ckpts

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

• Loss: 0.2980
• Accuracy: 0.9545

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: 1e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1628	1.0	223	0.7126	0.7727
0.6562	2.0	446	0.5069	0.8485
0.4199	3.0	669	0.3570	0.8990
0.325	4.0	892	0.2092	0.9394
0.2217	5.0	1115	0.2392	0.9444
0.1831	6.0	1338	0.2754	0.9293
0.1598	7.0	1561	0.3294	0.9343
0.1676	8.0	1784	0.2669	0.9495
0.1597	9.0	2007	0.3438	0.9293
0.1132	10.0	2230	0.3159	0.9444
0.1224	11.0	2453	0.2980	0.9545
0.095	12.0	2676	0.2970	0.9444
0.1087	13.0	2899	0.3449	0.9343
0.1254	14.0	3122	0.3198	0.9444

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2