End of training
Browse files- README.md +9 -24
- generation_config.json +2 -1
README.md
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
datasets:
|
|
@@ -14,8 +15,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
# speecht5_finetuned_kha
|
| 15 |
|
| 16 |
This model was trained from scratch on the audiofolder dataset.
|
| 17 |
-
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.4932
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -34,39 +33,25 @@ More information needed
|
|
| 34 |
### Training hyperparameters
|
| 35 |
|
| 36 |
The following hyperparameters were used during training:
|
| 37 |
-
- learning_rate:
|
| 38 |
-
- train_batch_size:
|
| 39 |
- eval_batch_size: 2
|
| 40 |
- seed: 42
|
| 41 |
-
- gradient_accumulation_steps:
|
| 42 |
-
- total_train_batch_size:
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
- lr_scheduler_warmup_steps: 500
|
| 46 |
-
- num_epochs:
|
| 47 |
- mixed_precision_training: Native AMP
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
-
|:-------------:|:--------:|:-----:|:---------------:|
|
| 53 |
-
| 0.4596 | 78.4314 | 1000 | 0.4390 |
|
| 54 |
-
| 0.4352 | 156.8627 | 2000 | 0.4409 |
|
| 55 |
-
| 0.4182 | 235.2941 | 3000 | 0.4419 |
|
| 56 |
-
| 0.4086 | 313.7255 | 4000 | 0.4521 |
|
| 57 |
-
| 0.4013 | 392.1569 | 5000 | 0.4637 |
|
| 58 |
-
| 0.3963 | 470.5882 | 6000 | 0.4660 |
|
| 59 |
-
| 0.3896 | 549.0196 | 7000 | 0.4761 |
|
| 60 |
-
| 0.3866 | 627.4510 | 8000 | 0.4869 |
|
| 61 |
-
| 0.385 | 705.8824 | 9000 | 0.4891 |
|
| 62 |
-
| 0.3806 | 784.3137 | 10000 | 0.4845 |
|
| 63 |
-
| 0.3827 | 862.7451 | 11000 | 0.4887 |
|
| 64 |
-
| 0.3788 | 941.1765 | 12000 | 0.4932 |
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|
| 68 |
|
| 69 |
-
- Transformers 4.
|
| 70 |
-
- Pytorch 2.
|
| 71 |
-
- Datasets
|
| 72 |
- Tokenizers 0.19.1
|
|
|
|
| 1 |
---
|
| 2 |
+
library_name: transformers
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
| 5 |
datasets:
|
|
|
|
| 15 |
# speecht5_finetuned_kha
|
| 16 |
|
| 17 |
This model was trained from scratch on the audiofolder dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 33 |
### Training hyperparameters
|
| 34 |
|
| 35 |
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 2e-05
|
| 37 |
+
- train_batch_size: 32
|
| 38 |
- eval_batch_size: 2
|
| 39 |
- seed: 42
|
| 40 |
+
- gradient_accumulation_steps: 64
|
| 41 |
+
- total_train_batch_size: 2048
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
- lr_scheduler_warmup_steps: 500
|
| 45 |
+
- num_epochs: 100
|
| 46 |
- mixed_precision_training: Native AMP
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
|
| 51 |
|
| 52 |
### Framework versions
|
| 53 |
|
| 54 |
+
- Transformers 4.44.2
|
| 55 |
+
- Pytorch 2.4.1+cu121
|
| 56 |
+
- Datasets 3.0.0
|
| 57 |
- Tokenizers 0.19.1
|
generation_config.json
CHANGED
|
@@ -5,5 +5,6 @@
|
|
| 5 |
"eos_token_id": 2,
|
| 6 |
"max_length": 1876,
|
| 7 |
"pad_token_id": 1,
|
| 8 |
-
"transformers_version": "4.
|
|
|
|
| 9 |
}
|
|
|
|
| 5 |
"eos_token_id": 2,
|
| 6 |
"max_length": 1876,
|
| 7 |
"pad_token_id": 1,
|
| 8 |
+
"transformers_version": "4.44.2",
|
| 9 |
+
"use_cache": false
|
| 10 |
}
|