mbart50-qg-aas / README.md
tiagoblima's picture
End of training
92fa24f
metadata
license: mit
base_model: facebook/mbart-large-50
tags:
  - generated_from_trainer
datasets:
  - tiagoblima/qg_squad_v1_pt
model-index:
  - name: mbart50-qg-aas
    results: []

mbart50-qg-aas

This model is a fine-tuned version of facebook/mbart-large-50 on the tiagoblima/qg_squad_v1_pt dataset. It achieves the following results on the evaluation set:

  • Loss: 5.1971

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: 0.005
  • train_batch_size: 64
  • eval_batch_size: 4
  • 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
7.1258 1.0 808 8.0165
5.7857 2.0 1616 7.2193
4.6138 3.0 2424 6.4846
3.6997 4.0 3232 5.7332
3.0051 5.0 4040 5.1971

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.0.0
  • Datasets 2.15.0
  • Tokenizers 0.15.0