Hugging Face's logo

apwic
/
indosum-base-0

Transformers
Safetensors
Indonesian
t5
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Community
indosum-base-0 / README.md
apwic's picture
apwic
End of training
898c512 verified
preview code
|
raw
history blame
2.02 kB
metadata
language:
  - id
license: apache-2.0
base_model: LazarusNLP/IndoNanoT5-base
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: indosum-base-0
    results: []

indosum-base-0

This model is a fine-tuned version of LazarusNLP/IndoNanoT5-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7149
  • Rouge1: 72.3288
  • Rouge2: 65.2271
  • Rougel: 69.3889
  • Rougelsum: 71.4606
  • Gen Len: 99.9773

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: 0.001
  • train_batch_size: 4
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5.0

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
1.5249 1.0 3566 0.8858 65.9377 58.0172 63.1 65.0124 103.8293
0.8202 2.0 7132 0.7492 69.7595 62.2139 66.6576 68.9014 101.3173
0.6232 3.0 10698 0.6953 70.7507 63.4429 67.7582 69.8526 97.7093
0.4728 4.0 14264 0.6717 70.8339 63.5583 67.7546 69.8695 97.7893
0.3238 5.0 17830 0.7149 71.7161 64.386 68.5898 70.7721 99.5373

Framework versions

  • Transformers 4.40.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1