|
|
--- |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- rouge |
|
|
model-index: |
|
|
- name: Image_Captioner_Mimic |
|
|
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. --> |
|
|
|
|
|
# Image_Captioner_Mimic |
|
|
|
|
|
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.0963 |
|
|
- Rouge1: 32.528 |
|
|
- Rouge2: 19.9922 |
|
|
- Rougel: 31.403 |
|
|
- Rougelsum: 31.9372 |
|
|
- Gen Len: 12.5584 |
|
|
|
|
|
## 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: linear |
|
|
- num_epochs: 30 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |
|
|
|:-------------:|:-----:|:------:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:| |
|
|
| 0.0597 | 1.0 | 24457 | 0.0567 | 37.8657 | 27.8087 | 37.4596 | 37.752 | 9.9527 | |
|
|
| 0.0533 | 2.0 | 48914 | 0.0526 | 39.2211 | 28.2036 | 38.5786 | 38.9976 | 10.7079 | |
|
|
| 0.0507 | 3.0 | 73371 | 0.0499 | 39.3449 | 28.3875 | 38.7151 | 39.0449 | 10.2091 | |
|
|
| 0.0457 | 4.0 | 97828 | 0.0479 | 39.8753 | 28.5 | 39.127 | 39.6178 | 11.2407 | |
|
|
| 0.0419 | 5.0 | 122285 | 0.0461 | 40.0478 | 28.797 | 39.3201 | 39.7468 | 10.3153 | |
|
|
| 0.0406 | 6.0 | 146742 | 0.0445 | 39.7923 | 28.4281 | 39.0583 | 39.4523 | 10.4186 | |
|
|
| 0.0373 | 7.0 | 171199 | 0.0429 | 39.954 | 28.535 | 39.2226 | 39.6457 | 10.6640 | |
|
|
| 0.0347 | 8.0 | 195656 | 0.0419 | 39.4329 | 28.0336 | 38.6815 | 39.0968 | 10.7775 | |
|
|
| 0.031 | 9.0 | 220113 | 0.0411 | 39.4524 | 28.1057 | 38.6998 | 39.0906 | 10.8397 | |
|
|
| 0.0286 | 10.0 | 244570 | 0.0407 | 39.1493 | 27.639 | 38.3784 | 38.8085 | 10.9530 | |
|
|
| 0.0261 | 11.0 | 269027 | 0.0408 | 38.8083 | 27.2206 | 37.9679 | 38.422 | 11.2390 | |
|
|
| 0.0249 | 12.0 | 293484 | 0.0412 | 38.3972 | 26.7316 | 37.5838 | 38.0409 | 11.4510 | |
|
|
| 0.0214 | 13.0 | 317941 | 0.0424 | 37.785 | 26.3302 | 36.9553 | 37.3764 | 11.4482 | |
|
|
| 0.0188 | 14.0 | 342398 | 0.0438 | 36.9552 | 25.3108 | 36.0278 | 36.4965 | 11.6232 | |
|
|
| 0.0174 | 15.0 | 366855 | 0.0458 | 35.6476 | 23.9574 | 34.6526 | 35.1259 | 11.6605 | |
|
|
| 0.0153 | 16.0 | 391312 | 0.0487 | 34.657 | 22.8337 | 33.5891 | 34.1343 | 12.2395 | |
|
|
| 0.013 | 17.0 | 415769 | 0.0518 | 33.5548 | 21.1569 | 32.4899 | 33.0394 | 12.2604 | |
|
|
| 0.0114 | 18.0 | 440226 | 0.0559 | 34.3809 | 22.0108 | 33.2698 | 33.8578 | 12.0861 | |
|
|
| 0.01 | 19.0 | 464683 | 0.0601 | 32.9062 | 20.3145 | 31.8147 | 32.3802 | 12.5176 | |
|
|
| 0.0081 | 20.0 | 489140 | 0.0651 | 32.9482 | 20.3862 | 31.865 | 32.3837 | 12.4577 | |
|
|
| 0.0069 | 21.0 | 513597 | 0.0698 | 32.3054 | 19.764 | 31.2178 | 31.7592 | 12.4939 | |
|
|
| 0.0057 | 22.0 | 538054 | 0.0751 | 31.7627 | 19.0106 | 30.6263 | 31.175 | 12.7530 | |
|
|
| 0.0048 | 23.0 | 562511 | 0.0793 | 31.8295 | 19.255 | 30.6958 | 31.2314 | 12.6077 | |
|
|
| 0.0041 | 24.0 | 586968 | 0.0834 | 32.1523 | 19.2017 | 30.9774 | 31.5383 | 12.7461 | |
|
|
| 0.0032 | 25.0 | 611425 | 0.0870 | 32.5379 | 20.0041 | 31.3903 | 31.9037 | 12.6848 | |
|
|
| 0.0025 | 26.0 | 635882 | 0.0903 | 32.6757 | 20.1388 | 31.5495 | 32.0827 | 12.5950 | |
|
|
| 0.0023 | 27.0 | 660339 | 0.0927 | 32.0874 | 19.3546 | 30.9125 | 31.4675 | 12.6290 | |
|
|
| 0.0019 | 28.0 | 684796 | 0.0947 | 32.6988 | 20.1847 | 31.5643 | 32.1143 | 12.5412 | |
|
|
| 0.0017 | 29.0 | 709253 | 0.0958 | 32.4574 | 19.7702 | 31.2955 | 31.8608 | 12.5558 | |
|
|
| 0.0014 | 30.0 | 733710 | 0.0963 | 32.528 | 19.9922 | 31.403 | 31.9372 | 12.5584 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.37.1 |
|
|
- Pytorch 1.13.1+cu117 |
|
|
- Datasets 2.15.0 |
|
|
- Tokenizers 0.15.1 |
|
|
|
