[bug] pearson_r p-value #81

New issue

Closed

opened 2025-10-14 17:44:42 -06:00 by navan · 0 comments

navan commented

2025-10-14 17:44:42 -06:00

Owner

Originally created by @albertma-evotec on 4/13/2021

In your DBTA.train() function in DTI.py file

     if verbose:
         print('Validation at Epoch '+ str(epo + 1) + ' with loss:' + str(loss_val.item())[:7] +', MSE: ' + str(mse)[:7] + ' , Pearson Correlation: '
	+ str(r2)[:7] + ' with p-value: ' + str(p_val)[:7] +' , Concordance Index: '+str(CI)[:7])

the p-value is often a very small number so that it is represented as the scientific e notation, i.e like 1.23456789e-9
str() it then [:7] will miss the e number part and lead to a wrong result.

str(1.23456789e-9)[:7] -> 1.23456 not what you actually wanted!

*Originally created by @albertma-evotec on 4/13/2021* In your DBTA.train() function in DTI.py file ``` if verbose: print('Validation at Epoch '+ str(epo + 1) + ' with loss:' + str(loss_val.item())[:7] +', MSE: ' + str(mse)[:7] + ' , Pearson Correlation: ' + str(r2)[:7] + ' with p-value: ' + str(p_val)[:7] +' , Concordance Index: '+str(CI)[:7]) ``` the p-value is often a very small number so that it is represented as the scientific e notation, i.e like 1.23456789e-9 str() it then [:7] will miss the e number part and lead to a wrong result. str(1.23456789e-9)[:7] -> 1.23456 not what you actually wanted!

navan

2025-10-14 17:44:43 -06:00

closed this issue
added the
bug
label

No labels

bug

enhancement

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: github/DeepPurpose#81

No description provided.

Rows
Columns