File size: 2,082 Bytes
7ac9e23 9df1e56 7ac9e23 |
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 |
---
license: other
base_model: aubmindlab/aragpt2-large
tags:
- generated_from_trainer
metrics:
- bleu
- rouge
model-index:
- name: res_nw_dj_aragpt2-large
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. -->
# res_nw_dj_aragpt2-large
This model is a fine-tuned version of [aubmindlab/aragpt2-large](https://huggingface.co/aubmindlab/aragpt2-large) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0718
- Bleu: 0.1063
- Rouge1: 0.4552
- Rouge2: 0.2232
- Rougel: 0.4516
## 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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 20.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Rouge1 | Rouge2 | Rougel |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:------:|
| 0.1323 | 1.0 | 5358 | 0.0757 | 0.0830 | 0.4072 | 0.1825 | 0.4035 |
| 0.0662 | 2.0 | 10716 | 0.0718 | 0.1063 | 0.4552 | 0.2232 | 0.4516 |
| 0.0526 | 3.0 | 16074 | 0.0727 | 0.1197 | 0.4753 | 0.2520 | 0.4719 |
| 0.0414 | 4.0 | 21432 | 0.0757 | 0.1274 | 0.4894 | 0.2644 | 0.4862 |
| 0.0325 | 5.0 | 26790 | 0.0819 | 0.1290 | 0.4910 | 0.2671 | 0.4875 |
| 0.0262 | 6.0 | 32148 | 0.0863 | 0.1297 | 0.4922 | 0.2665 | 0.4888 |
| 0.0221 | 7.0 | 37506 | 0.0930 | 0.1326 | 0.4960 | 0.2713 | 0.4923 |
### Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1
|