Hugging Face's logo

voidful
/
ssr-360m

Transformers
Safetensors
mmlmasr
Generated from Trainer
Model card Files Community
voidful commited on
Commit
03c7e71
·
verified ·
1 Parent(s): bfc84f7

Model save

Browse files
Files changed (1) hide show
  1. README.md +20 -18
README.md CHANGED
@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.1612
20
- - Model Preparation Time: 0.0058
21
- - Wer: 0.4343
22
 
23
  ## Model description
24
 
@@ -37,28 +37,30 @@ More information needed
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
- - learning_rate: 5e-05
41
- - train_batch_size: 10
42
- - eval_batch_size: 10
43
  - seed: 42
 
 
44
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
45
  - lr_scheduler_type: linear
46
  - num_epochs: 10
47
 
48
  ### Training results
49
 
50
- | Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Wer |
51
- |:-------------:|:-----:|:------:|:---------------:|:----------------------:|:------:|
52
- | 1.9429 | 1.0 | 10402 | 0.1859 | 0.0058 | 0.5157 |
53
- | 1.3785 | 2.0 | 20804 | -0.0288 | 0.0058 | 0.4419 |
54
- | 1.5886 | 3.0 | 31206 | -0.1098 | 0.0058 | 0.4158 |
55
- | 1.2236 | 4.0 | 41608 | -0.1546 | 0.0058 | 0.4035 |
56
- | 0.9884 | 5.0 | 52010 | -0.0799 | 0.0058 | 0.4199 |
57
- | 0.8967 | 6.0 | 62412 | -0.0511 | 0.0058 | 0.4184 |
58
- | 0.7347 | 7.0 | 72814 | 0.0044 | 0.0058 | 0.4228 |
59
- | 0.5243 | 8.0 | 83216 | 0.0490 | 0.0058 | 0.4245 |
60
- | 0.5616 | 9.0 | 93618 | 0.1234 | 0.0058 | 0.4321 |
61
- | 0.3841 | 10.0 | 104020 | 0.1612 | 0.0058 | 0.4343 |
62
 
63
 
64
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.0227
20
+ - Model Preparation Time: 0.0066
21
+ - Wer: 0.1206
22
 
23
  ## Model description
24
 
 
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
+ - learning_rate: 0.0005
41
+ - train_batch_size: 2
42
+ - eval_batch_size: 2
43
  - seed: 42
44
+ - gradient_accumulation_steps: 10
45
+ - total_train_batch_size: 20
46
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
47
  - lr_scheduler_type: linear
48
  - num_epochs: 10
49
 
50
  ### Training results
51
 
52
+ | Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Wer |
53
+ |:-------------:|:------:|:-----:|:---------------:|:----------------------:|:------:|
54
+ | 12.0652 | 0.9999 | 5200 | 0.5135 | 0.0066 | 0.4332 |
55
+ | 2.1537 | 1.9999 | 10400 | 0.0530 | 0.0066 | 0.1585 |
56
+ | 3.0568 | 2.9999 | 15600 | 0.0731 | 0.0066 | 0.2030 |
57
+ | 0.9142 | 3.9999 | 20800 | 0.0313 | 0.0066 | 0.1407 |
58
+ | 0.3436 | 4.9999 | 26000 | 0.0308 | 0.0066 | 0.1403 |
59
+ | 0.7558 | 5.9999 | 31200 | 0.0286 | 0.0066 | 0.1337 |
60
+ | 0.5733 | 6.9999 | 36400 | 0.0265 | 0.0066 | 0.1221 |
61
+ | 0.1573 | 7.9999 | 41600 | 0.0241 | 0.0066 | 0.1439 |
62
+ | 0.2248 | 8.9999 | 46800 | 0.0225 | 0.0066 | 0.1199 |
63
+ | 0.1501 | 9.9999 | 52000 | 0.0227 | 0.0066 | 0.1206 |
64
 
65
 
66
  ### Framework versions