Configuration Parsing Warning: In adapter_config.json: "peft.task_type" must be a string

whisper-large-v2-ft-tms-BTU6567_silence_base-on-car350-250422-v1

This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 2.8963

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: 8
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • total_train_batch_size: 64
  • total_eval_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.2
  • num_epochs: 30
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
13.985 1.0 1 14.5871
14.0198 2.0 2 14.5871
13.9792 3.0 3 14.5871
13.9951 4.0 4 14.5871
13.995 5.0 5 14.5871
14.0008 6.0 6 14.3061
13.7294 7.0 7 14.3061
13.7231 8.0 8 13.6134
13.0157 9.0 9 12.5096
11.682 10.0 10 10.4350
9.3305 11.0 11 7.3042
7.4152 12.0 12 6.3486
6.202 13.0 13 5.1358
5.0984 14.0 14 4.7347
4.8599 15.0 15 4.5979
4.7433 16.0 16 4.4747
4.6267 17.0 17 4.3350
4.4997 18.0 18 4.1838
4.3673 19.0 19 4.0262
4.2335 20.0 20 3.8706
4.0869 21.0 21 3.7173
3.9376 22.0 22 3.5768
3.7974 23.0 23 3.4486
3.674 24.0 24 3.3342
3.5471 25.0 25 3.2328
3.4338 26.0 26 3.1428
3.3387 27.0 27 3.0653
3.2617 28.0 28 2.9998
3.1746 29.0 29 2.9434
3.1137 30.0 30 2.8963

Framework versions

  • PEFT 0.13.0
  • Transformers 4.45.1
  • Pytorch 2.5.0+cu124
  • Datasets 3.0.1
  • Tokenizers 0.20.0
Downloads last month
3
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for dylanewbie/whisper-large-v2-ft-tms-BTU6567_silence_base-on-car350-250422-v1

Adapter
(257)
this model