roberta-base / README.md
dtorber's picture
Model save
bd9774e verified
|
raw
history blame
1.66 kB
metadata
license: mit
base_model: FacebookAI/roberta-base
tags:
  - generated_from_trainer
model-index:
  - name: roberta-base
    results: []

roberta-base

This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5272
  • Icm: 0.2161
  • Icmnorm: 0.6098
  • Fmeasure: 0.7394

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
  • distributed_type: multi-GPU
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Icm Icmnorm Fmeasure
No log 1.0 193 0.5272 0.2161 0.6098 0.7394
No log 2.0 386 0.6158 0.0402 0.5204 0.6763
0.5577 3.0 579 0.6748 0.1302 0.5661 0.7128

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2