metadata

license: apache-2.0
base_model: ntu-spml/distilhubert
tags:
  - generated_from_trainer
datasets:
  - audiofolder
metrics:
  - accuracy
model-index:
  - name: distilhubert-finetuned-donateacry
    results:
      - task:
          name: Audio Classification
          type: audio-classification
        dataset:
          name: audiofolder
          type: audiofolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8932584269662921

distilhubert-finetuned-donateacry

This model is a fine-tuned version of ntu-spml/distilhubert on the audiofolder dataset. It achieves the following results on the evaluation set:

Loss: 0.5034
Accuracy: 0.8933

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: 0.001
train_batch_size: 8
eval_batch_size: 8
seed: 123
gradient_accumulation_steps: 8
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.9888	11	0.9525	0.7303
No log	1.9775	22	1.2765	0.5393
No log	2.9663	33	0.6634	0.7978
No log	3.9551	44	0.6369	0.8202
No log	4.9438	55	0.5328	0.8596
No log	5.9326	66	0.5146	0.8652
No log	6.9213	77	0.5200	0.8764
No log	8.0	89	0.5213	0.8708
No log	8.9888	100	0.6062	0.8596
No log	9.9775	111	0.5938	0.8652
No log	10.9663	122	0.5247	0.8652
No log	11.9551	133	0.7004	0.8483
No log	12.9438	144	0.5388	0.8876
No log	13.9326	155	0.4856	0.8876
No log	14.9213	166	0.5380	0.8764
No log	16.0	178	0.5055	0.8876
No log	16.9888	189	0.5217	0.8876
No log	17.9775	200	0.5034	0.8933
No log	18.9663	211	0.4745	0.8876
No log	19.9551	222	0.4812	0.8876
No log	20.9438	233	0.4709	0.8820
No log	21.9326	244	0.4824	0.8876
No log	22.9213	255	0.4819	0.8876
No log	24.0	267	0.4877	0.8933
No log	24.7191	275	0.4866	0.8933

Framework versions

Transformers 4.42.3
Pytorch 2.3.1+cu118
Datasets 2.20.0
Tokenizers 0.19.1