File size: 2,230 Bytes
d491804 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
---
library_name: transformers
license: apache-2.0
base_model: openai/whisper-medium.en
tags:
- generated_from_trainer
model-index:
- name: whisper-medium.en-merged_dataset
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# whisper-medium.en-merged_dataset
This model is a fine-tuned version of [openai/whisper-medium.en](https://huggingface.co/openai/whisper-medium.en) on the None dataset.
It achieves the following results on the evaluation set:
- Cer: 890.9343
- Loss: 0.5805
## 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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- 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
- num_epochs: 3
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Cer | Validation Loss |
|:-------------:|:------:|:----:|:--------:|:---------------:|
| 3.1316 | 0.2626 | 500 | 327.6237 | 1.4389 |
| 1.2083 | 0.5252 | 1000 | 456.3455 | 0.9819 |
| 0.9269 | 0.7878 | 1500 | 737.9578 | 0.8079 |
| 0.7618 | 1.0504 | 2000 | 966.5418 | 0.7155 |
| 0.5428 | 1.3130 | 2500 | 665.7924 | 0.6660 |
| 0.5139 | 1.5756 | 3000 | 899.3139 | 0.6320 |
| 0.4948 | 1.8382 | 3500 | 965.0479 | 0.6049 |
| 0.4122 | 2.1008 | 4000 | 942.7040 | 0.5902 |
| 0.3145 | 2.3634 | 4500 | 969.4258 | 0.5819 |
| 0.2576 | 2.6261 | 5000 | 890.9343 | 0.5805 |
### Framework versions
- Transformers 4.51.0
- Pytorch 2.7.1+cu128
- Datasets 3.6.0
- Tokenizers 0.21.1
|