End of training

README.md

@@ -1,4 +1,5 @@
 ---
+library_name: transformers
 language:
 - ne
 license: apache-2.0
@@ -7,17 +8,35 @@ tags:
 - generated_from_trainer
 datasets:
 - openslr/openslr
+metrics:
+- wer
 model-index:
-- name: Whisper Medium Nepali - Kiran Pantha
-  results: []
+- name: Whisper Large Nepali - Kiran Pantha
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: OpenSLR54
+      type: openslr/openslr
+      config: default
+      split: test
+      args: 'config: ne, split: test'
+    metrics:
+    - name: Wer
+      type: wer
+      value: 48.043676069153776
 ---
 
 <!-- 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 Medium Nepali - Kiran Pantha
+# Whisper Large Nepali - Kiran Pantha
 
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the OpenSLR54 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3013
+- Wer: 48.0437
 
 ## Model description
 
@@ -36,20 +55,35 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 3e-05
-- train_batch_size: 8
-- eval_batch_size: 4
+- learning_rate: 1e-05
+- train_batch_size: 16
+- eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 16
-- total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 200
-- training_steps: 2000
+- lr_scheduler_warmup_steps: 100
+- training_steps: 1000
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Wer     |
+|:-------------:|:------:|:----:|:---------------:|:-------:|
+| 0.8572        | 0.4785 | 100  | 0.4851          | 76.0692 |
+| 0.4           | 0.9569 | 200  | 0.3592          | 64.0582 |
+| 0.2329        | 1.4354 | 300  | 0.3153          | 56.2329 |
+| 0.2098        | 1.9139 | 400  | 0.2918          | 53.5032 |
+| 0.1189        | 2.3923 | 500  | 0.2865          | 51.4104 |
+| 0.096         | 2.8708 | 600  | 0.2835          | 50.7734 |
+| 0.0565        | 3.3493 | 700  | 0.2984          | 50.9554 |
+| 0.0425        | 3.8278 | 800  | 0.2947          | 48.7716 |
+| 0.027         | 4.3062 | 900  | 0.3007          | 49.4995 |
+| 0.0174        | 4.7847 | 1000 | 0.3013          | 48.0437 |
+
 
 ### Framework versions
 
-- Transformers 4.44.0
-- Pytorch 2.3.1+cu121
-- Datasets 2.20.0
+- Transformers 4.44.2
+- Pytorch 2.4.0+cu121
+- Datasets 2.21.0
 - Tokenizers 0.19.1

generation_config.json

@@ -252,5 +252,5 @@
     "transcribe": 50359,
     "translate": 50358
   },
-  "transformers_version": "4.44.0"
+  "transformers_version": "4.44.2"
 }