End of training

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4936
-- Wer: 1.1225
+- Loss: 0.4692
+- Wer: 0.4020
 
 ## Model description
 
@@ -38,13 +38,13 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.001
+- learning_rate: 0.004
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - 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
-- lr_scheduler_warmup_steps: 100
+- lr_scheduler_warmup_steps: 500
 - num_epochs: 10
 - mixed_precision_training: Native AMP
 
@@ -52,16 +52,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Wer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| No log        | 1.0   | 13   | 6.6604          | 1.0    |
-| No log        | 2.0   | 26   | 3.7699          | 1.0    |
-| No log        | 3.0   | 39   | 3.1415          | 1.0    |
-| No log        | 4.0   | 52   | 2.6606          | 1.0    |
-| No log        | 5.0   | 65   | 1.5002          | 1.0    |
-| No log        | 6.0   | 78   | 1.0075          | 1.0    |
-| No log        | 7.0   | 91   | 0.7862          | 1.0    |
-| 3.2148        | 8.0   | 104  | 0.6659          | 1.2530 |
-| 3.2148        | 9.0   | 117  | 0.5483          | 1.0079 |
-| 3.2148        | 10.0  | 130  | 0.4936          | 1.1225 |
+| 1.8364        | 1.0   | 13   | 1.7411          | 0.8907 |
+| 1.8276        | 2.0   | 26   | 1.5980          | 0.8391 |
+| 1.6815        | 3.0   | 39   | 1.4269          | 0.7822 |
+| 1.4402        | 4.0   | 52   | 1.2306          | 0.7050 |
+| 1.3527        | 5.0   | 65   | 1.0515          | 0.6605 |
+| 1.1983        | 6.0   | 78   | 0.8829          | 0.5821 |
+| 0.99          | 7.0   | 91   | 0.7602          | 0.5550 |
+| 0.9221        | 8.0   | 104  | 0.6441          | 0.4770 |
+| 0.8035        | 9.0   | 117  | 0.5535          | 0.4503 |
+| 0.6713        | 10.0  | 130  | 0.4692          | 0.4020 |
 
 
 ### Framework versions

adapter.dzo.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7b4005517cbf8b768de3351f2bf5b185a7aef4477942fbfa1a2d65642fd4ce2c
+oid sha256:846fe268463735741dd24384d5f8da520d59a2c35af6ad2cfccd061a008d64d5
 size 8952268