|
--- |
|
tags: |
|
- xsum_v1_last |
|
- generated_from_trainer |
|
datasets: |
|
- xsum |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: pegasus-large-finetune-xsum |
|
results: |
|
- task: |
|
name: Sequence-to-sequence Language Modeling |
|
type: text2text-generation |
|
dataset: |
|
name: xsum |
|
type: xsum |
|
args: default |
|
metrics: |
|
- name: Rouge1 |
|
type: rouge |
|
value: 5.0462 |
|
--- |
|
|
|
<!-- 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-large-finetune-xsum |
|
|
|
This model is a fine-tuned version of [google/pegasus-large](https://huggingface.co/google/pegasus-large) on the xsum dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 10.0826 |
|
- Rouge1: 5.0462 |
|
- Rouge2: 0.6914 |
|
- Rougel: 3.5071 |
|
- Rougelsum: 3.9548 |
|
|
|
## 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: 5.6e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:| |
|
| 11.4044 | 1.0 | 13 | 10.7501 | 5.5154 | 0.5561 | 3.8425 | 4.2435 | |
|
| 10.5741 | 2.0 | 26 | 10.2309 | 5.4282 | 0.7228 | 3.5759 | 4.0538 | |
|
| 10.0146 | 3.0 | 39 | 10.0826 | 5.0462 | 0.6914 | 3.5071 | 3.9548 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0 |
|
- Pytorch 1.13.0 |
|
- Datasets 2.6.1 |
|
- Tokenizers 0.11.0 |
|
|