README.md

metadata

tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: disfluency-large-3
    results: []

disfluency-large-3

This model is a fine-tuned version of vinai/phobert-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0364
• Precision: 0.9849
• Recall: 0.9802
• F1: 0.9825
• Accuracy: 0.9936

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	140	0.0713	0.8955	0.9165	0.9059	0.9816
No log	2.0	280	0.0334	0.9706	0.9730	0.9718	0.9925
No log	3.0	420	0.0584	0.9656	0.9609	0.9633	0.9880
0.1335	4.0	560	0.0352	0.9742	0.9742	0.9742	0.9922
0.1335	5.0	700	0.0539	0.9651	0.9633	0.9642	0.9894
0.1335	6.0	840	0.0293	0.9730	0.9754	0.9742	0.9924
0.1335	7.0	980	0.0364	0.9849	0.9802	0.9825	0.9936
0.0146	8.0	1120	0.0343	0.9795	0.9778	0.9786	0.9941
0.0146	9.0	1260	0.0268	0.9802	0.9814	0.9808	0.9947
0.0146	10.0	1400	0.0427	0.9682	0.9688	0.9685	0.9918
0.0076	11.0	1540	0.0429	0.9576	0.9633	0.9605	0.9899
0.0076	12.0	1680	0.0343	0.9735	0.9730	0.9732	0.9933
0.0076	13.0	1820	0.0305	0.9801	0.9754	0.9777	0.9939
0.0076	14.0	1960	0.0437	0.9765	0.9742	0.9753	0.9924
0.0047	15.0	2100	0.0363	0.9778	0.9778	0.9778	0.9939

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3