CoolFace
Modelpublic

Luciano/bertimbau-base-finetuned-lener-br-finetuned-brazilian_court_decisions

sourceHugging Facemitupdated 3y agoView on Hugging Face
1likes28downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

bertimbau-base-finetuned-lener-br-finetuned-braziliancourtdecisions

This model is a fine-tuned version of Luciano/bertimbau-base-finetuned-lener-br on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 1.8017
  • —Accuracy: 0.7698

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
  • —trainbatchsize: 8
  • —evalbatchsize: 8
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 15

Training results

Training LossEpochStepValidation LossAccuracy
No log1.04050.77900.6535
0.82762.08100.67390.7277
0.58183.012150.87670.7302
0.41474.016200.82290.7896
0.2875.020250.98740.7921
0.2876.024301.23010.7772
0.17277.028351.28640.7946
0.11798.032401.50970.7772
0.07099.036451.47720.7921
0.043710.040501.55810.7797
0.043711.044551.63170.7896
0.031812.048601.72950.7822
0.015813.052651.73330.7797
0.010814.056701.80080.7772
0.013715.060751.80170.7698

Framework versions

  • —Transformers 4.22.0
  • —Pytorch 1.12.1+cu113
  • —Datasets 2.4.0
  • —Tokenizers 0.12.1