-
-
Save karamanbk/8af50168240621516e5722e4196d1533 to your computer and use it in GitHub Desktop.
This is very much detailed and very informative.
Thank you for the great code and explanation, I would be really nice if we could have explore more of the ways that are possible to increase the model accuracy. As my models are not that accurate unfortunately.
Just one question, how can a 'Customer ID', which is a actually a categorical data, part of the training dataset? If 'Customer ID' is '12747.0', it does not make any sense in the training data, as it could be any other number like '435666666666' or 'ABCD' or '536TGK5'.
Now, if you remove the 'Customer ID' from training, how will you test on the test dataset by predicting which of the customers will buy in the next week or so?
Thank you for the great code and explanation, I would be really nice if we could have explore more of the ways that are possible to increase the model accuracy. As my models are not that accurate unfortunately.
There's probably an overfitting problem as the accuracy for the test set is way lower than that of the training set.
Accuracy of XGB classifier on training set: 0.92
Accuracy of XGB classifier on test set: 0.62
Awesome