|
--- |
|
tags: |
|
- paraphrasing |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: pegasus-xsum-finetuned-paws-parasci |
|
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. --> |
|
|
|
# pegasus-xsum-finetuned-paws-parasci |
|
|
|
This model is a fine-tuned version of [domenicrosati/pegasus-xsum-finetuned-paws](https://huggingface.co/domenicrosati/pegasus-xsum-finetuned-paws) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.2256 |
|
- Rouge1: 61.8854 |
|
- Rouge2: 43.1061 |
|
- Rougel: 57.421 |
|
- Rougelsum: 57.4417 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- training_steps: 4000 |
|
- mixed_precision_training: Native AMP |
|
- label_smoothing_factor: 0.1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:| |
|
| No log | 0.05 | 1000 | 3.8024 | 49.471 | 24.8024 | 43.4857 | 43.5552 | |
|
| No log | 0.09 | 2000 | 3.6533 | 49.1046 | 24.4038 | 43.0189 | 43.002 | |
|
| No log | 0.14 | 3000 | 3.5867 | 49.5026 | 24.748 | 43.3059 | 43.2923 | |
|
| No log | 0.19 | 4000 | 3.5613 | 49.4319 | 24.5444 | 43.2225 | 43.1965 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0 |
|
- Pytorch 1.11.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.12.1 |
|
|
