End of training
Browse files
README.md
CHANGED
|
@@ -8,6 +8,8 @@ tags:
|
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
- mozilla-foundation/common_voice_16_0
|
|
|
|
|
|
|
| 11 |
model-index:
|
| 12 |
- name: Whisper Small Persian
|
| 13 |
results: []
|
|
@@ -19,6 +21,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
# Whisper Small Persian
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [makhataei/Whisper-Small-Common-Voice](https://huggingface.co/makhataei/Whisper-Small-Common-Voice) on the Common Voice 16.0 dataset.
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -37,7 +42,7 @@ More information needed
|
|
| 37 |
### Training hyperparameters
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
-
- learning_rate: 1e-
|
| 41 |
- train_batch_size: 10
|
| 42 |
- eval_batch_size: 10
|
| 43 |
- seed: 42
|
|
@@ -45,8 +50,24 @@ The following hyperparameters were used during training:
|
|
| 45 |
- total_train_batch_size: 40
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
-
- lr_scheduler_warmup_steps:
|
| 49 |
-
- training_steps:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
|
| 51 |
### Framework versions
|
| 52 |
|
|
|
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
- mozilla-foundation/common_voice_16_0
|
| 11 |
+
metrics:
|
| 12 |
+
- wer
|
| 13 |
model-index:
|
| 14 |
- name: Whisper Small Persian
|
| 15 |
results: []
|
|
|
|
| 21 |
# Whisper Small Persian
|
| 22 |
|
| 23 |
This model is a fine-tuned version of [makhataei/Whisper-Small-Common-Voice](https://huggingface.co/makhataei/Whisper-Small-Common-Voice) on the Common Voice 16.0 dataset.
|
| 24 |
+
It achieves the following results on the evaluation set:
|
| 25 |
+
- Loss: 0.8716
|
| 26 |
+
- Wer: 52.4832
|
| 27 |
|
| 28 |
## Model description
|
| 29 |
|
|
|
|
| 42 |
### Training hyperparameters
|
| 43 |
|
| 44 |
The following hyperparameters were used during training:
|
| 45 |
+
- learning_rate: 1e-08
|
| 46 |
- train_batch_size: 10
|
| 47 |
- eval_batch_size: 10
|
| 48 |
- seed: 42
|
|
|
|
| 50 |
- total_train_batch_size: 40
|
| 51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
- lr_scheduler_type: linear
|
| 53 |
+
- lr_scheduler_warmup_steps: 50
|
| 54 |
+
- training_steps: 1000
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
| 60 |
+
| 0.0002 | 0.16 | 100 | 0.8707 | 51.8989 |
|
| 61 |
+
| 0.0002 | 0.31 | 200 | 0.8713 | 52.1447 |
|
| 62 |
+
| 0.0002 | 0.47 | 300 | 0.8716 | 52.1377 |
|
| 63 |
+
| 0.0002 | 0.62 | 400 | 0.8717 | 52.2907 |
|
| 64 |
+
| 0.0015 | 0.78 | 500 | 0.8718 | 52.4762 |
|
| 65 |
+
| 0.0002 | 0.93 | 600 | 0.8716 | 52.4878 |
|
| 66 |
+
| 0.0002 | 1.09 | 700 | 0.8716 | 52.4901 |
|
| 67 |
+
| 0.0001 | 1.25 | 800 | 0.8716 | 52.4832 |
|
| 68 |
+
| 0.0002 | 1.4 | 900 | 0.8716 | 52.4832 |
|
| 69 |
+
| 0.0002 | 1.56 | 1000 | 0.8716 | 52.4832 |
|
| 70 |
+
|
| 71 |
|
| 72 |
### Framework versions
|
| 73 |
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 966995080
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a9bef7d4de4a11eb5bbc03aeee592ac8eb8cb8438c4881bc88945b53b1b2880
|
| 3 |
size 966995080
|
runs/Jan01_17-49-01_Software-AI/events.out.tfevents.1704118744.Software-AI.385155.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2029f91a240cdaeb81e6c3c7b2aec539d40daab8d33dc57d599954b25bec8c41
|
| 3 |
+
size 14875
|