.0267 | 0.0043 | 0.0267 | 0.0060 | 0.2415 |
| No log | 0.69 | 200 | 4.2574 | 0.0437 | 0.0268 | 0.0437 | 0.0152 | 0.2711 |
| No log | 0.86 | 250 | 4.1553 | 0.0510 | 0.0154 | 0.0510 | 0.0168 | 0.3005 |
| 4.3075 | 1.04 | 300 | 4.0633 | 0.0728 | 0.0249 | 0.0728 | 0.0306 | 0.3226 |
| 4.3075 | 1.21 | 350 | 3.9895 | 0.0728 | 0.0598 | 0.0728 | 0.0338 | 0.3243 |
| 4.3075 | 1.38 | 400 | 3.9076 | 0.0752 | 0.0415 | 0.0752 | 0.0307 | 0.3296 |
| 4.3075 | 1.55 | 450 | 3.8252 | 0.0874 | 0.0392 | 0.0874 | 0.0408 | 0.3410 |
| 4.3075 | 1.73 | 500 | 3.7704 | 0.0825 | 0.0408 | 0.0825 | 0.0357 | 0.3425 |
| 4.3075 | 1.9 | 550 | 3.6929 | 0.0874 | 0.0619 | 0.0874 | 0.0390 | 0.3488 |
| 3.851 | 2.07 | 600 | 3.6301 | 0.0898 | 0.0534 | 0.0898 | 0.0412 | 0.3500 |
| 3.851 | 2.24 | 650 | 3.6033 | 0.0995 | 0.0559 | 0.0995 | 0.0480 | 0.3595 |
| 3.851 | 2.42 | 700 | 3.5565 | 0.0874 | 0.0443 | 0.0874 | 0.0398 | 0.3481 |
| 3.851 | 2.59 | 750 | 3.5117 | 0.1141 | 0.0660 | 0.1141 | 0.0652 | 0.3684 |
| 3.851 | 2.76 | 800 | 3.4704 | 0.1141 | 0.0582 | 0.1141 | 0.0638 | 0.3680 |
| 3.851 | 2.93 | 850 | 3.4492 | 0.1141 | 0.0682 | 0.1141 | 0.0583 | 0.3672 |
| 3.5754 | 3.11 | 900 | 3.4026 | 0.1189 | 0.0770 | 0.1189 | 0.0668 | 0.3728 |
| 3.5754 | 3.28 | 950 | 3.3955 | 0.1214 | 0.0477 | 0.1214 | 0.0612 | 0.3709 |
| 3.5754 | 3.45 | 1000 | 3.3561 | 0.1262 | 0.0585 | 0.1262 | 0.0683 | 0.3757 |
| 3.5754 | 3.62 | 1050 | 3.3244 | 0.1262 | 0.0806 | 0.1262 | 0.0731 | 0.3750 |
| 3.5754 | 3.8 | 1100 | 3.3235 | 0.1286 | 0.0629 | 0.1286 | 0.0655 | 0.3760 |
| 3.5754 | 3.97 | 1150 | 3.2740 | 0.1262 | 0.0946 | 0.1262 | 0.0793 | 0.3772 |
| 3.406 | 4.14 | 1200 | 3.2410 | 0.1311 | 0.1106 | 0.1311 | 0.0762 | 0.3813 |
| 3.406 | 4.31 | 1250 | 3.2280 | 0.1335 | 0.0856 | 0.1335 | 0.0740 | 0.3816 |
| 3.406 | 4.49 | 1300 | 3.2170 | 0.1408 | 0.0648 | 0.1408 | 0.0740 | 0.3867 |
| 3.406 | 4.66 | 1350 | 3.1892 | 0.1238 | 0.0752 | 0.1238 | 0.0712 | 0.3755 |
| 3.406 | 4.83 | 1400 | 3.1660 | 0.1505 | 0.1149 | 0.1505 | 0.0970 | 0.3949 |
| 3.2992 | 5.0 | 1450 | 3.1378 | 0.1456 | 0.0956 | 0.1456 | 0.0885 | 0.3908 |
| 3.2992 | 5.18 | 1500 | 3.1343 | 0.1481 | 0.0903 | 0.1481 | 0.0867 | 0.3932 |
| 3.2992 | 5.35 | 1550 | 3.1113 | 0.1553 | 0.0996 | 0.1553 | 0.0953 | 0.3976 |
| 3.2992 | 5.52 | 1600 | 3.0820 | 0.1626 | 0.1355 | 0.1626 | 0.1094 | 0.4058 |
| 3.2992 | 5.69 | 1650 | 3.0808 | 0.1699 | 0.1526 | 0.1699 | 0.1191 | 0.4102 |
| 3.2992 | 5.87 | 1700 | 3.0625 | 0.1772 | 0.1424 | 0.1772 | 0.1234 | 0.4160 |
| 3.2023 | 6.04 | 1750 | 3.0465 | 0.1820 | 0.1168 | 0.1820 | 0.1165 | 0.4170 |
| 3.2023 | 6.21 | 1800 | 3.0341 | 0.1675 | 0.1161 | 0.1675 | 0.1080 | 0.4075 |
| 3.2023 | 6.38 | 1850 | 3.0194 | 0.1869 | 0.1078 | 0.1869 | 0.1146 | 0.4204 |
| 3.2023 | 6.56 | 1900 | 3.0086 | 0.1942 | 0.1256 | 0.1942 | 0.1208 | 0.4272 |
| 3.2023 | 6.73 | 1950 | 3.0076 | 0.1845 | 0.1154 | 0.1845 | 0.1166 | 0.4187 |
| 3.2023 | 6.9 | 2000 | 2.9890 | 0.1845 | 0.1440 | 0.1845 | 0.1200 | 0.4204 |
| 3.1308 | 7.08 | 2050 | 2.9780 | 0.1966 | 0.1506 | 0.1966 | 0.1305 | 0.4296 |
| 3.1308 | 7.25 | 2100 | 2.9687 | 0.1820 | 0.1354 | 0.1820 | 0.1178 | 0.4194 |
| 3.1308 | 7.42 | 2150 | 2.9540 | 0.1966 | 0.1385 | 0.1966 | 0.1284 | 0.4286 |
| 3.1308 | 7.59 | 2200 | 2.9549 | 0.2063 | 0.1477 | 0.2063 | 0.1359 | 0.4354 |
| 3.1308 | 7.77 | 2250 | 2.9422 | 0.1990 | 0.1375 | 0.1990 | 0.1276 | 0.4301 |
| 3.1308 | 7.94 | 2300 | 2.9304 | 0.2039 | 0.1606 | 0.2039 | 0.1354 | 0.4335 |
| 3.0777 | 8.11 | 2350 | 2.9227 | 0.2063 | 0.1636 | 0.2063 | 0.1372 | 0.4359 |
| 3.0777 | 8.28 | 2400 | 2.9125 | 0.2136 | 0.1800 | 0.2136 | 0.1470 | 0.4410 |
| 3.0777 | 8.46 | 2450 | 2.9142 | 0.2136 | 0.1805 | 0.2136 | 0.1443 | 0.4403 |
| 3.0777 | 8.63 | 2500 | 2.9170 | 0.2184 | 0.1860 | 0.2184 | 0.1500 | 0.4437 |
| 3.0777 | 8.8 | 2550 | 2.9022 | 0.2136 | 0.1828 | 0.2136 | 0.1456 | 0.4403 |
| 3.0777 | 8.97 | 2600 | 2.8939 | 0.2136 | 0.1660 | 0.2136 | 0.1425 | 0.4403 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.15.1