jialicheng/unlearn_speech_commands_hubert-base_scrub_10_42

superb_ks_42

This model is a fine-tuned version of facebook/hubert-base-ls960 on the superb dataset. It achieves the following results on the evaluation set:

• Loss: 369.8440
• Accuracy: 0.6209
• Test Accuracy: 0.6209
• Df Accuracy: 0.1405
• Unlearn Overall Accuracy: 0.7402
• Unlearn Time: 1236.5284

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: 32
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Overall Accuracy	Unlearn Overall Accuracy	Time
0.0	1.0	1438	369.8440	0.1405	0.7402	0.7402	-1
0.0	2.0	2876	1400.2289	0.1405	0.7402	0.7402	-1
0.0	3.0	4314	2678.3972	0.1405	0.7402	0.7402	-1
0.0	4.0	5752	3715.6882	0.1405	0.7402	0.7402	-1
0.0	5.0	7190	4112.1831	0.1405	0.7402	0.7402	-1

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.2+cu118
• Datasets 2.18.0
• Tokenizers 0.15.2