CindyDelage/Challenge_HuggingFace_DFG_FrugalAI

FrugalAI

This model is a fine-tuned version of facebook/wav2vec2-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3650
• Accuracy: 0.9365

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.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: 0.05
• num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5669	0.9998	3527	0.2941	0.9211
1.3782	1.9998	7054	0.4237	0.9216
1.3008	2.9998	10581	0.4684	0.9113
1.2606	3.9998	14108	0.4323	0.9204
1.0533	4.9998	17635	0.3708	0.9331
0.8921	5.9998	21162	0.3650	0.9365

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0