README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - hebrew_news
metrics:
  - rouge
model-index:
  - name: mt5-small-news-final
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: hebrew_news
          type: hebrew_news
          config: hebrew_news
          split: train
          args: hebrew_news
        metrics:
          - name: Rouge1
            type: rouge
            value: 0.4946

mt5-small-news-final

This model is a fine-tuned version of google/mt5-small on the hebrew_news dataset. It achieves the following results on the evaluation set:

• Loss: nan
• Rouge1: 0.4946
• Rouge2: 0.0432
• Rougel: 0.4922
• Rougelsum: 0.4919
• Gen Len: 6.0117

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.0	1.0	1029	nan	0.4946	0.0432	0.4922	0.4919	6.0117

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.11.0