metadata
license: apache-2.0
base_model: t5-base
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- accuracy
model-index:
- name: t5-base_cola_dense
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: glue
type: glue
config: cola
split: validation
args: cola
metrics:
- name: Accuracy
type: accuracy
value: 0.8370086289549377
t5-base_cola_dense
This model is a fine-tuned version of t5-base on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 0.4482
- Accuracy: 0.8370
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: 32
- eval_batch_size: 64
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.6484 | 0.07 | 10 | 0.6254 | 0.6913 |
0.6188 | 0.15 | 20 | 0.6181 | 0.6913 |
0.625 | 0.22 | 30 | 0.6137 | 0.6913 |
0.5744 | 0.3 | 40 | 0.6063 | 0.6913 |
0.6055 | 0.37 | 50 | 0.5963 | 0.6913 |
0.5723 | 0.45 | 60 | 0.5788 | 0.6913 |
0.5777 | 0.52 | 70 | 0.5527 | 0.6913 |
0.5332 | 0.6 | 80 | 0.5117 | 0.7354 |
0.4662 | 0.67 | 90 | 0.5060 | 0.7843 |
0.4936 | 0.75 | 100 | 0.4717 | 0.7929 |
0.4898 | 0.82 | 110 | 0.5304 | 0.8015 |
0.4844 | 0.9 | 120 | 0.4771 | 0.8006 |
0.4297 | 0.97 | 130 | 0.4673 | 0.7987 |
0.4658 | 1.04 | 140 | 0.4927 | 0.8063 |
0.3992 | 1.12 | 150 | 0.4884 | 0.8121 |
0.4752 | 1.19 | 160 | 0.4838 | 0.8102 |
0.3934 | 1.27 | 170 | 0.4714 | 0.8092 |
0.4662 | 1.34 | 180 | 0.5192 | 0.7929 |
0.4404 | 1.42 | 190 | 0.4719 | 0.8111 |
0.3746 | 1.49 | 200 | 0.5077 | 0.8015 |
0.4465 | 1.57 | 210 | 0.4425 | 0.8073 |
0.3829 | 1.64 | 220 | 0.4844 | 0.8130 |
0.4021 | 1.72 | 230 | 0.4659 | 0.8169 |
0.4225 | 1.79 | 240 | 0.4277 | 0.8130 |
0.4297 | 1.87 | 250 | 0.4677 | 0.8150 |
0.3476 | 1.94 | 260 | 0.4455 | 0.8207 |
0.4159 | 2.01 | 270 | 0.5063 | 0.8188 |
0.3371 | 2.09 | 280 | 0.4648 | 0.8265 |
0.3383 | 2.16 | 290 | 0.5451 | 0.8178 |
0.3175 | 2.24 | 300 | 0.4551 | 0.8303 |
0.3553 | 2.31 | 310 | 0.4899 | 0.8303 |
0.3138 | 2.39 | 320 | 0.4887 | 0.8265 |
0.3196 | 2.46 | 330 | 0.4632 | 0.8265 |
0.3132 | 2.54 | 340 | 0.5126 | 0.8207 |
0.3167 | 2.61 | 350 | 0.4661 | 0.8245 |
0.3757 | 2.69 | 360 | 0.4596 | 0.8245 |
0.3346 | 2.76 | 370 | 0.4650 | 0.8265 |
0.3018 | 2.84 | 380 | 0.4672 | 0.8284 |
0.3338 | 2.91 | 390 | 0.4822 | 0.8293 |
0.3496 | 2.99 | 400 | 0.4677 | 0.8322 |
0.248 | 3.06 | 410 | 0.4349 | 0.8332 |
0.2804 | 3.13 | 420 | 0.5308 | 0.8322 |
0.292 | 3.21 | 430 | 0.4757 | 0.8284 |
0.249 | 3.28 | 440 | 0.5145 | 0.8284 |
0.315 | 3.36 | 450 | 0.6137 | 0.8322 |
0.2996 | 3.43 | 460 | 0.5499 | 0.8341 |
0.2986 | 3.51 | 470 | 0.4774 | 0.8332 |
0.3124 | 3.58 | 480 | 0.5733 | 0.8284 |
0.2809 | 3.66 | 490 | 0.4938 | 0.8341 |
0.213 | 3.73 | 500 | 0.5208 | 0.8332 |
0.3106 | 3.81 | 510 | 0.4609 | 0.8322 |
0.2226 | 3.88 | 520 | 0.5320 | 0.8274 |
0.3108 | 3.96 | 530 | 0.5457 | 0.8255 |
0.2456 | 4.03 | 540 | 0.4865 | 0.8322 |
0.223 | 4.1 | 550 | 0.5540 | 0.8313 |
0.1884 | 4.18 | 560 | 0.5363 | 0.8341 |
0.1934 | 4.25 | 570 | 0.5706 | 0.8332 |
0.1793 | 4.33 | 580 | 0.5814 | 0.8322 |
0.2952 | 4.4 | 590 | 0.5305 | 0.8360 |
0.2915 | 4.48 | 600 | 0.5104 | 0.8332 |
0.259 | 4.55 | 610 | 0.5076 | 0.8428 |
0.2453 | 4.63 | 620 | 0.5188 | 0.8351 |
0.1903 | 4.7 | 630 | 0.5396 | 0.8399 |
0.2573 | 4.78 | 640 | 0.5584 | 0.8332 |
0.2787 | 4.85 | 650 | 0.5340 | 0.8360 |
0.2256 | 4.93 | 660 | 0.5175 | 0.8351 |
0.257 | 5.0 | 670 | 0.4482 | 0.8370 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3