Hugging Face's logo

xuliu15
/
Frisian_32r_LoRA_10h_new

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
Frisian_32r_LoRA_10h_new / README.md
xuliu15's picture
xuliu15
End of training
1911f1a verified
preview code
|
raw
history blame contribute delete
1.68 kB
metadata
license: apache-2.0
library_name: peft
tags:
  - generated_from_trainer
base_model: openai/whisper-small
datasets:
  - mozilla-foundation/common_voice_6_1
model-index:
  - name: LoRA-Frisian-10h-new
    results: []

LoRA-Frisian-10h-new

This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_6_1 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7553

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: 8
  • 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: 50
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
1.0227 1.0 935 1.0968
0.826 2.0 1870 0.8901
0.7334 3.0 2805 0.8065
0.648 4.0 3740 0.7682
0.655 5.0 4675 0.7553

Framework versions

  • PEFT 0.10.0
  • Transformers 4.40.1
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1