mageec commited on
Commit
2068eeb
·
verified ·
1 Parent(s): 5f0a0c4

updated epoch 3

Browse files
Files changed (1) hide show
  1. README.md +6 -8
README.md CHANGED
@@ -37,8 +37,6 @@ language:
37
  - ja
38
  pipeline_tag: audio-classification
39
  ---
40
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
41
- should probably proofread and complete it, then remove this comment. -->
42
 
43
  # WAV2VEC2_CAPSTONE_MODEL
44
 
@@ -57,22 +55,22 @@ More information needed
57
  ### Training hyperparameters
58
  The following hyperparameters were used during training:
59
  - learning_rate: 0.0003
60
- - train_batch_size: 7
61
- - eval_batch_size: 7
62
  - seed: 42
63
- - gradient_accumulation_steps: 11
64
- - total_train_batch_size: 77
65
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
66
  - lr_scheduler_type: linear
67
  - lr_scheduler_warmup_ratio: 0.1
68
- - num_epochs: 7
69
  - mixed_precision_training: Native AMP
70
  ### Training results
71
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 score |
72
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|
73
  | 0.9339 | 1.0 | 776 | 1.4214 | 0.7162 | 0.7094 |
74
  | 0.5663 | 2.0 | 1552 | 1.0182 | 0.8318 | 0.8277 |
75
- | 0.6128 | 3.0 | 1541 | 0.6117 | 0.8205 | 0.8203 |
76
  | 0.4882 | 4.0 | 2055 | 0.5092 | 0.8474 | 0.8478 |
77
  | 0.2305 | 5.0 | 2569 | 0.5244 | 0.8643 | 0.8639 |
78
  | 0.1219 | 6.0 | 3082 | 0.4169 | 0.8966 | 0.8966 |
 
37
  - ja
38
  pipeline_tag: audio-classification
39
  ---
 
 
40
 
41
  # WAV2VEC2_CAPSTONE_MODEL
42
 
 
55
  ### Training hyperparameters
56
  The following hyperparameters were used during training:
57
  - learning_rate: 0.0003
58
+ - train_batch_size: 8
59
+ - eval_batch_size: 8
60
  - seed: 42
61
+ - gradient_accumulation_steps: 10
62
+ - total_train_batch_size: 80
63
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
64
  - lr_scheduler_type: linear
65
  - lr_scheduler_warmup_ratio: 0.1
66
+ - num_epochs: 8
67
  - mixed_precision_training: Native AMP
68
  ### Training results
69
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 score |
70
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|
71
  | 0.9339 | 1.0 | 776 | 1.4214 | 0.7162 | 0.7094 |
72
  | 0.5663 | 2.0 | 1552 | 1.0182 | 0.8318 | 0.8277 |
73
+ | 0.4408 | 3.0 | 2328 | 0.6117 | 0.8795 | 0.8784 |
74
  | 0.4882 | 4.0 | 2055 | 0.5092 | 0.8474 | 0.8478 |
75
  | 0.2305 | 5.0 | 2569 | 0.5244 | 0.8643 | 0.8639 |
76
  | 0.1219 | 6.0 | 3082 | 0.4169 | 0.8966 | 0.8966 |