File size: 1,994 Bytes
d53a0cd |
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 |
---
license: mit
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: final_bart
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. -->
# final_bart
This model is a fine-tuned version of [gogamza/kobart-base-v2](https://huggingface.co/gogamza/kobart-base-v2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.6848
- Rouge1: 35.7722
- Rouge2: 12.5127
- Rougel: 23.3002
- Rdass: 0.6248
- Bleu1: 30.5261
- Bleu2: 17.6264
- Bleu3: 10.3974
- Bleu4: 5.4348
- Gen Len: 53.47
## 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: 3e-05
- train_batch_size: 64
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rdass | Bleu1 | Bleu2 | Bleu3 | Bleu4 | Gen Len |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:------:|:-------:|:-------:|:-------:|:------:|:-------:|
| 2.1542 | 1.5 | 1000 | 2.7491 | 33.5554 | 11.2371 | 22.006 | 0.6093 | 27.9938 | 15.5354 | 8.2494 | 4.42 | 50.08 |
| 2.0071 | 2.99 | 2000 | 2.6813 | 35.0501 | 12.2759 | 22.6669 | 0.6155 | 29.6866 | 17.1396 | 9.7016 | 5.3559 | 54.04 |
| 1.8694 | 4.49 | 3000 | 2.6848 | 35.7722 | 12.5127 | 23.3002 | 0.6248 | 30.5261 | 17.6264 | 10.3974 | 5.4348 | 53.47 |
### Framework versions
- Transformers 4.25.1
- Pytorch 1.13.1+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2
|