Pandas Dataframe 样本的补码

Question

import pandas as pd

df = pd.read_csv("train.csv")

sample = df.sample(10)

sample.to_csv("train_subset.csv")

I want to sample 10 random rows from a given csv file (train.csv) and store it as a new csv file train_subset.csv. The code above achieves that. Now I also want to store all the rows that weren't sampled into a file train_remaining.csv.

How can I implement that? How do I find which rows were sampled?

Answer 1

You can use

df.index.difference(sample.index)

where sample.index is the sected sample index.

And then use it for select the complementary dataframe:

complementary = df.iloc[df.index.difference(sample.index)]

Answer 2

I would suggest using sklearns train_test_split.

http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html

This will allow you to take a percentage of the rows that are randomly selected.