|
--- |
|
license: apache-2.0 |
|
base_model: google/mt5-large |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
model-index: |
|
- name: cs_mT5-large2_0.01_50_v0.1 |
|
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. --> |
|
|
|
# cs_mT5-large2_0.01_50_v0.1 |
|
|
|
This model is a fine-tuned version of [google/mt5-large](https://huggingface.co/google/mt5-large) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 7.9807 |
|
- Bleu: 0.6899 |
|
- Gen Len: 19.0 |
|
|
|
## 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: 0.01 |
|
- 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: 50 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:| |
|
| 2.6703 | 1.0 | 6 | 8.3163 | 0.0 | 19.0 | |
|
| 3.3427 | 2.0 | 12 | 6.5085 | 0.5004 | 19.0 | |
|
| 2.8652 | 3.0 | 18 | 7.0200 | 0.6899 | 19.0 | |
|
| 2.9454 | 4.0 | 24 | 7.3333 | 0.2191 | 19.0 | |
|
| 2.7918 | 5.0 | 30 | 7.5745 | 0.4671 | 12.0 | |
|
| 3.5645 | 6.0 | 36 | 6.3676 | 0.0 | 19.0 | |
|
| 3.0885 | 7.0 | 42 | 7.0359 | 0.6908 | 19.0 | |
|
| 3.5374 | 8.0 | 48 | 6.8709 | 0.1154 | 12.3333 | |
|
| 3.3746 | 9.0 | 54 | 6.4090 | 0.0 | 19.0 | |
|
| 2.5927 | 10.0 | 60 | 6.7357 | 0.6381 | 19.0 | |
|
| 2.581 | 11.0 | 66 | 6.5953 | 0.2635 | 19.0 | |
|
| 4.1786 | 12.0 | 72 | 7.8617 | 0.2068 | 19.0 | |
|
| 2.8545 | 13.0 | 78 | 6.2553 | 0.1628 | 19.0 | |
|
| 2.8925 | 14.0 | 84 | 6.6297 | 0.612 | 19.0 | |
|
| 3.2424 | 15.0 | 90 | 7.0312 | 0.7377 | 19.0 | |
|
| 2.379 | 16.0 | 96 | 6.9121 | 0.6562 | 19.0 | |
|
| 2.2356 | 17.0 | 102 | 6.9446 | 0.2759 | 15.0 | |
|
| 3.0548 | 18.0 | 108 | 7.5770 | 0.1529 | 19.0 | |
|
| 2.4637 | 19.0 | 114 | 7.1444 | 0.4497 | 19.0 | |
|
| 2.96 | 20.0 | 120 | 6.8181 | 0.3779 | 11.0 | |
|
| 2.2016 | 21.0 | 126 | 6.6893 | 0.6562 | 19.0 | |
|
| 1.9774 | 22.0 | 132 | 7.3802 | 0.3807 | 19.0 | |
|
| 1.6734 | 23.0 | 138 | 6.7319 | 0.5405 | 19.0 | |
|
| 3.1958 | 24.0 | 144 | 7.1645 | 1.2379 | 19.0 | |
|
| 3.1363 | 25.0 | 150 | 7.7097 | 0.3794 | 19.0 | |
|
| 2.4353 | 26.0 | 156 | 6.9324 | 0.2522 | 14.0 | |
|
| 2.8675 | 27.0 | 162 | 6.7989 | 0.1488 | 19.0 | |
|
| 1.7486 | 28.0 | 168 | 7.1052 | 0.7123 | 19.0 | |
|
| 2.775 | 29.0 | 174 | 7.0195 | 0.7393 | 19.0 | |
|
| 1.8752 | 30.0 | 180 | 6.9133 | 0.2119 | 19.0 | |
|
| 1.7576 | 31.0 | 186 | 7.2143 | 0.2641 | 19.0 | |
|
| 2.2793 | 32.0 | 192 | 7.0029 | 1.1166 | 19.0 | |
|
| 1.98 | 33.0 | 198 | 6.9954 | 0.5348 | 19.0 | |
|
| 1.4242 | 34.0 | 204 | 7.5163 | 0.2088 | 19.0 | |
|
| 2.413 | 35.0 | 210 | 7.0622 | 0.1433 | 19.0 | |
|
| 1.2191 | 36.0 | 216 | 7.0088 | 0.5307 | 12.0 | |
|
| 1.5944 | 37.0 | 222 | 7.7706 | 0.1948 | 19.0 | |
|
| 1.0044 | 38.0 | 228 | 7.7163 | 0.8485 | 18.4286 | |
|
| 1.4428 | 39.0 | 234 | 7.4919 | 0.6033 | 19.0 | |
|
| 3.0175 | 40.0 | 240 | 7.4158 | 0.5109 | 19.0 | |
|
| 1.3632 | 41.0 | 246 | 6.9819 | 0.4326 | 11.0 | |
|
| 1.8384 | 42.0 | 252 | 7.0156 | 0.6215 | 18.2381 | |
|
| 1.3237 | 43.0 | 258 | 7.2082 | 0.4826 | 19.0 | |
|
| 1.1516 | 44.0 | 264 | 7.5088 | 0.4745 | 18.8571 | |
|
| 1.3893 | 45.0 | 270 | 7.7298 | 0.4527 | 19.0 | |
|
| 1.0125 | 46.0 | 276 | 7.8458 | 0.832 | 14.0 | |
|
| 0.8954 | 47.0 | 282 | 7.9101 | 0.7754 | 17.5714 | |
|
| 1.8111 | 48.0 | 288 | 7.9713 | 0.6899 | 19.0 | |
|
| 1.2008 | 49.0 | 294 | 7.9821 | 0.6899 | 19.0 | |
|
| 1.5131 | 50.0 | 300 | 7.9807 | 0.6899 | 19.0 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 1.13.1+cu117 |
|
- Datasets 2.17.0 |
|
- Tokenizers 0.15.2 |
|
|
