update model card README.md

README.md

@@ -1,41 +1,38 @@
 ---
-language:
-- be
 license: apache-2.0
 tags:
-- whisper-event
 - generated_from_trainer
 datasets:
-- mozilla-foundation/common_voice_11_0
+- common_voice_11_0
 metrics:
 - wer
 model-index:
-- name: Whisper Tiny Belarusian
+- name: whisper-tiny-be-test
   results:
   - task:
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: mozilla-foundation/common_voice_11_0 be
-      type: mozilla-foundation/common_voice_11_0
+      name: common_voice_11_0
+      type: common_voice_11_0
       config: be
       split: validation
       args: be
     metrics:
     - name: Wer
       type: wer
-      value: 51.28205128205128
+      value: 46.7032967032967
 ---
 
 <!-- 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. -->
 
-# Whisper Tiny Belarusian
+# whisper-tiny-be-test
 
-This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the mozilla-foundation/common_voice_11_0 be dataset.
+This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the common_voice_11_0 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4686
-- Wer: 51.2821
+- Loss: 0.4282
+- Wer: 46.7033
 
 ## Model description
 
@@ -61,7 +58,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 10
-- training_steps: 200
+- training_steps: 300
 - mixed_precision_training: Native AMP
 
 ### Training results
@@ -88,6 +85,16 @@ The following hyperparameters were used during training:
 | 0.6803        | 0.9   | 180  | 0.4852          | 55.8608 |
 | 0.4813        | 0.95  | 190  | 0.4686          | 51.2821 |
 | 0.4952        | 1.0   | 200  | 0.4624          | 51.4652 |
+| 0.3956        | 0.03  | 210  | 0.4690          | 52.0147 |
+| 0.3719        | 0.07  | 220  | 0.4673          | 52.7473 |
+| 0.3168        | 0.1   | 230  | 0.4499          | 51.4652 |
+| 0.3582        | 0.13  | 240  | 0.4525          | 46.8864 |
+| 0.2475        | 0.17  | 250  | 0.4612          | 52.3810 |
+| 0.2988        | 0.2   | 260  | 0.4346          | 49.8168 |
+| 0.2749        | 0.23  | 270  | 0.4249          | 48.9011 |
+| 0.3368        | 0.27  | 280  | 0.4388          | 46.5201 |
+| 0.2574        | 0.3   | 290  | 0.4309          | 46.7033 |
+| 0.2921        | 0.33  | 300  | 0.4282          | 46.7033 |
 
 
 ### Framework versions

train.log

@@ -242,3 +242,5 @@
 {'loss': 0.2574, 'learning_rate': 4.482758620689655e-06, 'epoch': 0.3}
 {'eval_loss': 0.43085092306137085, 'eval_wer': 46.7032967032967, 'eval_runtime': 18.1023, 'eval_samples_per_second': 3.535, 'eval_steps_per_second': 0.11, 'epoch': 0.3}
 {'loss': 0.2921, 'learning_rate': 1.0344827586206898e-06, 'epoch': 0.33}
+{'eval_loss': 0.4282010793685913, 'eval_wer': 46.7032967032967, 'eval_runtime': 18.1178, 'eval_samples_per_second': 3.532, 'eval_steps_per_second': 0.11, 'epoch': 0.33}
+{'train_runtime': 1208.0467, 'train_samples_per_second': 7.947, 'train_steps_per_second': 0.248, 'train_loss': 0.10500287771224975, 'epoch': 0.33}