GiordanoB's picture
update model card README.md
ae682a4
|
raw
history blame
1.73 kB
metadata
license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: mt5-base-finetuned-summarization-V2
    results: []

mt5-base-finetuned-summarization-V2

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

  • Loss: 8.3409
  • Rouge1: 6.1259
  • Rouge2: 1.4637
  • Rougel: 5.3192
  • Rougelsum: 5.7739
  • Gen Len: 9.9286

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
No log 1.0 15 10.0266 6.7528 2.8064 5.9938 6.4352 10.0
No log 2.0 30 8.4159 6.1259 1.4637 5.3192 5.7739 10.0714
No log 3.0 45 8.3409 6.1259 1.4637 5.3192 5.7739 9.9286

Framework versions

  • Transformers 4.18.0
  • Pytorch 1.11.0
  • Datasets 2.1.0
  • Tokenizers 0.12.1