End of training

README.md

@@ -6,6 +6,11 @@ license: apache-2.0
 base_model: openai/whisper-base
 tags:
 - generated_from_trainer
+metrics:
+- wer
+- precision
+- recall
+- f1
 model-index:
 - name: ./whisper-base-ea-1hrsd
   results: []
@@ -18,18 +23,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the Afrispeech-200 dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 0.6967
-- eval_wer_ortho: 0.2832
-- eval_wer: 0.2358
-- eval_cer: 0.1039
-- eval_precision: 0.8574
-- eval_Recall: 0.8573
-- eval_F1: 0.8565
-- eval_runtime: 167.5223
-- eval_samples_per_second: 3.779
-- eval_steps_per_second: 0.478
-- epoch: 0.7958
-- step: 300
+- Loss: 0.7341
+- Wer Ortho: 0.3002
+- Wer: 0.2495
+- Cer: 0.1214
+- Precision: 0.8529
+- Recall: 0.8536
+- F1: 0.8523
 
 ## Model description
 
@@ -58,6 +58,17 @@ The following hyperparameters were used during training:
 - training_steps: 500
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Wer Ortho | Wer    | Cer    | Precision | Recall | F1     |
+|:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:---------:|:------:|:------:|
+| 0.9512        | 0.3344 | 100  | 0.9677          | 0.3278    | 0.2780 | 0.1320 | 0.8298    | 0.8355 | 0.8319 |
+| 0.7473        | 0.6689 | 200  | 0.8168          | 0.3054    | 0.2477 | 0.1169 | 0.8454    | 0.8479 | 0.8459 |
+| 0.7431        | 1.0033 | 300  | 0.7681          | 0.2922    | 0.2427 | 0.1247 | 0.8537    | 0.8507 | 0.8512 |
+| 0.5364        | 1.3378 | 400  | 0.7507          | 0.2946    | 0.2470 | 0.1240 | 0.8490    | 0.8484 | 0.8476 |
+| 0.5233        | 1.6722 | 500  | 0.7341          | 0.3002    | 0.2495 | 0.1214 | 0.8529    | 0.8536 | 0.8523 |
+
+
 ### Framework versions
 
 - Transformers 4.51.3

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6adb3ba54566bd72217130179ebc78735c0311721401b5ae441c6cb20cae2dae
+oid sha256:21b9b5d520ce8f7cdff0b68c1986323e32c1f04280595f48a9617a6fb6c86fc3
 size 290403936