Modelpublic

bigmorning/whisper_input_decoder_no_lob_with_force__0015

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face

0likes11downloads

Model Card

whisperinputdecodernolobwithforce__0015

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: 4.1748
—Train Accuracy: 0.0130
—Train Wermet: 0.7090
—Validation Loss: 3.3528
—Validation Accuracy: 0.0127
—Validation Wermet: 0.7856
—Epoch: 14

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', 'learningrate': 1e-05, 'decay': 0.0, 'beta1': 0.9, 'beta2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weightdecay_rate': 0.01}
—training_precision: float32

Training results

Train Loss	Train Accuracy	Train Wermet	Validation Loss	Validation Accuracy	Validation Wermet	Epoch
5.4192	0.0107	1.1095	3.9799	0.0114	0.9502	0
4.7193	0.0116	0.8751	3.9335	0.0114	0.9414	1
4.6743	0.0117	0.8498	3.9949	0.0112	0.9682	2
4.6508	0.0117	0.8454	3.8900	0.0114	0.9453	3
4.6308	0.0118	0.8337	3.8853	0.0114	0.9495	4
4.6102	0.0118	0.8215	3.8884	0.0115	0.9205	5
4.5940	0.0118	0.8132	3.8409	0.0116	0.9007	6
4.5703	0.0119	0.7971	3.8224	0.0116	0.9098	7
4.5470	0.0120	0.7822	3.8013	0.0116	0.8938	8
4.5219	0.0120	0.7679	3.7776	0.0117	0.8829	9
4.4859	0.0121	0.7519	3.7360	0.0118	0.8411	10
4.4408	0.0123	0.7412	3.6972	0.0118	0.8593	11
4.3774	0.0124	0.7240	3.6035	0.0121	0.8234	12
4.2906	0.0127	0.7168	3.5057	0.0123	0.8130	13
4.1748	0.0130	0.7090	3.3528	0.0127	0.7856	14

Framework versions

—Transformers 4.33.0.dev0
—TensorFlow 2.13.0
—Tokenizers 0.13.3