End of training

README.md

@@ -16,13 +16,10 @@ model-index:
     dataset:
       name: PolyAI/minds14
       type: PolyAI/minds14
-      config: en-US
-      split: train
-      args: en-US
     metrics:
     - name: Wer
       type: wer
-      value: 36.12750885478158
+      value: 0.33530106257378983
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,9 +29,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6316
-- Wer Ortho: 35.7187
-- Wer: 36.1275
+- Loss: 0.5824
+- Wer Ortho: 0.3424
+- Wer: 0.3353
 
 ## Model description
 
@@ -60,19 +57,22 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant_with_warmup
 - lr_scheduler_warmup_steps: 50
-- training_steps: 500
-- mixed_precision_training: Native AMP
+- training_steps: 250
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer     |
-|:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
-| 0.0012        | 17.86 | 500  | 0.6316          | 35.7187   | 36.1275 |
+| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer    |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
+| 0.3677        | 1.72  | 50   | 0.5198          | 0.3849    | 0.3648 |
+| 0.1925        | 3.45  | 100  | 0.5038          | 0.3671    | 0.3518 |
+| 0.0836        | 5.17  | 150  | 0.5206          | 0.3547    | 0.3406 |
+| 0.0265        | 6.9   | 200  | 0.5520          | 0.3627    | 0.3518 |
+| 0.008         | 8.62  | 250  | 0.5824          | 0.3424    | 0.3353 |
 
 
 ### Framework versions
 
-- Transformers 4.35.2
-- Pytorch 2.1.0+cu118
-- Datasets 2.15.0
-- Tokenizers 0.15.0
+- Transformers 4.32.0
+- Pytorch 2.1.0+cu121
+- Datasets 2.16.1
+- Tokenizers 0.13.3

generation_config.json

@@ -35,7 +35,7 @@
   "forced_decoder_ids": [
     [
       1,
-      50322
+      50259
     ],
     [
       2,
@@ -148,7 +148,7 @@
     "<|yo|>": 50325,
     "<|zh|>": 50260
   },
-  "language": "sinhalese",
+  "language": "english",
   "max_initial_timestamp_index": 1,
   "max_length": 448,
   "no_timestamps_token_id": 50363,
@@ -249,5 +249,5 @@
     "transcribe": 50359,
     "translate": 50358
   },
-  "transformers_version": "4.35.2"
+  "transformers_version": "4.32.0"
 }