How to find location of already downloaded datasets with Scikit Learn on Linux?

Question

I use Debian Linux and I'm trying to find where my already downloaded datasets were stored.

On /home/usuario/scikit_learn_data, there insn't any file. I thougth all downloaded datasets would be stored here (on download, I didn't specified any custom location).

There's my code to download datasets:

import os
from sklearn import datasets

current_directory = os.getcwd()
print(current_directory)

iris = datasets.load_iris()
diabetes = datasets.load_diabetes()

datasets.get_data_home() 
print(iris)

AS you can see, I downloaded two datasets. But, as I said before, there's isn't any file on /home/usuario/scikit_learn_data. So, Where are the downloaded data files???

I am trying to figure out where are my downloaded datasets files. I have little memory (this is a business requirement) on the computer I am working with, so I have to manage to delete these files. But I just can not find them!

Answer 1

I think some datasets like iris are installed along with the sklearn package. I was able to find them in the following location:

~/.local/lib/python3.10/site-packages/sklearn/datasets/data/