faster-whisper-medium-chichewa

README.md

@@ -4,6 +4,8 @@ license: apache-2.0
 base_model: openai/whisper-medium
 tags:
 - generated_from_trainer
+metrics:
+- wer
 model-index:
 - name: medium-model
   results: []
@@ -15,6 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
 # medium-model
 
 This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.2383
+- Wer: 0.6660
+- Cer: 0.2985
 
 ## Model description
 
@@ -34,20 +40,31 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
-- train_batch_size: 6
+- train_batch_size: 2
 - eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 12
+- total_train_batch_size: 4
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 500
 - training_steps: 5000
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch   | Step | Validation Loss | Wer    | Cer    |
+|:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
+| 0.0575        | 13.1579 | 1000 | 1.9806          | 0.6920 | 0.3415 |
+| 0.0118        | 26.3158 | 2000 | 2.0510          | 0.6764 | 0.3321 |
+| 0.0039        | 39.4737 | 3000 | 2.1584          | 0.6647 | 0.2982 |
+| 0.0011        | 52.6316 | 4000 | 2.2279          | 0.6997 | 0.3405 |
+| 0.0009        | 65.7895 | 5000 | 2.2383          | 0.6660 | 0.2985 |
+
+
 ### Framework versions
 
 - Transformers 4.48.0
-- Pytorch 2.6.0+cu124
+- Pytorch 2.8.0+cu126
 - Datasets 3.6.0
-- Tokenizers 0.21.2
+- Tokenizers 0.21.4

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77d0c4dbc02f5c791d090067a62dad5c06c328c903f5c232901ecb5d8e855560
+oid sha256:cb3cabbf5fb0424ff63ab7230121e01a041e7c7521d32fac4482092cc583fd70
 size 3055544304

runs/Nov05_06-56-09_e4580a34292e/events.out.tfevents.1762325781.e4580a34292e.483.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb81d03a9d65d5bca6d1d18593e135a1089e3e0cfd8d99d1a46955bd9f62147e
+size 30149

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ddd5d9ef933889bf05b7f386f9a142abc9dd0d03347d1393ac1eec05a95d2f9a
-size 5496
+oid sha256:d3b243009ae2adc29d06fdcfe6aa897dec00a74ccf2d7e238f906a91388ecb71
+size 5905