Model save

README.md

@@ -1,11 +1,8 @@
 ---
 library_name: transformers
 license: cc-by-nc-4.0
-base_model: facebook/mms-1b-all
+base_model: csikasote/mms-1b-all-swagen-combined-m50f50-dnn-42-0.09
 tags:
-- automatic-speech-recognition
-- swagen
-- mms
 - generated_from_trainer
 model-index:
 - name: mms-1b-all-swagen-combined-m50f50-dnn-42-0.09
@@ -17,10 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mms-1b-all-swagen-combined-m50f50-dnn-42-0.09
 
-This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the SWAGEN - DAT dataset.
+This model is a fine-tuned version of [csikasote/mms-1b-all-swagen-combined-m50f50-dnn-42-0.09](https://huggingface.co/csikasote/mms-1b-all-swagen-combined-m50f50-dnn-42-0.09) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3616
-- Cer: 0.0603
+- Loss: 0.3460
+- Cer: 0.0870
 
 ## Model description
 
@@ -47,19 +44,19 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 16
 - 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: 250
-- num_epochs: 15.0
+- lr_scheduler_warmup_ratio: 0.05
+- lr_scheduler_warmup_steps: 100
+- num_epochs: 30.0
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss | Cer    |
 |:-------------:|:------:|:----:|:---------------:|:------:|
-| 2.1832        | 1.9723 | 250  | 2.4165          | 0.7603 |
-| 0.2443        | 3.9407 | 500  | 0.3617          | 0.0603 |
-| 0.8610        | 5.9091 | 750  | 0.3945          | 0.0791 |
-| 0.0804        | 7.8775 | 1000 | 0.4205          | 0.0874 |
-| 0.1375        | 9.8458 | 1250 | 0.4118          | 0.0884 |
+| 0.1212        | 0.7905 | 100  | 0.2882          | 0.0667 |
+| 0.6450        | 1.5771 | 200  | 0.3139          | 0.0766 |
+| 0.1175        | 2.3636 | 300  | 0.3341          | 0.0834 |
+| 0.2012        | 3.1502 | 400  | 0.3460          | 0.0870 |
 
 
 ### Framework versions