File size: 4,112 Bytes
35b1733 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 |
---
library_name: transformers
license: mit
base_model: VietAI/vit5-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: vit5-base_nli
results: []
---
<!-- 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. -->
# vit5-base_nli
This model is a fine-tuned version of [VietAI/vit5-base](https://huggingface.co/VietAI/vit5-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.9361
- Accuracy: 0.7508
- Precision Macro: 0.7512
- Recall Macro: 0.7507
- F1 Macro: 0.7508
- F1 Weighted: 0.7508
## 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: 128
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 256
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision Macro | Recall Macro | F1 Macro | F1 Weighted |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------------:|:------------:|:--------:|:-----------:|
| 1.2042 | 1.0 | 72 | 1.0168 | 0.4772 | 0.5084 | 0.4784 | 0.4539 | 0.4536 |
| 1.0195 | 2.0 | 144 | 0.7645 | 0.6723 | 0.6767 | 0.6724 | 0.6711 | 0.6709 |
| 0.6388 | 3.0 | 216 | 0.6869 | 0.7126 | 0.7199 | 0.7121 | 0.7128 | 0.7127 |
| 0.5149 | 4.0 | 288 | 0.6967 | 0.7428 | 0.7453 | 0.7432 | 0.7427 | 0.7425 |
| 0.2882 | 5.0 | 360 | 0.7899 | 0.7375 | 0.7440 | 0.7375 | 0.7376 | 0.7374 |
| 0.2238 | 6.0 | 432 | 0.9740 | 0.7313 | 0.7398 | 0.7319 | 0.7300 | 0.7298 |
| 0.1326 | 7.0 | 504 | 1.0921 | 0.7344 | 0.7372 | 0.7350 | 0.7337 | 0.7335 |
| 0.096 | 8.0 | 576 | 1.2234 | 0.7366 | 0.7420 | 0.7361 | 0.7366 | 0.7366 |
| 0.0755 | 9.0 | 648 | 1.3014 | 0.7326 | 0.7355 | 0.7324 | 0.7332 | 0.7330 |
| 0.0505 | 10.0 | 720 | 1.3717 | 0.7397 | 0.7414 | 0.7395 | 0.7400 | 0.7399 |
| 0.0419 | 11.0 | 792 | 1.4521 | 0.7392 | 0.7429 | 0.7389 | 0.7394 | 0.7393 |
| 0.0301 | 12.0 | 864 | 1.5602 | 0.7428 | 0.7433 | 0.7428 | 0.7430 | 0.7429 |
| 0.0213 | 13.0 | 936 | 1.7194 | 0.7450 | 0.7457 | 0.7448 | 0.7450 | 0.7450 |
| 0.0171 | 14.0 | 1008 | 1.7975 | 0.7450 | 0.7475 | 0.7448 | 0.7449 | 0.7449 |
| 0.018 | 15.0 | 1080 | 1.7963 | 0.7525 | 0.7528 | 0.7525 | 0.7526 | 0.7526 |
| 0.0084 | 16.0 | 1152 | 1.8312 | 0.7512 | 0.7517 | 0.7512 | 0.7513 | 0.7513 |
| 0.0083 | 17.0 | 1224 | 1.8834 | 0.7525 | 0.7531 | 0.7526 | 0.7526 | 0.7525 |
| 0.0089 | 18.0 | 1296 | 1.9212 | 0.7561 | 0.7568 | 0.7561 | 0.7562 | 0.7561 |
| 0.0064 | 19.0 | 1368 | 1.9379 | 0.7508 | 0.7512 | 0.7507 | 0.7508 | 0.7508 |
| 0.0082 | 20.0 | 1440 | 1.9361 | 0.7508 | 0.7512 | 0.7507 | 0.7508 | 0.7508 |
### Framework versions
- Transformers 4.55.0
- Pytorch 2.7.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4
|