Error got when score python model in MM

[wuyb28]

Describe the issue A clear and concise description of the issue you're experiencing.

To Reproduce using sastcl sample to register pymodel to monthly env, and found all three dmcas_*.json files have the same size. https://github.com/sassoftware/python-sasctl/blob/master/examples/pzmm_binary_classification_model_import.ipynb

Go to SAS Model Manager. Upload the hmeq.csv and score

Expected behavior The task should score a table without error.

Stack Trace If you're experiencing an exception, include the full stack trace and error message.

Version What version of sasctl are you using? 1.9.2

[wuyb28]

I changed the sample code and replace DecisionTree to logistic regression and got soring success. So it related to algorithm?

[wuyb28]

More detail python code, score log & model zip file. py&logs.zip LogisticRegression.zip RandomForest.zip

[smlindauer]

The fitstat, roc, and lift charts being the same is noted in #167 and fixed in 2e1fa34 (unreleased as of yet).

[smlindauer]

You may be running in to a version mismatch between your development environment and the Viya environment. Can you check the versions of scikit-learn and pandas you are using on each?

[wuyb28]

I'm using latest anaconda.

And my viya env is our monthly env https://ssemonthly.demo.sas.com/SASLogon/timedout?callbackUrl=/SASModelManager/.

Could you tell me which anaconda , scikit-learn and pandas version is suitable or is tested? Thank you!

[wuyb28]

I change my anaconda with Anaconda3-2020.07-Windows-x86_64.exe with python 3.8.3, and got success! Previous one is python 3.10. Thanks for your advise!@smlindauer

[wuyb28]

Now I'm using the updated sasctl src and got different size of json files but some model's lift chart may not be correct.