JovialValley/model_syllable_onSet0

<!-- 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. -->

modelsyllableonSet0

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:

• —Loss: 0.1789
• —0 Precision: 1.0
• —0 Recall: 0.9688
• —0 F1-score: 0.9841
• —0 Support: 32
• —1 Precision: 0.9667
• —1 Recall: 1.0
• —1 F1-score: 0.9831
• —1 Support: 29
• —2 Precision: 1.0
• —2 Recall: 1.0
• —2 F1-score: 1.0
• —2 Support: 29
• —3 Precision: 1.0
• —3 Recall: 1.0
• —3 F1-score: 1.0
• —3 Support: 8
• —Accuracy: 0.9898
• —Macro avg Precision: 0.9917
• —Macro avg Recall: 0.9922
• —Macro avg F1-score: 0.9918
• —Macro avg Support: 98
• —Weighted avg Precision: 0.9901
• —Weighted avg Recall: 0.9898
• —Weighted avg F1-score: 0.9898
• —Weighted avg Support: 98
• —Wer: 0.4059
• —Mtrix: [[0, 1, 2, 3], [0, 31, 1, 0, 0], [1, 0, 29, 0, 0], [2, 0, 0, 29, 0], [3, 0, 0, 0, 8]]

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: 0.0003
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —gradientaccumulationsteps: 2
• —totaltrainbatch_size: 16
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —lrschedulerwarmup_steps: 200
• —num_epochs: 70
• —mixedprecisiontraining: Native AMP

Training results

Framework versions

• —Transformers 4.25.1
• —Pytorch 1.13.0+cu116
• —Datasets 2.8.0
• —Tokenizers 0.13.2