Importing JSONencoder solved the AttributeError, but now i got a new one. đź™‚

It says “TypeError : __init__() takes 1 positional argument but 2 were given

o = default(o) ” in _interencode from encode.py file. ]]>

The cnn_models are in the cnn_models directory on the GitHub page. You can try to copy the code directly into the notebook, or clone the entire repository including the cnn_models directory.

Please let me know if this fixes your problem.

]]>i am stuck in the Reading the shapefile code. I got the AttributeError: module ‘simplejson’ has no attribute ‘JSONencoder’.

Any help please. ]]>

The correlation matrix was actually made on the glass dataset, which does not contain any categorical data. But I did use this opportunity to update the code and explicitly mention the dangers of one-hot encoding categorical data, and also added a few additional sections. ]]>

Thanks for the feedback! I’ll update it as soon as I have time. ]]>

thanks for your examples. We used it for educational purposes, however please revisit the section where you transform non-numerical values to numerical values.

If I understand you correctly, you run correlations on numerical values which do not have any ‘numerical meaning’ but are strictly categorical (‘glass type’); the results (other than ‘ 0’) therefore are arbitrary and strongly depend only on the ordering of the types. ]]>