Modelpublic

ahmedmohamed55/ast-tennis

sourceHugging Facebsd-3-clauseupdated 2mo agoView on Hugging Face

0likes211downloads

Model Card

ast-tennis

This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.1345
—Accuracy: 0.9828
—F1: 0.9815

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: 1e-05
—trainbatchsize: 16
—evalbatchsize: 16
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 15
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	15	0.5728	0.8103	0.8084
0.8515	2.0	30	0.2800	0.9138	0.9143
0.3041	3.0	45	0.1779	0.9310	0.9262
0.0946	4.0	60	0.1539	0.9310	0.9262
0.0946	5.0	75	0.0899	0.9483	0.9426
0.0256	6.0	90	0.0683	0.9828	0.9815
0.0062	7.0	105	0.0793	0.9655	0.9623
0.0017	8.0	120	0.1586	0.9310	0.9274
0.0017	9.0	135	0.1103	0.9828	0.9815
0.0008	10.0	150	0.1142	0.9828	0.9815
0.0006	11.0	165	0.1460	0.9655	0.9633
0.0005	12.0	180	0.1489	0.9655	0.9633
0.0005	13.0	195	0.1356	0.9828	0.9815
0.0005	14.0	210	0.1343	0.9828	0.9815
0.0005	15.0	225	0.1345	0.9828	0.9815

Framework versions

—Transformers 5.13.1
—Pytorch 2.11.0+cu128
—Datasets 4.0.0
—Tokenizers 0.22.2