README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: wav2vec2-base-finetuned-ks
    results: []

wav2vec2-base-finetuned-ks

This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5131
• Accuracy: 0.8545

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• lr_scheduler_warmup_steps: 500
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	8	0.6930	0.4182
0.6924	2.0	16	0.6835	0.6
0.6802	3.0	24	0.6618	0.6909
0.649	4.0	32	0.6318	0.6364
0.6306	5.0	40	0.6066	0.6545
0.6306	6.0	48	0.5750	0.6545
0.597	7.0	56	0.5462	0.6909
0.5569	8.0	64	0.5165	0.7455
0.5379	9.0	72	0.4870	0.7818
0.4663	10.0	80	0.4436	0.8727
0.4663	11.0	88	0.4749	0.8364
0.4581	12.0	96	0.3678	0.8909
0.4244	13.0	104	0.4135	0.8545
0.3686	14.0	112	0.3410	0.9455
0.308	15.0	120	0.3530	0.8909
0.308	16.0	128	0.4086	0.8364
0.2677	17.0	136	0.3900	0.8545
0.2419	18.0	144	0.2027	0.9455
0.1973	19.0	152	0.3836	0.8727
0.1483	20.0	160	0.3371	0.8909
0.1483	21.0	168	0.5328	0.8
0.2048	22.0	176	0.3533	0.8909
0.2232	23.0	184	0.4151	0.8727
0.1864	24.0	192	0.2016	0.9273
0.2669	25.0	200	0.5131	0.8545

Framework versions

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