---
license: apache-2.0
base_model: facebook/bart-large
tags:
- generated_from_trainer
metrics:
- rouge
- wer
model-index:
- name: BART_1st_STAGE_SUMMARIZER_v3
  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. -->

# BART_1st_STAGE_SUMMARIZER_v3

This model is a fine-tuned version of [facebook/bart-large](https://huggingface.co/facebook/bart-large) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0858
- Rouge1: 0.7512
- Rouge2: 0.5341
- Rougel: 0.6975
- Rougelsum: 0.702
- Wer: 0.3693

## 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: 2e-05
- train_batch_size: 6
- eval_batch_size: 6
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:------:|
| No log        | 0.21  | 250  | 1.3647          | 0.7195 | 0.4836 | 0.6589 | 0.6639    | 0.4174 |
| 1.954         | 0.42  | 500  | 1.2505          | 0.7301 | 0.4985 | 0.6698 | 0.6744    | 0.406  |
| 1.954         | 0.63  | 750  | 1.2094          | 0.7341 | 0.5051 | 0.6764 | 0.6799    | 0.3973 |
| 1.3635        | 0.84  | 1000 | 1.1771          | 0.7387 | 0.5144 | 0.6837 | 0.688     | 0.3902 |
| 1.3635        | 1.05  | 1250 | 1.1862          | 0.7422 | 0.5177 | 0.6853 | 0.6907    | 0.3892 |
| 1.2246        | 1.26  | 1500 | 1.1514          | 0.7416 | 0.5182 | 0.686  | 0.6905    | 0.3869 |
| 1.2246        | 1.48  | 1750 | 1.1415          | 0.7448 | 0.5219 | 0.6884 | 0.693     | 0.3837 |
| 1.1592        | 1.69  | 2000 | 1.1235          | 0.7459 | 0.5252 | 0.691  | 0.6953    | 0.38   |
| 1.1592        | 1.9   | 2250 | 1.1210          | 0.7481 | 0.5284 | 0.6921 | 0.697     | 0.3778 |
| 1.1084        | 2.11  | 2500 | 1.1074          | 0.7487 | 0.5295 | 0.6939 | 0.6987    | 0.376  |
| 1.1084        | 2.32  | 2750 | 1.0985          | 0.7505 | 0.5317 | 0.6961 | 0.7009    | 0.3738 |
| 1.0452        | 2.53  | 3000 | 1.0907          | 0.7507 | 0.5324 | 0.6962 | 0.7006    | 0.3718 |
| 1.0452        | 2.74  | 3250 | 1.0884          | 0.7512 | 0.5339 | 0.6973 | 0.7023    | 0.3702 |
| 1.0253        | 2.95  | 3500 | 1.0858          | 0.7512 | 0.5341 | 0.6975 | 0.702     | 0.3693 |


### Framework versions

- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2