hadiqa123
/

train_model

@@ -12,10 +12,10 @@ should probably proofread and complete it, then remove this comment. -->
 # train_model
-This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 4.6009
-- Wer: 0.9988
 ## Model description
@@ -38,27 +38,45 @@ The following hyperparameters were used during training:
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 10
-- num_epochs: 10
 - mixed_precision_training: Native AMP
 ### Training results
-| Training Loss | Epoch | Step | Validation Loss | Wer    |
-|:-------------:|:-----:|:----:|:---------------:|:------:|
-| 17.8271       | 1.11  | 10   | 54.8523         | 1.0    |
-| 11.71         | 2.22  | 20   | 15.0626         | 0.9988 |
-| 4.6536        | 3.33  | 30   | 7.3361          | 0.9988 |
-| 3.6026        | 4.44  | 40   | 5.7365          | 0.9988 |
-| 3.3475        | 5.56  | 50   | 5.0688          | 0.9988 |
-| 3.2072        | 6.67  | 60   | 4.7947          | 0.9988 |
-| 3.1514        | 7.78  | 70   | 4.6413          | 0.9988 |
-| 3.1043        | 8.89  | 80   | 4.6918          | 0.9988 |
-| 3.1093        | 10.0  | 90   | 4.6009          | 0.9988 |
 ### Framework versions

 # train_model
+This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.5418
+- Wer: 0.3477
 ## Model description
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 1000
+- num_epochs: 30
 - mixed_precision_training: Native AMP
 ### Training results
+| Training Loss | Epoch | Step  | Validation Loss | Wer    |
+|:-------------:|:-----:|:-----:|:---------------:|:------:|
+| 3.4778        | 1.0   | 500   | 1.7193          | 0.9881 |
+| 0.8342        | 2.01  | 1000  | 0.5422          | 0.5306 |
+| 0.4304        | 3.01  | 1500  | 0.4456          | 0.4634 |
+| 0.2998        | 4.02  | 2000  | 0.4095          | 0.4283 |
+| 0.2348        | 5.02  | 2500  | 0.4446          | 0.4216 |
+| 0.1921        | 6.02  | 3000  | 0.5314          | 0.3949 |
+| 0.1576        | 7.03  | 3500  | 0.4374          | 0.4033 |
+| 0.1435        | 8.03  | 4000  | 0.6605          | 0.4036 |
+| 0.1261        | 9.04  | 4500  | 0.4944          | 0.3887 |
+| 0.1107        | 10.04 | 5000  | 0.4507          | 0.3806 |
+| 0.0994        | 11.04 | 5500  | 0.4927          | 0.3733 |
+| 0.0891        | 12.05 | 6000  | 0.5067          | 0.3754 |
+| 0.0862        | 13.05 | 6500  | 0.4767          | 0.3691 |
+| 0.0702        | 14.06 | 7000  | 0.4982          | 0.3739 |
+| 0.0648        | 15.06 | 7500  | 0.5233          | 0.3736 |
+| 0.0599        | 16.06 | 8000  | 0.5338          | 0.3694 |
+| 0.0588        | 17.07 | 8500  | 0.5675          | 0.3568 |
+| 0.0587        | 18.07 | 9000  | 0.5689          | 0.3657 |
+| 0.0461        | 19.08 | 9500  | 0.5803          | 0.3639 |
+| 0.0443        | 20.08 | 10000 | 0.5427          | 0.3654 |
+| 0.0436        | 21.08 | 10500 | 0.5441          | 0.3662 |
+| 0.035         | 22.09 | 11000 | 0.5511          | 0.3601 |
+| 0.0338        | 23.09 | 11500 | 0.4968          | 0.3581 |
+| 0.0327        | 24.1  | 12000 | 0.5254          | 0.3553 |
+| 0.0274        | 25.1  | 12500 | 0.5212          | 0.3524 |
+| 0.0246        | 26.1  | 13000 | 0.5445          | 0.3495 |
+| 0.0263        | 27.11 | 13500 | 0.5291          | 0.3500 |
+| 0.0228        | 28.11 | 14000 | 0.5378          | 0.3458 |
+| 0.0223        | 29.12 | 14500 | 0.5418          | 0.3477 |
 ### Framework versions