minddigger008's picture
Upload 11 files
d3c1e6f verified
|
raw
history blame
1.3 kB
metadata
license: mit
base_model: minhtoan/t5-finetune-cnndaily-news
tags:
  - generated_from_trainer
model-index:
  - name: t5-finetune-cnndaily-news-sumsum
    results: []

t5-finetune-cnndaily-news-sumsum

This model is a fine-tuned version of minhtoan/t5-finetune-cnndaily-news on the sumsum dataset.

Model description

A light weight model for light dialogue summarization tasks.

Intended uses & limitations

Dialogue summarization. Not intended for large conversation.

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • gradient_accumulation_steps: 16
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 1

Training results

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2