|
--- |
|
library_name: transformers |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: gpt22gpt2-gpt2-medium-cnn-dailymail-seed42 |
|
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. --> |
|
|
|
# gpt22gpt2-gpt2-medium-cnn-dailymail-seed42 |
|
|
|
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: 1.7212 |
|
- Rouge1: 0.2011 |
|
- Rouge2: 0.0413 |
|
- Rougel: 0.1250 |
|
- Rougelsum: 0.1887 |
|
|
|
## 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: 16 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 1000 |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
|
|:-------------:|:------:|:-----:|:---------------:|:------:|:------:|:------:|:---------:| |
|
| 2.8287 | 0.2229 | 2000 | 2.6199 | 0.1926 | 0.0340 | 0.1248 | 0.1800 | |
|
| 2.4721 | 0.4458 | 4000 | 2.2632 | 0.1971 | 0.0382 | 0.1284 | 0.1816 | |
|
| 2.2324 | 0.6687 | 6000 | 2.0400 | 0.2164 | 0.0487 | 0.1345 | 0.2025 | |
|
| 2.1025 | 0.8916 | 8000 | 1.9285 | 0.1827 | 0.0330 | 0.1175 | 0.1691 | |
|
| 1.8926 | 1.1145 | 10000 | 1.8595 | 0.1712 | 0.0288 | 0.1124 | 0.1598 | |
|
| 1.8605 | 1.3374 | 12000 | 1.8191 | 0.1919 | 0.0371 | 0.1230 | 0.1790 | |
|
| 1.8327 | 1.5603 | 14000 | 1.7931 | 0.1838 | 0.0346 | 0.1189 | 0.1724 | |
|
| 1.8209 | 1.7832 | 16000 | 1.7606 | 0.1847 | 0.0342 | 0.1169 | 0.1733 | |
|
| 1.7879 | 2.0061 | 18000 | 1.7484 | 0.1949 | 0.0390 | 0.1227 | 0.1830 | |
|
| 1.6658 | 2.2290 | 20000 | 1.7460 | 0.1910 | 0.0358 | 0.1201 | 0.1798 | |
|
| 1.6547 | 2.4519 | 22000 | 1.7386 | 0.1906 | 0.0360 | 0.1196 | 0.1791 | |
|
| 1.6426 | 2.6748 | 24000 | 1.7256 | 0.1953 | 0.0387 | 0.1226 | 0.1833 | |
|
| 1.6495 | 2.8977 | 26000 | 1.7212 | 0.2011 | 0.0413 | 0.1250 | 0.1887 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.0 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|
