filippo571's picture
Upload TFDistilBertForSequenceClassification
1a6e1e9
|
raw
history blame
2.04 kB
metadata
license: apache-2.0
tags:
  - generated_from_keras_callback
model-index:
  - name: pull_request_comments_model
    results: []

pull_request_comments_model

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

  • Train Loss: 0.4068
  • Validation Loss: 0.5871
  • Train Accuracy: 0.8205
  • Epoch: 9

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', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 140, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
  • training_precision: float32

Training results

Train Loss Validation Loss Train Accuracy Epoch
1.2952 1.1288 0.4957 0
1.1276 0.9616 0.6581 1
0.9647 0.8319 0.6838 2
0.8187 0.7676 0.7265 3
0.7088 0.7020 0.7436 4
0.6026 0.6590 0.7863 5
0.5296 0.6246 0.7949 6
0.4712 0.6000 0.8120 7
0.4359 0.5900 0.8291 8
0.4068 0.5871 0.8205 9

Framework versions

  • Transformers 4.26.1
  • TensorFlow 2.9.0
  • Datasets 2.10.1
  • Tokenizers 0.13.2