End of training

README.md

@@ -0,0 +1,84 @@
+---
+library_name: peft
+language:
+- fr
+license: apache-2.0
+base_model: openai/whisper-small
+tags:
+- base_model:adapter:openai/whisper-small
+- lora
+- transformers
+metrics:
+- wer
+model-index:
+- name: 52Hz Small Fr - IMT Atlantique X 52 Hertz
+  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. -->
+
+# 52Hz Small Fr - IMT Atlantique X 52 Hertz
+
+This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Premier dataset organisé de 52 Hertz dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5418
+- Wer: 44.3978
+
+## 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: 0.0001
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: constant_with_warmup
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 15
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer      |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 1.4744        | 1.0   | 23   | 1.4905          | 103.3613 |
+| 1.2047        | 2.0   | 46   | 1.1278          | 61.3445  |
+| 0.8717        | 3.0   | 69   | 0.8429          | 54.2017  |
+| 0.5948        | 4.0   | 92   | 0.6974          | 50.2801  |
+| 0.4514        | 5.0   | 115  | 0.6192          | 50.0     |
+| 0.3894        | 6.0   | 138  | 0.5874          | 48.8796  |
+| 0.3317        | 7.0   | 161  | 0.5678          | 47.6190  |
+| 0.2881        | 8.0   | 184  | 0.5390          | 49.8599  |
+| 0.2545        | 9.0   | 207  | 0.5351          | 45.7983  |
+| 0.2414        | 10.0  | 230  | 0.5308          | 46.6387  |
+| 0.1757        | 11.0  | 253  | 0.5375          | 46.0784  |
+| 0.173         | 12.0  | 276  | 0.5290          | 45.3782  |
+| 0.1261        | 13.0  | 299  | 0.5295          | 43.6975  |
+| 0.1224        | 14.0  | 322  | 0.5263          | 45.3782  |
+| 0.0956        | 15.0  | 345  | 0.5418          | 44.3978  |
+
+
+### Framework versions
+
+- PEFT 0.18.1
+- Transformers 4.57.3
+- Pytorch 2.9.1+cu130
+- Datasets 4.4.2
+- Tokenizers 0.22.2

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87b0e5cc46172ffd6e08486102bb6a960a39683cb196a65b912b9bec72eaf6f9
+oid sha256:6240ade302c93fd00747d93eccb849221e434b6cbdf9bc8a8c2463b606a6c132
 size 7098064