orhermansaffar

def chooseBestKforKMeans(scaled_data, k_range):
    ans = []
    for k in k_range:
        scaled_inertia = kMeansRes(scaled_data, k)
        ans.append((k, scaled_inertia))
    results = pd.DataFrame(ans, columns = ['k','Scaled Inertia']).set_index('k')
    best_k = results.idxmin()[0]
    return best_k, results

orhermansaffar

# How to use TimeBasedCV
data_for_modeling=pd.read_csv('data.csv', parse_dates=['record_date'])
tscv = TimeBasedCV(train_period=30,
                   test_period=7,
                   freq='days')
for train_index, test_index in tscv.split(data_for_modeling,
                   validation_split_date=datetime.date(2019,2,1), date_column='record_date'):
    print(train_index, test_index)

# get number of splits

orhermansaffar

import pandas as pd
import datetime
from datetime import datetime as dt
from dateutil.relativedelta import *

class TimeBasedCV(object):
    '''
    Parameters 
    ----------
    train_period: int