Model save

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6249
-- Wer: 73.7172
+- Loss: 0.4687
+- Wer: 50.5902
 
 ## Model description
 
@@ -39,23 +39,25 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 3.75e-05
 - train_batch_size: 16
-- eval_batch_size: 4
+- eval_batch_size: 1
 - seed: 42
 - gradient_accumulation_steps: 2
 - total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 312
-- num_epochs: 5
+- lr_scheduler_type: constant
+- lr_scheduler_warmup_steps: 50
+- training_steps: 200
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Wer      |
-|:-------------:|:------:|:----:|:---------------:|:--------:|
-| 2.2169        | 1.2698 | 40   | 1.7869          | 241.5580 |
-| 1.0037        | 2.5397 | 80   | 0.8910          | 78.5521  |
-| 0.5907        | 3.8095 | 120  | 0.6249          | 73.7172  |
+| Training Loss | Epoch  | Step | Validation Loss | Wer     |
+|:-------------:|:------:|:----:|:---------------:|:-------:|
+| 0.6584        | 1.2698 | 40   | 0.5936          | 65.0084 |
+| 0.3575        | 2.5397 | 80   | 0.4684          | 55.3120 |
+| 0.2254        | 3.8095 | 120  | 0.4344          | 50.0    |
+| 0.1486        | 5.0794 | 160  | 0.4562          | 52.1079 |
+| 0.0868        | 6.3492 | 200  | 0.4687          | 50.5902 |
 
 
 ### Framework versions

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa28468ac9b52d0b1f851838fb26d241bc1fa9fe621087a2fedf8be18e932834
+oid sha256:e0e9a8a6a2405f1ef747c57eb67d6430089edf114f56b97ad0d448acd182f8f2
 size 151099494