metadata

library_name: transformers
license: apache-2.0
base_model: openai/whisper-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: whisper-base-finetuned-bmd
    results: []

whisper-base-finetuned-bmd

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

Loss: 1.3900
Accuracy: 0.3235
F1: 0.3095
Precision: 0.4512
Recall: 0.3235
Sensitivity: 0.3235

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: 16
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Sensitivity
No log	0.8571	3	1.1235	0.2647	0.1108	0.0701	0.2647	0.2647
No log	2.0	7	1.1253	0.2647	0.1108	0.0701	0.2647	0.2647
1.0608	2.8571	10	1.1178	0.2647	0.1162	0.0744	0.2647	0.2647
1.0608	4.0	14	1.1206	0.2941	0.1740	0.4886	0.2941	0.2941
1.0608	4.8571	17	1.1225	0.2941	0.2083	0.3475	0.2941	0.2941
0.9214	6.0	21	1.1167	0.4412	0.4108	0.5259	0.4412	0.4412
0.9214	6.8571	24	1.0754	0.5	0.4625	0.4954	0.5	0.5
0.9214	8.0	28	1.1578	0.4118	0.3959	0.4304	0.4118	0.4118
0.6179	8.8571	31	1.2143	0.3824	0.3663	0.4120	0.3824	0.3824
0.6179	10.0	35	1.3170	0.4118	0.4169	0.5174	0.4118	0.4118
0.6179	10.8571	38	1.3484	0.3529	0.3484	0.4777	0.3529	0.3529
0.3513	12.0	42	1.3904	0.3235	0.3095	0.4512	0.3235	0.3235
0.3513	12.8571	45	1.3900	0.3235	0.3095	0.4512	0.3235	0.3235

Framework versions

Transformers 4.46.2
Pytorch 2.5.0+cu121
Datasets 3.1.0
Tokenizers 0.20.3