PremalMatalia commited on
Commit
8b3a927
1 Parent(s): 5b797ee

Updated Hyperparameters

Browse files
Files changed (1) hide show
  1. README.md +8 -5
README.md CHANGED
@@ -22,20 +22,23 @@ Tensorflow version (GPU?): 2.5.0 (False)</br>
22
 
23
  ## Hyperparameters
24
  ```
 
 
 
 
 
25
  batch_size=8
 
26
  n_epochs=6
27
  base_LM_model = "deepset/roberta-base-squad2"
28
- max_seq_len=386
29
- doc_stride=128
30
  learning_rate=1.5e-5
31
  adam_epsilon=1e-5
32
  adam_beta1=0.95
33
  adam_beta2=0.999
34
  warmup_steps=100
35
  weight_decay=0.01
36
- n_best_size=20
37
- max_answer_length=30
38
- min_null_score=7.0
39
  ```
40
  ##### There is a special threshold value CLS_threshold=-3 used to more accurately identify no answers [Logic will be available in GitHub Repo [TBD]
41
 
22
 
23
  ## Hyperparameters
24
  ```
25
+ max_seq_len=386
26
+ doc_stride=128
27
+ n_best_size=20
28
+ max_answer_length=30
29
+ min_null_score=7.0
30
  batch_size=8
31
+
32
  n_epochs=6
33
  base_LM_model = "deepset/roberta-base-squad2"
 
 
34
  learning_rate=1.5e-5
35
  adam_epsilon=1e-5
36
  adam_beta1=0.95
37
  adam_beta2=0.999
38
  warmup_steps=100
39
  weight_decay=0.01
40
+ optimizer=AdamW
41
+ lr_scheduler="polynomial"
 
42
  ```
43
  ##### There is a special threshold value CLS_threshold=-3 used to more accurately identify no answers [Logic will be available in GitHub Repo [TBD]
44