whisper_nosp_0010 / README.md
bigmorning's picture
Upload TFWhisperForConditionalGeneration
a6c5c94
metadata
license: apache-2.0
tags:
  - generated_from_keras_callback
model-index:
  - name: whisper_nosp_0010
    results: []

whisper_nosp_0010

This model is a fine-tuned version of openai/whisper-tiny on an unknown dataset. It achieves the following results on the evaluation set:

  • Train Loss: 0.7431
  • Train Accuracy: 0.0199
  • Validation Loss: 0.9603
  • Validation Accuracy: 0.0196
  • 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': 'AdamWeightDecay', 'learning_rate': 1e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
  • training_precision: float32

Training results

Train Loss Train Accuracy Validation Loss Validation Accuracy Epoch
7.5559 0.0010 6.3853 0.0013 0
6.3227 0.0021 5.7023 0.0038 1
4.9825 0.0063 3.6302 0.0109 2
2.9413 0.0126 2.1959 0.0154 3
1.9349 0.0157 1.6630 0.0172 4
1.4741 0.0171 1.3813 0.0181 5
1.1975 0.0181 1.2161 0.0186 6
1.0048 0.0188 1.0990 0.0191 7
0.8598 0.0194 1.0165 0.0194 8
0.7431 0.0199 0.9603 0.0196 9

Framework versions

  • Transformers 4.25.0.dev0
  • TensorFlow 2.9.2
  • Datasets 2.6.1
  • Tokenizers 0.13.2