How to I train a model for a certain number of rows in python?

Question

I am creating a model and have 1876 rows. I would like to use 1000 of them for training and 876 of them for testing. I am unsure how to do this, currently, I have the following:

df_train, df_test = train_test_split(df, train_size=0.80, test_size=0.20, shuffle=False)

This is by the percentage which is not the goal currently but it is all I know how to do. Does anyone have suggestions of how I could change this code to get exactly 1000 training and 876 testing rows? I understand that this is likely a sub-optimal split. Thank you in advance!

Answer 1

You can use

df_train = df.iloc[:1000]
df_test = df.iloc[1000:]

Answer 2

As @blorgon wrote you can use integers instead of percentages but also Pandas has the sample function, which samples n items randomly (so you can choose n=1000) here is the documentation: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.sample.html

Answer 3

test_size=0.20 means 20% of 1876 will result in the test data. use test_size=.46

 1876*.46=1876-862=1014 (test size)