Model save

README.md

@@ -1,6 +1,8 @@
 ---
 tags:
 - generated_from_trainer
+metrics:
+- wer
 model-index:
 - name: whisper-tiny.en
   results: []
@@ -12,6 +14,9 @@ should probably proofread and complete it, then remove this comment. -->
 # whisper-tiny.en
 
 This model was trained from scratch on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8196
+- Wer: 41.9679
 
 ## Model description
 
@@ -31,18 +36,22 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-06
-- train_batch_size: 4
-- eval_batch_size: 4
+- train_batch_size: 2
+- eval_batch_size: 2
 - seed: 42
 - gradient_accumulation_steps: 8
-- total_train_batch_size: 32
+- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
 - num_epochs: 10
+- mixed_precision_training: Native AMP
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss | Wer     |
+|:-------------:|:-----:|:----:|:---------------:|:-------:|
+| 0.6241        | 6.02  | 500  | 0.8196          | 41.9679 |
 
 
 ### Framework versions

generation_config.json

@@ -47,10 +47,11 @@
     ]
   ],
   "is_multilingual": false,
-  "max_initial_timestamp_index": 1,
+  "max_initial_timestamp_index": 50,
   "max_length": 448,
   "no_timestamps_token_id": 50362,
   "pad_token_id": 50256,
+  "prev_sot_token_id": 50360,
   "return_timestamps": false,
   "suppress_tokens": [
     1,
@@ -144,5 +145,5 @@
     50360,
     50361
   ],
-  "transformers_version": "4.33.0.dev0"
+  "transformers_version": "4.35.2"
 }

runs/Dec30_14-09-57_1f62fb92ba29/events.out.tfevents.1735567798.1f62fb92ba29.2191.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2fac9bbe84fcffe28f2881d38514261b38813812d39b3347a5de4b8e2e9f4450
-size 7078
+oid sha256:bec8697a6c8668ed9511a6e177bf7cbf9377bb4708685fd2f3d8ed1693eb8f5e
+size 7903