Hey guys. I wanted to enquire about the best practices

Question

Hey guys. I wanted to enquire about the best practices

for analysing data that has a lot of ID columns. These are numerical data points but serve more of a categorical value if that makes sense? In a lot of preparing dataset people have simply dropped the ID columns. The ID themselves are randomly generated however the thing they denote might have a seasionality based on that ID column for example a table with a lot of product details and only a few numerical columns and/or categorical columns.

0

23.05.2019

2 ответов

34 просмотра

Robert

Change the type of those columns to "object" or "category". Then, pandas will treat the columns as strings or factors, respectively. For instance df["id_column"] = df["id_column"].astype("category")

0

07.06.2019

Helen Helen · Accepted Answer

Helen Helen

try pandas.describe method

0

04.06.2019

Похожие чаты

Hey guys. I wanted to enquire about the best practices

2 ответов

Похожие вопросы