metadata

license: mit
tags:
  - generated_from_keras_callback
model-index:
  - name: pmfsl/bertimbau-base-finetuned-rte
    results:
      - task:
          type: text-classification
          name: Natural Lenguage Inference
        dataset:
          type: assin2
          name: ASSIN2
        metrics:
          - type: accuracy
            value: 0.877859477124183
          - type: f1
            value: 0.8860083873427372
datasets:
  - assin2
metrics:
  - accuracy
  - f1
language:
  - pt

pmfsl/bertimbau-base-finetuned-rte

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

Train Loss: 0.0326
Validation Loss: 0.1834
Test Loss: 0.5695
Train Accuracy: 0.9531
Train F1: 0.9534
Test Accuracy: 0.8778
Test F1: 0.8860
Epoch: 4

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:

optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-05, 'decay_steps': 505, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
training_precision: float32

Training results

Train Loss	Validation Loss	Train Accuracy	Train F1	Epoch
0.3846	0.2204	0.9152	0.9191	0
0.1981	0.1577	0.9442	0.9455	1
0.1026	0.1348	0.9509	0.9511	2
0.0593	0.1492	0.9531	0.9542	3
0.0326	0.1834	0.9531	0.9534	4

Framework versions

Transformers 4.27.4
TensorFlow 2.12.0
Datasets 2.11.0
Tokenizers 0.13.2