antonvinny/gs3-test
By antonvinny · ↓ 2 · ♥ 1 · Updated 2025-01-16
What's in this repo
This page lists the files in antonvinny/gs3-test, including decoder_model_merged_quantized.onnx, decoder_model_quantized.onnx, decoder_with_past_model_quantized.onnx and more.
- 4 ONNX models (.onnx files)
- 29 files, including 18 .json, 4 .onnx, 2 .txt, 1 .md, 1 .safetensors, 1 .1
- Last updated 2025-01-16
ONNX files
| File | Size | Download |
|---|---|---|
| onnx/decoder_model_merged_quantized.onnx | — | Download |
| onnx/decoder_model_quantized.onnx | — | Download |
| onnx/decoder_with_past_model_quantized.onnx | — | Download |
| onnx/encoder_model_quantized.onnx | — | Download |
How to use
These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.
README
whisper-tiny-gs3
This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0001
- Wer: 3.0326
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: 8
- seed: 42
- 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_steps: 100
- training_steps: 600
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0013 | 50.0 | 200 | 0.0007 | 3.0326 |
| 0.0002 | 100.0 | 400 | 0.0002 | 3.0326 |
| 0.0001 | 150.0 | 600 | 0.0001 | 3.0326 |
Framework versions
- Transformers 4.48.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0
Model info
Created: 2025-01-16
Updated: 2025-01-16
View on Hugging Face