Last answered:

25 Feb 2023

Posted on:

25 Aug 2021

1

Module4. Credit Risk Modeling. Section 6. Could you post the 4 test and train csv files.

Module4. Credit Risk Modeling. Section 6. Could you post the 4 test and train csv files.



My loan_data_inputs_train.shape
(373028, 283)
Unlike the instructors 324 features.
I am not able to fix the problem. I am not able to proceed with section 6 as the dataframes for "the error at my end" are a complete mismatch.
I have tried running the code provided at the end of each section and ended with a similar discrepancy.
Could anybody could just share the 4 test and train (inputs and targets) csv files from end of Section 5.
Appreciate it.

Thank you and Regards.






KeyError: "Passing list-likes to .loc or [] with any missing labels is no longer supported. The following labels were missing: Index(['home_ownership:RENT_OTHER_NONE_ANY', 'addr_state:ND_NE_IA_NV_FL_HI_AL',\n       'addr_state:NM_VA', 'addr_state:OK_TN_MO_LA_MD_NC',\n
7 answers ( 0 marked as helpful)
Posted on:

01 Nov 2021

2

I'm having a very similar problem and I'm unable to proceed with following along within this section.

Posted on:

09 Nov 2021

2

I am also facing the same issue

Posted on:

03 Nov 2022

0

Same issue here, is this course still alive? I pay a lot for this service compared to other courses.

Posted on:

11 Nov 2022

0

One way to solve this error is apply the preprocessing steps to your testing and training data together. What I mean by that is apply the same steps twice one for training data and one for testing data.

Posted on:

15 Nov 2022

0

Yes, I already tried that but that unfortunately also gives errors with one of the files.

Posted on:

17 Nov 2022

0

Any update on this issue? Could the .csv files be provided?

Posted on:

25 Feb 2023

0

My solution to this problem :

1. put max_iter=10000 in LogisticRegression:
    reg = LogisticRegression(max_iter=10000)
2. put "fillna(0)" inside the fit:
    reg.fit(inputs_train.fillna(0), loan_data_targets_train)

The result is very similar

Submit an answer