|
--- |
|
base_model: /exports/eddie/scratch/s1970716/models/summarization/longt5_xl_sfd_bp_20/checkpoint-280 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- learn3r/summ_screen_fd_bp |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: longt5_xl_sfd_bp_40 |
|
results: |
|
- task: |
|
name: Summarization |
|
type: summarization |
|
dataset: |
|
name: learn3r/summ_screen_fd_bp |
|
type: learn3r/summ_screen_fd_bp |
|
metrics: |
|
- name: Rouge1 |
|
type: rouge |
|
value: 40.6965 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# longt5_xl_sfd_bp_40 |
|
|
|
This model is a fine-tuned version of [/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_sfd_bp_20/checkpoint-280](https://huggingface.co//exports/eddie/scratch/s1970716/models/summarization/longt5_xl_sfd_bp_20/checkpoint-280) on the learn3r/summ_screen_fd_bp dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.8277 |
|
- Rouge1: 40.6965 |
|
- Rouge2: 17.2793 |
|
- Rougel: 27.8429 |
|
- Rougelsum: 39.0726 |
|
- Gen Len: 294.0890 |
|
|
|
## 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: 0.001 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 32 |
|
- total_train_batch_size: 256 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: constant |
|
- num_epochs: 20.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:--------:| |
|
| 0.1033 | 0.97 | 14 | 3.1096 | 40.3355 | 16.0557 | 27.4642 | 38.6436 | 279.7507 | |
|
| 0.0836 | 1.95 | 28 | 3.0361 | 38.2411 | 16.5448 | 26.6409 | 36.5841 | 368.4659 | |
|
| 0.0717 | 2.99 | 43 | 2.9389 | 32.0114 | 13.7953 | 22.278 | 30.726 | 489.2047 | |
|
| 0.0614 | 3.97 | 57 | 3.0221 | 32.969 | 13.7053 | 22.7428 | 31.6951 | 477.7240 | |
|
| 0.1275 | 4.94 | 71 | 2.8277 | 40.6965 | 17.2793 | 27.8429 | 39.0726 | 294.0890 | |
|
| 0.0511 | 5.98 | 86 | 3.0433 | 33.6479 | 15.0729 | 23.5443 | 32.3304 | 476.8457 | |
|
| 0.0666 | 6.96 | 100 | 3.1150 | 37.743 | 16.2368 | 26.2524 | 36.1313 | 390.4362 | |
|
| 0.0398 | 8.0 | 115 | 3.2225 | 41.3177 | 16.6663 | 28.7806 | 39.5914 | 203.4006 | |
|
| 0.0396 | 8.97 | 129 | 3.1462 | 39.9605 | 16.6732 | 28.3459 | 38.226 | 123.8309 | |
|
| 0.0466 | 9.95 | 143 | 3.2545 | 40.7977 | 16.9616 | 27.427 | 38.8973 | 298.5579 | |
|
| 0.043 | 10.99 | 158 | 3.3188 | 36.6349 | 16.1781 | 25.1327 | 35.1793 | 425.1395 | |
|
| 0.0538 | 11.97 | 172 | 2.8277 | 36.7878 | 15.1186 | 24.9774 | 35.275 | 394.8605 | |
|
| 0.028 | 12.94 | 186 | 3.4398 | 42.9644 | 18.1812 | 29.1539 | 41.0465 | 188.1780 | |
|
| 0.1056 | 13.98 | 201 | 3.3348 | 41.1626 | 17.1605 | 27.6558 | 39.2548 | 261.2967 | |
|
| 0.0303 | 14.96 | 215 | 3.0238 | 42.2372 | 17.7292 | 28.8099 | 40.3325 | 231.6083 | |
|
| 0.0234 | 16.0 | 230 | 3.3485 | 41.714 | 17.7161 | 27.9345 | 39.8519 | 306.1602 | |
|
| 0.0263 | 16.97 | 244 | 3.2419 | 42.0014 | 17.2719 | 28.499 | 40.2024 | 210.7122 | |
|
| 0.0225 | 17.95 | 258 | 3.3453 | 41.7766 | 17.7154 | 28.4692 | 39.9749 | 248.5786 | |
|
| 0.0225 | 18.99 | 273 | 3.4441 | 42.1727 | 17.598 | 28.5122 | 40.4005 | 248.6380 | |
|
| 0.0211 | 19.48 | 280 | 3.3211 | 42.5239 | 17.4102 | 28.6868 | 40.6537 | 200.3798 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.14.1 |
|
|
