How to do the ADF Test for each column of my Dataframe in Python?

Not sure about your data in particular but you can use the statsmodels library for that. Here is the docs about the ADF test : https://www.statsmodels.org/stable/generated/statsmodels.tsa.stattools.adfuller.html

This is a function you can apply to columns of DataFrame:

from statsmodels.tsa.stattools import adfuller
import pandas as pd


def adfuller_test(series, signif=0.05):
"""
Perform Augmented Dickey-Fuller to test for Stationarity of the given series
and print report. Null Hypothesis: Data has unit root and is non-stationary.

series: time series in pd.Series format
signif: significance level for P-value to reject Null Hypothesis
"""
    x = adfuller(series, autolag='AIC')

    #using dictionary saves different data types (float, int, boolean)
    output = {'Test Statistic': x[0], 
              'P-value': x[1], 
              'Number of lags': x[2], 
              'Number of observations': x[3],
              f'Reject (signif. level {signif})': x[1] < signif }

    for key, val in x[4].items():
         output[f'Critical value {key}'] = val

return pd.Series(output)

You can follow like df_train.apply(lambda x: adfuller_test(x), axis=0) to get a summary.