retrain_first1epoch

alexziweiwang/retrain_first1epoch

retrain_first1epoch is a machine learning model from alexziweiwang. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

4% of all-time downloads

All-time downloads

221

Public

Repo size

2.5 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin1.3 GB · 100%

At a glance

Library: transformers
Model type: wav2vec2
Access: Public
Created: Nov 28, 2022
Updated: Nov 28, 2022
SHA: bcb298cb

Model card

From the Hugging Face model README

retrain_first1epoch

This model is a fine-tuned version of alexziweiwang/exp21-uaspeech-foundation on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 8.2238
Acc: 0.24
Wer: 1.0
Correct: 48
Total: 200
Strlen: 200

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: 9e-06
train_batch_size: 2
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 8
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Acc	Wer	Correct	Total	Strlen
No log	0.02	5	13.8048	0.22	1.0237	44	200	200
12.2209	0.04	10	13.6869	0.22	1.0257	44	200	200
12.2209	0.06	15	13.5691	0.225	1.0296	45	200	200
12.4299	0.08	20	13.4590	0.24	1.0375	48	200	200
12.4299	0.11	25	13.3508	0.235	1.0395	47	200	200
11.0298	0.13	30	13.2241	0.25	1.0375	50	200	200
11.0298	0.15	35	13.0757	0.245	1.0336	49	200	200
10.5248	0.17	40	12.9277	0.245	1.0316	49	200	200
10.5248	0.19	45	12.7784	0.25	1.0316	50	200	200
10.8585	0.21	50	12.6346	0.25	1.0277	50	200	200
10.8585	0.23	55	12.4939	0.25	1.0277	50	200	200
10.7046	0.25	60	12.3472	0.25	1.0257	50	200	200
10.7046	0.27	65	12.1962	0.25	1.0237	50	200	200
10.8031	0.3	70	12.0537	0.25	1.0257	50	200	200
10.8031	0.32	75	11.9088	0.25	1.0237	50	200	200
10.859	0.34	80	11.7693	0.25	1.0257	50	200	200
10.859	0.36	85	11.6214	0.25	1.0198	50	200	200
9.7886	0.38	90	11.4699	0.25	1.0178	50	200	200
9.7886	0.4	95	11.3182	0.25	1.0138	50	200	200
10.4627	0.42	100	11.1609	0.25	1.0119	50	200	200
10.4627	0.44	105	11.0017	0.25	1.0138	50	200	200
10.0619	0.46	110	10.8520	0.25	1.0138	50	200	200
10.0619	0.48	115	10.7096	0.25	1.0138	50	200	200
8.7443	0.51	120	10.5629	0.25	1.0138	50	200	200
8.7443	0.53	125	10.4111	0.25	1.0119	50	200	200
9.675	0.55	130	10.2606	0.25	1.0119	50	200	200
9.675	0.57	135	10.1125	0.245	1.0119	49	200	200
9.1918	0.59	140	9.9708	0.24	1.0040	48	200	200
9.1918	0.61	145	9.8248	0.24	1.0040	48	200	200
9.6798	0.63	150	9.6785	0.24	1.0040	48	200	200
9.6798	0.65	155	9.5309	0.24	1.0040	48	200	200
9.0181	0.67	160	9.3867	0.24	1.0040	48	200	200
9.0181	0.7	165	9.2432	0.24	1.0040	48	200	200
7.7446	0.72	170	9.1053	0.24	1.0040	48	200	200
7.7446	0.74	175	8.9743	0.24	1.0040	48	200	200
8.0251	0.76	180	8.8538	0.24	1.0040	48	200	200
8.0251	0.78	185	8.7473	0.24	1.0020	48	200	200
7.9652	0.8	190	8.6516	0.24	1.0020	48	200	200
7.9652	0.82	195	8.5661	0.24	1.0020	48	200	200
7.9537	0.84	200	8.4887	0.24	1.0020	48	200	200
7.9537	0.86	205	8.4206	0.24	1.0	48	200	200
7.2889	0.89	210	8.3644	0.24	1.0	48	200	200
7.2889	0.91	215	8.3169	0.24	1.0	48	200	200
7.8974	0.93	220	8.2789	0.24	1.0	48	200	200
7.8974	0.95	225	8.2514	0.24	1.0	48	200	200
6.9118	0.97	230	8.2330	0.24	1.0	48	200	200
6.9118	0.99	235	8.2238	0.24	1.0	48	200	200

Framework versions

Transformers 4.23.1
Pytorch 1.12.1+cu113
Datasets 1.18.3
Tokenizers 0.13.2

Library: transformers
Type: wav2vec2
Created: Nov 28, 2022
Updated: Nov 28, 2022