basic_train_basic_test 1000 similar params: per_device_train_batch_size=16, # bylo 16 a pod tim 1 gradient_accumulation_steps=1, warmup_steps=250, max_steps=2000
This model is a fine-tuned version of openai/whisper-small on the xbilek25/basic_train_set_en_last3cs_1000 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1421
- Wer: 14.5731
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: 16
- 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_steps: 250
- training_steps: 2000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.0008 | 7.03 | 500 | 0.1370 | 16.9296 |
0.0003 | 15.03 | 1000 | 0.1390 | 14.3956 |
0.0002 | 23.03 | 1500 | 0.1412 | 14.6125 |
0.0002 | 31.02 | 2000 | 0.1421 | 14.5731 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.15.2
- Downloads last month
- 7
Finetuned from
Dataset used to train xbilek25/whisper-small-train-basic_1000_v1.0_shuffled
Evaluation results
- Wer on xbilek25/basic_train_set_en_last3cs_1000self-reported14.573