whisper_charsplit_new_round3__0016

This model is a fine-tuned version of bigmorning/whisper_charsplit_new_round2__0061 on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0000
• Train Accuracy: 0.0795
• Train Wermet: 7.7277
• Validation Loss: 0.5752
• Validation Accuracy: 0.0771
• Validation Wermet: 6.8671
• Epoch: 15

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	Train Wermet	Validation Loss	Validation Accuracy	Validation Wermet	Epoch
0.0009	0.0795	7.9492	0.5730	0.0769	7.2856	0
0.0015	0.0795	8.4221	0.5756	0.0769	7.1487	1
0.0012	0.0795	7.8476	0.5699	0.0769	6.5976	2
0.0010	0.0795	7.6843	0.5740	0.0769	6.9513	3
0.0014	0.0795	8.0796	0.5763	0.0768	7.4043	4
0.0019	0.0795	7.7274	0.5724	0.0769	6.4922	5
0.0008	0.0795	7.3468	0.5734	0.0769	6.1909	6
0.0009	0.0795	7.2393	0.5816	0.0769	6.5734	7
0.0010	0.0795	7.5822	0.5755	0.0769	6.6613	8
0.0004	0.0795	7.3807	0.5698	0.0770	7.0671	9
0.0001	0.0795	7.7157	0.5681	0.0771	6.8391	10
0.0001	0.0795	7.7540	0.5725	0.0771	6.9281	11
0.0001	0.0795	7.7721	0.5726	0.0771	6.8911	12
0.0000	0.0795	7.8163	0.5721	0.0771	6.8876	13
0.0000	0.0795	7.7745	0.5741	0.0771	6.8770	14
0.0000	0.0795	7.7277	0.5752	0.0771	6.8671	15

Framework versions

• Transformers 4.32.0.dev0
• TensorFlow 2.12.0
• Tokenizers 0.13.3