Commit
•
48a95c0
1
Parent(s):
5bfda33
End of training
Browse files- README.md +21 -5
- generation_config.json +1 -1
README.md
CHANGED
|
@@ -2,9 +2,11 @@
|
|
| 2 |
language:
|
| 3 |
- hi
|
| 4 |
license: apache-2.0
|
| 5 |
-
base_model:
|
| 6 |
tags:
|
| 7 |
- generated_from_trainer
|
|
|
|
|
|
|
| 8 |
model-index:
|
| 9 |
- name: base
|
| 10 |
results: []
|
|
@@ -15,7 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# base
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -35,8 +40,8 @@ More information needed
|
|
| 35 |
|
| 36 |
The following hyperparameters were used during training:
|
| 37 |
- learning_rate: 1e-05
|
| 38 |
-
- train_batch_size:
|
| 39 |
-
- eval_batch_size:
|
| 40 |
- seed: 42
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
|
@@ -44,9 +49,20 @@ The following hyperparameters were used during training:
|
|
| 44 |
- training_steps: 5000
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
### Framework versions
|
| 48 |
|
| 49 |
- Transformers 4.35.2
|
| 50 |
- Pytorch 2.1.0+cu121
|
| 51 |
-
- Datasets 2.16.
|
| 52 |
- Tokenizers 0.15.0
|
|
|
|
| 2 |
language:
|
| 3 |
- hi
|
| 4 |
license: apache-2.0
|
| 5 |
+
base_model: reproductionguru/voicetest6
|
| 6 |
tags:
|
| 7 |
- generated_from_trainer
|
| 8 |
+
metrics:
|
| 9 |
+
- wer
|
| 10 |
model-index:
|
| 11 |
- name: base
|
| 12 |
results: []
|
|
|
|
| 17 |
|
| 18 |
# base
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [reproductionguru/voicetest6](https://huggingface.co/reproductionguru/voicetest6) on the tutorial Voice 11.0 dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.4757
|
| 23 |
+
- Wer: 21.0355
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
| 42 |
- learning_rate: 1e-05
|
| 43 |
+
- train_batch_size: 8
|
| 44 |
+
- eval_batch_size: 8
|
| 45 |
- seed: 42
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
|
|
|
| 49 |
- training_steps: 5000
|
| 50 |
- mixed_precision_training: Native AMP
|
| 51 |
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
| 56 |
+
| 0.4953 | 0.4 | 1000 | 0.6332 | 30.4347 |
|
| 57 |
+
| 0.408 | 0.8 | 2000 | 0.5400 | 25.3584 |
|
| 58 |
+
| 0.2074 | 1.2 | 3000 | 0.5097 | 23.7629 |
|
| 59 |
+
| 0.1796 | 1.61 | 4000 | 0.4885 | 21.6339 |
|
| 60 |
+
| 0.1583 | 2.01 | 5000 | 0.4757 | 21.0355 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
### Framework versions
|
| 64 |
|
| 65 |
- Transformers 4.35.2
|
| 66 |
- Pytorch 2.1.0+cu121
|
| 67 |
+
- Datasets 2.16.1
|
| 68 |
- Tokenizers 0.15.0
|
generation_config.json
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"_from_model_config": true,
|
| 3 |
"begin_suppress_tokens": [
|
| 4 |
220,
|
| 5 |
50257
|
|
@@ -9,5 +8,6 @@
|
|
| 9 |
"eos_token_id": 50257,
|
| 10 |
"max_length": 448,
|
| 11 |
"pad_token_id": 50257,
|
|
|
|
| 12 |
"transformers_version": "4.35.2"
|
| 13 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"begin_suppress_tokens": [
|
| 3 |
220,
|
| 4 |
50257
|
|
|
|
| 8 |
"eos_token_id": 50257,
|
| 9 |
"max_length": 448,
|
| 10 |
"pad_token_id": 50257,
|
| 11 |
+
"return_timestamps": false,
|
| 12 |
"transformers_version": "4.35.2"
|
| 13 |
}
|