columns = ['Accelerations Zone 5','Decelerations Zone 5','Max Speed','Sprints','HMLD'] # Need to check which more columns we need to add?

for column in columns: for i in [7,14,21,28]: df_new = extract_records_based_on_days(data,data_match_day,'Session Date',column,i) train_match_compare(df_new,data_match_day,'Session Date',column,i) v_1,v_2,v_3,v_4 = cacl_mean_sd(df_new[column],data_match_day[column]) visualize_distributions(df_new[column],data_match_day[column],two_sample_data_setup_for_cohens_d(v_1,v_2,v_3,v_4),interpret_cohens_d(two_sample_data_setup_for_cohens_d(v_1,v_2,v_3,v_4)))

The above function produces Cohen's d values for 7 days, 14days, 21days, and 28days of training vs match record. The function produces a graph.

I would like you to create a function that will create a data frame for those columns and recording the Cohen's d values for each column for 7, 14, 21, and 28days each.

Thank you!