kejian commited on
Commit
30d1ded
1 Parent(s): 6e5f7da

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -3
README.md CHANGED
@@ -71,9 +71,11 @@ More information needed
71
 
72
  The following hyperparameters were used during training:
73
  - learning_rate: 0.0007
74
- - train_batch_size: 64
75
- - eval_batch_size: 32
76
  - seed: 42
 
 
77
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
78
  - lr_scheduler_type: linear
79
  - lr_scheduler_warmup_ratio: 0.01
@@ -177,4 +179,4 @@ The following hyperparameters were used during training:
177
  'weight_decay': 0.1}}
178
 
179
  # Wandb URL:
180
- https://wandb.ai/kejian/uncategorized/runs/3a3fq724
 
71
 
72
  The following hyperparameters were used during training:
73
  - learning_rate: 0.0007
74
+ - train_batch_size: 32
75
+ - eval_batch_size: 16
76
  - seed: 42
77
+ - gradient_accumulation_steps: 2
78
+ - total_train_batch_size: 64
79
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
80
  - lr_scheduler_type: linear
81
  - lr_scheduler_warmup_ratio: 0.01
 
179
  'weight_decay': 0.1}}
180
 
181
  # Wandb URL:
182
+ https://wandb.ai/kejian/uncategorized/runs/1efu1obk