Hugging Face's logo

arsalanu
/
bert-base-uncased-squad

Question Answering
Transformers
PyTorch
Graphcore
bert
Generated from Trainer
Model card Files
xet
 Community
2
arsalanu commited on
Commit
c8f3f5c
路
1 Parent(s): 9890283

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -32,16 +32,16 @@ More information needed
32
 
33
  The following hyperparameters were used during training:
34
  - learning_rate: 9e-05
35
- - train_batch_size: 1
36
  - eval_batch_size: 8
37
  - seed: 42
38
  - distributed_type: IPU
39
  - gradient_accumulation_steps: 32
40
- - total_train_batch_size: 512
41
  - total_eval_batch_size: 512
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: cosine
44
- - lr_scheduler_warmup_ratio: 0.1
45
  - num_epochs: 3
46
  - training precision: Mixed Precision
47
 
 
32
 
33
  The following hyperparameters were used during training:
34
  - learning_rate: 9e-05
35
+ - train_batch_size: 2
36
  - eval_batch_size: 8
37
  - seed: 42
38
  - distributed_type: IPU
39
  - gradient_accumulation_steps: 32
40
+ - total_train_batch_size: 2048
41
  - total_eval_batch_size: 512
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: cosine
44
+ - lr_scheduler_warmup_ratio: 0.25
45
  - num_epochs: 3
46
  - training precision: Mixed Precision
47