RatioClothing-Project/ratio_eda.py at master · ericpowell03/RatioClothing-Project · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns

def make_ratio_df(file_id_userid, file_id, file_sizing_data):
	'''
	INPUT: path to csv files of userids for id, orders, and sizing data
	OUTPUT: new data frame of only repeated orders and the new dataframe of the total orders
	'''
	id_userid = pd.read_csv(file_id_userid)
	ids = pd.read_csv(file_id)
	ratio = pd.read_csv(file_sizing_data)

	orders  = pd.merge(id_userid, ids, 'left', 'id')
	total_orders = pd.merge(orders, ratio, 'inner', 'user_id')
	index = total_orders.duplicated('user_id').loc

	df_duplicates = total_orders.ix[total_orders.duplicated('user_id'), :]

	df_duplicates.to_csv('data/repeat_orders')
	file_loc = 'data/repeat_orders'
	df = clean_data(file_loc)
	df.to_csv('data/cleaned_ratio_data')

	return total_orders, df

def make_retention_df(total_orders):
	'''
	INPUT: dataframe of total orders made
	OUTPUT: dataframe with a churn column for if customers were purchased again
	'''
	total_orders.to_csv('data/total_orders')
	num_shirt = pd.DataFrame(total_orders.groupby('user_id').size())
	num_shirt.rename(columns={0: 'retention'}, inplace=True)
	num_shirt.reset_index(inplace=True)
	retention = pd.merge(total_orders, num_shirt, 'inner', 'user_id')
	retention.eval('churn = retention <= 1')
	retention.churn = retention.churn.astype(int)
	drop_columns = ['id_y', 'id_x', 'retention']
	retention.drop(drop_columns, axis=1, inplace=True)
	retention.drop_duplicates('user_id', inplace=True)
	retention.to_csv('data/churn')

	file_loc_churn = 'data/churn'
	df_retention = clean_data(file_loc_churn)
	df_retention.to_csv('data/cleaned_churn')
	return df_retention

def clean_data(file_loc):
	'''
	INPUT: path to file location
	OUTPUT: cleaned dataframe with columns dropped, dummy variables, and filled-in default values
	'''
	df = pd.read_csv(file_loc)
	drop_columns = ['created_at', 'updated_at', 'source', 'birthday_month', 'posture', 'watch_wrist', 'watch_size', 'rise_inches', 'pocket_size','lastorderdate', 'shoulder_left', 'shoulder_right', 'shoulder_slope', 'estimated_birth_year', 'button_count', 'button_stance', 'posture_alteration', 'lastorderid', 'back_pleats']
	df.drop(drop_columns, axis=1, inplace=True)

	fill_columns = ['build', 'jacket_length', 'tuck', 'fit']
	df = fill_nans(df, fill_columns)

	df = dummies(df, 'build')
	df = dummies(df, 'sleeve_fit')
	df = dummies(df, 'tuck')
	df = dummies(df, 'fit')
	drop_dummies = ['Average', 'Standard', 'Both', 'Traditional']
	df.drop(drop_dummies, inplace=True, axis=1)

	return df

def dummies(df, column_list):
	df = pd.concat([df, pd.get_dummies(df[column_list])], axis=1)
	df.drop(column_list, inplace=True, axis=1)
	return df

def fill_nans(df, columns):
	for column in columns:
		if column == 'build':
			df['build'].fillna('Average', inplace=True)
		elif column == 'jacket_length':
			df['jacket_length'].fillna(0, inplace=True)
		elif column == 'tuck':
			df['tuck'].fillna('Both', inplace=True)
		elif column == 'fit':
			df['fit'].fillna('Slim', inplace=True)
	return df

def descriptions(df):
	df.describe()
	column_list = ['height_inches', 'age_years', 'weight_pounds']
	hist = histogram(df, column_list, bins=50)
	return df.describe(), hist

def plot_corr(df):
	'''
	Heat map of correlations
	'''
	plt.clf()
	drop_columns = ['Unnamed: 0', 'id_x', 'id_y', 'armpit_inches', 'torso_length_inches', 'leg_length_inches', 'knee_inches', 'arm_left_inches', 'arm_right_inches', 'thigh_inches', 'user_id', 'forearm_alteration', 'short_sleeve_opening', 'cuff', 'bicep', 'yoke', 'seat', 'arm_left_inches', 'arm_right_inches', 'Slim', 'Very Slim', 'Untucked', 'Tucked', 'Fitted', 'Slim', 'Muscular', 'Full', 'Fit', 'asym_cuff', 'asym_sleeve', 'center_back_arm_right_inches', 'center_back_arm_left_inches', 'overarm_inches', 'pant_inseam_inches', 'short_sleeve_length', 'neck', 'sleeve', 'chest', 'waist', 'length', 'jacket_length', 'shoulder_inches', 'height_inches', 'age_years', 'shirt_sleeve_inches']
	#cmap = sns.palplot(sns.color_palette("Blues"))
	sns.heatmap(df.drop(drop_columns, axis=1).corr(), cmap='YlOrRd')
	plt.title('Correlation Plot')
	plt.xticks(rotation=90)
	plt.yticks(rotation=0)
	plt.savefig('images/heat_map.png', dpi=300)

def histogram(df, column_list, bins=10):
	df[column_list].hist(bins=bins)
	plt.show()

if __name__=='__main__':

	file_id_userid = 'data/id_userid.csv'
	file_id = 'data/id.csv'
	file_sizing_data = 'data/ratio_sizing_data.csv'

	total_orders, df = make_ratio_df(file_id_userid, file_id, file_sizing_data)

	make_retention_df(total_orders)