---
tags:
- generated_from_trainer
datasets:
- cnn_dailymail
metrics:
- rouge
model-index:
- name: pegasus-newsroom-cnn-adam8bit-bs4x64acc_3
  results:
  - task:
      name: Sequence-to-sequence Language Modeling
      type: text2text-generation
    dataset:
      name: cnn_dailymail
      type: cnn_dailymail
      args: 3.0.0
    metrics:
    - name: Rouge1
      type: rouge
      value: 44.268
---

<!-- 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-newsroom-cnn-adam8bit-bs4x64acc_3

This model is a fine-tuned version of [oMateos2020/pegasus-newsroom-cnn-adam8bit-bs4x64acc_2](https://huggingface.co/oMateos2020/pegasus-newsroom-cnn-adam8bit-bs4x64acc_2) on the cnn_dailymail dataset.
It achieves the following results on the evaluation set:
- Loss: 2.8566
- Rouge1: 44.268
- Rouge2: 21.5816
- Rougel: 31.3753
- Rougelsum: 41.2339
- Gen Len: 71.3505

## 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: 6.4e-05
- train_batch_size: 4
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.4
- num_epochs: 1
- mixed_precision_training: Native AMP
- label_smoothing_factor: 0.1

### Training results

| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2  | Rougel  | Rougelsum | Gen Len |
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:-------:|:---------:|:-------:|
| 2.907         | 1.0   | 1121 | 2.8566          | 44.268 | 21.5816 | 31.3753 | 41.2339   | 71.3505 |


### Framework versions

- Transformers 4.20.1
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1