Kaggle grocery competition

The issue is RAM, not hard drive space.

t2.medium has 4 GB RAM. You’re probably trying to load the 5 GB train.csv file all at once into a dataframe. Since it cannot fit in RAM, it might be causing a crash.

Read the entire thread linked below - solutions to this problem are posted here:


Part of becoming a programmer is learning to say something other than “it doesn’t work”, which is what you are saying when you say “job gets killed instantly.” You need to be specific as in “I get the following error when I run this specific program.”

I will surely take care of this in future.

I am using random forest to practice what is taught in class using Grocery dataset. I was wondering weather predictions coming from random forest are bounded between minimum and maximum of unit_sales per category feed in to the model. So if there is a increasing/decreasing trend in unit_sales I doubt that random forest will be able to capture that in the predictions.

Well spotted - yes they are bounded and it’s a major issue to consider. Although since in this case it’s only 2 weeks ahead to predict, I doubt it’s a problem here

1 Like