Model save

README.md

@@ -14,19 +14,19 @@ model-index:
 <!-- 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. -->
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/keviinkibe/huggingface/runs/3b4u4ih8)
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/keviinkibe/huggingface/runs/3b4u4ih8)
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/keviinkibe/huggingface/runs/7iywhhxw)
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/keviinkibe/huggingface/runs/7iywhhxw)
 # whisper-small-finetuned-finetuned
 
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the common_voice_16_1 dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 4.2187
-- eval_wer: 277.0115
-- eval_runtime: 37.2641
-- eval_samples_per_second: 0.268
-- eval_steps_per_second: 0.027
-- epoch: 19.05
-- step: 20
+- eval_loss: 3.4388
+- eval_wer: 101.3774
+- eval_runtime: 370.6019
+- eval_samples_per_second: 0.27
+- eval_steps_per_second: 0.011
+- epoch: 6.008
+- step: 100
 
 ## Model description
 
@@ -46,18 +46,19 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 16
-- eval_batch_size: 16
+- train_batch_size: 32
+- eval_batch_size: 32
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- training_steps: 20
+- lr_scheduler_type: constant_with_warmup
+- lr_scheduler_warmup_steps: 10
+- training_steps: 500
 - mixed_precision_training: Native AMP
 
 ### Framework versions
 
 - PEFT 0.11.1
 - Transformers 4.42.3
-- Pytorch 2.2.2+cu121
+- Pytorch 2.4.1+cu121
 - Datasets 2.19.2
 - Tokenizers 0.19.1