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 Small 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: 60.43956043956044
+      value: 55.67765567765568
 ---
 
 <!-- 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 Small 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: 0.6197
-- Wer: 60.4396
+- Loss: 0.5387
+- Wer: 55.6777
 
 ## Model description
 
@@ -54,14 +51,14 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0001
+- learning_rate: 3.1578947368421056e-06
 - train_batch_size: 32
 - eval_batch_size: 32
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 5
-- training_steps: 100
+- training_steps: 150
 - mixed_precision_training: Native AMP
 
 ### Training results
@@ -78,6 +75,11 @@ The following hyperparameters were used during training:
 | 0.7832        | 0.8   | 80   | 0.6129          | 65.9341 |
 | 0.6031        | 0.9   | 90   | 0.5877          | 61.3553 |
 | 0.6678        | 1.0   | 100  | 0.5759          | 61.5385 |
+| 0.4611        | 0.07  | 110  | 0.5625          | 57.6923 |
+| 0.4451        | 0.13  | 120  | 0.5636          | 56.5934 |
+| 0.3615        | 0.2   | 130  | 0.5490          | 61.1722 |
+| 0.4055        | 0.27  | 140  | 0.5382          | 55.1282 |
+| 0.2946        | 0.33  | 150  | 0.5387          | 55.6777 |
 
 
 ### Framework versions

train.log

@@ -56,3 +56,5 @@
 {'loss': 0.4055, 'learning_rate': 8.96551724137931e-06, 'epoch': 0.27}
 {'eval_loss': 0.5382302403450012, 'eval_wer': 55.12820512820513, 'eval_runtime': 22.4274, 'eval_samples_per_second': 2.854, 'eval_steps_per_second': 0.089, 'epoch': 0.27}
 {'loss': 0.2946, 'learning_rate': 2.0689655172413796e-06, 'epoch': 0.33}
+{'eval_loss': 0.53872150182724, 'eval_wer': 55.67765567765568, 'eval_runtime': 20.4177, 'eval_samples_per_second': 3.135, 'eval_steps_per_second': 0.098, 'epoch': 0.33}
+{'train_runtime': 451.4438, 'train_samples_per_second': 10.633, 'train_steps_per_second': 0.332, 'train_loss': 0.13119232177734375, 'epoch': 0.33}