3_epoch_noFirVox

This model is a fine-tuned version of openai/whisper-tiny on the /home/investigacion/disco4TB/workspace_pablo/firvox_whisper_research/finetunnig/dataset/dataset_parquet/dataset_1000x6_noFirVox_correctedpaths.parquet dataset. It achieves the following results on the evaluation set:

• Loss: 0.4373
• Accuracy: 0.87

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 32
• seed: 0
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 2
• 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: 0.1
• num_epochs: 3.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9061	1.0	80	0.7686	0.7856
0.4682	2.0	160	0.5186	0.8389
0.286	3.0	240	0.4373	0.87

Framework versions

• Transformers 4.44.1
• Pytorch 1.11.0
• Datasets 2.19.1
• Tokenizers 0.19.1