dgx2_whisper_base_teacher_student_distill_att_loss_libri360_epochs_50_batch_8

This model is a fine-tuned version of rohitp1/subh_whisper_base_teacher_student_distill_att_loss_libri360_epochs_100_batch_4_ckpt1500 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.1009
• Wer: 7.2700

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: 0.0005
• train_batch_size: 8
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 256
• total_train_batch_size: 2048
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.2
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.5913	0.98	100	0.6243	6.3702
0.6028	1.97	200	0.6798	6.6119
0.6546	2.95	300	0.7314	6.9170
0.7386	3.94	400	1.3793	7.0099
1.2318	4.92	500	0.9823	7.0099
1.2304	5.91	600	1.1009	7.2700

Framework versions

• Transformers 4.28.1
• Pytorch 1.12.1
• Datasets 2.8.0
• Tokenizers 0.13.2