README.md

metadata

tags:
  - generated_from_trainer
datasets:
  - indosum
metrics:
  - rouge
model-index:
  - name: t5-base-indonesian-summarization-cased-finetuned-indosum
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: indosum
          type: indosum
          config: indosum_fold0_source
          split: validation
          args: indosum_fold0_source
        metrics:
          - name: Rouge1
            type: rouge
            value: 0.3278

t5-base-indonesian-summarization-cased-finetuned-indosum

This model is a fine-tuned version of panggi/t5-base-indonesian-summarization-cased on the indosum dataset. It achieves the following results on the evaluation set:

• Loss: 0.5884
• Rouge1: 0.3278
• Rouge2: 0.2868
• Rougel: 0.32
• Rougelsum: 0.3198
• Gen Len: 19.0

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: 3
• eval_batch_size: 3
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.3815	1.0	4754	0.5584	0.3281	0.2866	0.3202	0.32	19.0
0.3207	2.0	9508	0.5884	0.3278	0.2868	0.32	0.3198	19.0

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3