End of training

README.md

@@ -11,7 +11,7 @@ datasets:
 metrics:
 - wer
 model-index:
-- name: Whisper Small acholi
+- name: Whisper base acholi
   results:
   - task:
       name: Automatic Speech Recognition
@@ -22,18 +22,18 @@ model-index:
     metrics:
     - name: Wer
       type: wer
-      value: 121.88961646398504
+      value: 122.26379794200186
 ---
 
 <!-- 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 acholi
+# Whisper base acholi
 
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Sunbird dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.8028
-- Wer: 121.8896
+- Loss: 2.8895
+- Wer: 122.2638
 
 ## Model description
 
@@ -52,26 +52,29 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-05
+- learning_rate: 1e-05
 - train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 1000
-- training_steps: 1000
+- training_steps: 4000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Wer      |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 2.8441        | 3.32  | 1000 | 2.8028          | 121.8896 |
+| 3.2321        | 3.32  | 1000 | 2.9610          | 140.3181 |
+| 2.5056        | 6.64  | 2000 | 2.7358          | 116.9317 |
+| 2.0671        | 9.97  | 3000 | 2.7957          | 144.9953 |
+| 1.7382        | 13.29 | 4000 | 2.8895          | 122.2638 |
 
 
 ### Framework versions
 
 - Transformers 4.38.0.dev0
 - Pytorch 2.1.0+cu121
-- Datasets 2.16.1
-- Tokenizers 0.15.0
+- Datasets 2.17.1
+- Tokenizers 0.15.2

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8321f7da3ac5765efe140817f42ab6b04e6bac7c0e41f115ce5b11c643e9f15d
+oid sha256:d539c48a948fa363ae9b87a8557cba4106e3f3b340fa505e6ec6ea5d9971808a
 size 290401888