In [8]:

# Python ≥3.5 (ideally)
import platform
import sys, getopt
assert sys.version_info >= (3, 5)
import csv

# Import Dependencies
%matplotlib inline

# Math Operations
import numpy as np
from math import pi

# Datetime
import datetime
from datetime import date
import time

# Data Preprocessing
import pandas as pd    # version 1.0.3
import os    #  used to read the csv filenames
import re
import random
from io import BytesIO
from pathlib import Path

# Reading directories
import glob
import os

# Working with JSON
import json
from pandas.io.json import json_normalize
from ast import literal_eval

# Data Visualisation
import matplotlib as mpl
import matplotlib.pyplot as plt
import seaborn as sns
plt.style.use('seaborn-whitegrid')
import missingno as msno    # visually display missing data

# Display in Jupyter
from IPython.display import Image, YouTubeVideo
from IPython.core.display import HTML

# Ignore Warnings
import warnings
warnings.filterwarnings(action="ignore", message="^internal gelsd")

print('Setup Complete')

Setup Complete

In [9]:

# Python / module versions used here for reference
print('Python: {}'.format(platform.python_version()))
print('NumPy: {}'.format(np.__version__))
print('pandas: {}'.format(pd.__version__))
print('matplotlib: {}'.format(mpl.__version__))
print('Seaborn: {}'.format(sns.__version__))

Python: 3.7.6
NumPy: 1.18.1
pandas: 1.0.1
matplotlib: 3.1.3
Seaborn: 0.10.0

In [10]:

# Set up initial paths to subfolders
base_dir = os.path.join('..', '..', '..', )
data_dir = os.path.join(base_dir, 'data')
data_dir_fbref = os.path.join(base_dir, 'data', 'fbref')
data_dir_stratabet = os.path.join(base_dir, 'data', 'stratabet')
img_dir = os.path.join(base_dir, 'img')
fig_dir = os.path.join(base_dir, 'img', 'fig')
video_dir = os.path.join(base_dir, 'video')

In [11]:

pd.set_option('display.max_columns', None)

Chances¶

Files A¶

In [12]:

# Combine individual csv files to form one DataFrame, df_raw, using glob
all_files = glob.glob(data_dir_stratabet + '/downloaded/collection/Chances' + "/*.csv")    # Creates a list of all csv files

li = []    # pd.concat takes a list of DataFrames as an argument

for filename in all_files:
    df_raw_temp = pd.read_csv(filename, index_col=None, header=0)
    li.append(df_raw_temp)

df_chances = pd.concat(li, axis=0, ignore_index=True)    # ignore_index=True as we don't want pandas to try an align row indexes

In [13]:

df_chances = df_chances.sort_values(['competition', 'kickoffDate', 'kickoffTime'], ascending=[True, True, True])

In [14]:

df_chances

Out[14]:

	index	competition	gsm_id	kickoffDate	kickoffTime	hometeam_team1	awayteam_team2	icon	chanceRating	team	type	time	player	location_x	location_y	bodyPart	shotQuality	defPressure	numDefPlayers	numAttPlayers	outcome	primaryPlayer	primaryType	primaryLocation_x	primaryLocation_y	secondaryPlayer	secondaryType
7116	2444	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	verygoodchance	verygoodchance	Melbourne Victory	Open Play	15:57	Baro	3.0	28.0	Head	2	4	2	0	Missed	J. Troisi	Free Kick	-120.0	37.0	-	-
7117	2445	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	poorchance	poorchance	Brisbane Roar	Open Play	00:52	B. Borello	1.0	79.0	Left	1	1	1	0	Missed	T. Broich	Open Play Pass	65.0	86.0	-	-
7118	2446	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	poorchance	poorchance	Melbourne Victory	Open Play	06:16	F. Ben Khalfallah	-43.0	86.0	Right	1	3	3	0	Defended	-	-	-	-	-	-
7119	2447	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	goodchance	goodchance	Melbourne Victory	Open Play	09:22	F. Ben Khalfallah	-26.0	41.0	Left	2	3	3	0	Defended	M. Austin	Cross High	103.0	32.0	-	-
7120	2448	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	poorchance	poorchance	Melbourne Victory	Open Play	09:32	J. Troisi	-6.0	99.0	Right	2	2	3	0	Defended	-	-	-	-	-	-
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
65405	139	TurSL	2291443	2017-06-03	14:00:00	Galatasaray	Konyaspor	poorchance	poorchance	Galatasaray	Open Play	29:38	S. Inan	-41.0	124.0	Right	2	0	2	1	Missed	-	-	-	-	-	-
65406	140	TurSL	2291443	2017-06-03	14:00:00	Galatasaray	Konyaspor	fairlygoodchance	fairlygoodchance	Galatasaray	Dangerous Moment	03:50	W. Sneijder	-3.0	29.0	Right	0	4	1	0	Defended	S. Gumus	Cross Low	-38.0	0.0	-	-
65407	141	TurSL	2291443	2017-06-03	14:00:00	Galatasaray	Konyaspor	fairlygoodchance	fairlygoodchance	Galatasaray	Open Play	37:33	T. Cigerci	13.0	36.0	Other	2	4	3	1	Defended	W. Sneijder	Corner	136.0	0.0	L. Podolski	Corner Won
65408	142	TurSL	2291443	2017-06-03	14:00:00	Galatasaray	Konyaspor	poorchance	poorchance	Galatasaray	Open Play	43:20	M. Linnes	47.0	93.0	Right	2	0	3	0	Missed	Bruma	Open Play Pass	92.0	98.0	S. Inan	Open Play Pass
65409	143	TurSL	2291443	2017-06-03	14:00:00	Galatasaray	Konyaspor	poorchance	poorchance	Konyaspor	Open Play	44:43	M. Mbamba	17.0	77.0	Right	2	0	4	1	Defended	-	-	-	-	-	-

95927 rows × 27 columns

In [15]:

lst_competitions = list(df_chances['competition'].unique())

In [16]:

for i, g in df_chances.groupby('competition'):
    g.to_csv(data_dir_stratabet + '/raw/chances/individual_competitions/stratabet_events_chances_{}.csv'.format(i), header=True, index_label=True)

In [17]:

 #df_chances.to_csv(data_dir_stratabet + '/raw/chances/' + 'stratabet_chances_all.csv', index=None, header=True)

Files B¶

In [23]:

# Combine individual csv files to form one DataFrame, df_raw, using glob
all_files = glob.glob(data_dir_stratabet + '/downloaded/danclarke93' + "/*.csv")    # Creates a list of all csv files

li = []    # pd.concat takes a list of DataFrames as an argument

for filename in all_files:
    df_raw_temp = pd.read_csv(filename, index_col=None, header=0)
    li.append(df_raw_temp)

df_chances = pd.concat(li, axis=0, ignore_index=True)    # ignore_index=True as we don't want pandas to try an align row indexes

In [24]:

df_chances = df_chances.sort_values(['competition', 'kickoffDate', 'kickoffTime'], ascending=[True, True, True])

In [25]:

df_chances

Out[25]:

	Unnamed: 0	competition	gsm_id	kickoffDate	kickoffTime	hometeam_team1	awayteam_team2	icon	chanceRating	team	type	time	player	location_x	location_y	bodyPart	shotQuality	defPressure	numDefPlayers	numAttPlayers	outcome	primaryPlayer	primaryType	primaryLocation_x	primaryLocation_y	secondaryPlayer	secondaryType
21344	11404	EngCh	2467215	2017-08-04	18:45:00	Nottingham Forest	Millwall	goodchance	goodchance	Nottingham Forest	Open Play	74:07	Z. Clough	-16.0	56.0	Left	3	5	2	0	Defended	K. Dowell	Open Play Pass	-58.0	32.0	-	-
21345	11401	EngCh	2467215	2017-08-04	18:45:00	Nottingham Forest	Millwall	poorchance	poorchance	Nottingham Forest	Open Play	80:18	B. McKay	12.0	102.0	Right	3	4	2	0	Defended	A. Traore	Open Play Pass	102.0	171.0	-	-
21346	11402	EngCh	2467215	2017-08-04	18:45:00	Nottingham Forest	Millwall	poorchance	poorchance	Millwall	Open Play	77:58	F. Onyedinma	-62.0	16.0	Right	3	5	2	0	Saved	-	-	-	-	-	-
21347	11403	EngCh	2467215	2017-08-04	18:45:00	Nottingham Forest	Millwall	verygoodchance	verygoodchance	Millwall	Dangerous Moment	75:18	S. Morison	32.0	12.0	Left	0	0	3	1	Defended	J. Wallace	Cross High	-70.0	22.0	-	-
21348	11405	EngCh	2467215	2017-08-04	18:45:00	Nottingham Forest	Millwall	fairlygoodchance	fairlygoodchance	Millwall	Open Play	68:04	L. Gregory	11.0	45.0	Left	1	5	3	0	Missed	S. Morison	Open Play Pass	29.0	35.0	J. Wallace	Cross Low
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
21422	80263	SpaPr	2529514	2018-03-12	20:00:00	Deportivo Alaves	Real Betis	fairlygoodchance	fairlygoodchance	Deportivo Alaves	Open Play	88:24	Bojan	10.0	60.0	Right	2	1	4	0	Missed	H. Perez	Cross High	-115.0	56.0	Munir	Open Play Pass
21423	10892	SpaPr	2529514	2018-03-12	20:00:00	Deportivo Alaves	Real Betis	goal	Very Good	Real Betis	Open Play	77:08	Loren Moron	-47.0	30.0	Right	3	2	2	0	-	R. Boudebouz	Open Play Pass	-94.0	160.0	Fabian Ruiz	Open Play Pass
21424	10891	SpaPr	2529514	2018-03-12	20:00:00	Deportivo Alaves	Real Betis	goal	Very Good	Deportivo Alaves	Open Play	66:06	Ruben Sobrino	-35.0	47.0	Right	3	3	1	0	-	Martin	Open Play Pass	-9.0	196.0	Tomas Pina	Open Play Pass
21425	10890	SpaPr	2529514	2018-03-12	20:00:00	Deportivo Alaves	Real Betis	goal	Great	Real Betis	Open Play	43:20	Javi Garcia	30.0	12.0	Right	4	1	1	0	-	A. Mandi	Shot (Opposition Rebound)	34.0	33.0	R. Boudebouz	Corner
21426	10889	SpaPr	2529514	2018-03-12	20:00:00	Deportivo Alaves	Real Betis	goal	Great	Real Betis	Open Play	22:39	Loren Moron	-38.0	44.0	Right	3	1	1	0	-	Victor Camarasa	Open Play Pass	-56.0	112.0	Francis Guerrero	Throw in

27254 rows × 27 columns

In [26]:

lst_competitions = list(df_chances['competition'].unique())

In [27]:

for i, g in df_chances.groupby('competition'):
    g.to_csv(data_dir_stratabet + '/raw/chances/individual_competitions/stratabet_events_chances_{}.csv'.format(i), header=True, index_label=True)

Mins Played¶

In [21]:

# Combine individual csv files to form one DataFrame, df_raw, using glob
all_files = glob.glob(data_dir_stratabet + '/downloaded/collection/Mins Played' + "/*.csv")    # Creates a list of all csv files

li = []    # pd.concat takes a list of DataFrames as an argument

for filename in all_files:
    df_raw_temp = pd.read_csv(filename, index_col=None, header=0)
    li.append(df_raw_temp)

df_mins = pd.concat(li, axis=0, ignore_index=True)    # ignore_index=True as we don't want pandas to try an align row indexes

In [26]:

df_mins = df_mins.sort_values(['competition', 'kick_off_dt'], ascending=[True, True])

In [27]:

df_mins

Out[27]:

	index	comp_id	competition	end_added	end_min	fixture_id	fixture_week	kick_off_dt	player	player_id	season	season_id	start_added	start_min	start_status	team	team_id	location	mins_played
15018	0	283	AusAL	0	90	2238877	1.0	2016-10-07	M. Theo	17526	2016/2017	12512	0	0	Start	Brisbane Roar	4321	home	90
15019	1	283	AusAL	0	90	2238877	1.0	2016-10-07	J. Hingert	88322	2016/2017	12512	0	0	Start	Brisbane Roar	4321	home	90
15020	2	283	AusAL	0	90	2238877	1.0	2016-10-07	C. Brown	193345	2016/2017	12512	0	0	Start	Brisbane Roar	4321	home	90
15021	3	283	AusAL	0	90	2238877	1.0	2016-10-07	J. North	17558	2016/2017	12512	0	0	Start	Brisbane Roar	4321	home	90
15022	4	283	AusAL	0	90	2238877	1.0	2016-10-07	L. DeVere	56394	2016/2017	12512	0	0	Start	Brisbane Roar	4321	home	90
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
3535	372	19	TurSL	0	90	2506732	5.0	2017-09-18	J. Jonsson	196079	2017/2018	14277	0	-1	Unused Sub	Konyaspor	2223	away	0
3536	373	19	TurSL	0	90	2506732	5.0	2017-09-18	D. Milosevic	215347	2017/2018	14277	0	71	Sub	Konyaspor	2223	away	19
3537	374	19	TurSL	0	90	2506732	5.0	2017-09-18	V. Bora	150431	2017/2018	14277	0	-1	Unused Sub	Konyaspor	2223	away	0
3538	375	19	TurSL	0	90	2506732	5.0	2017-09-18	M. Araz	154283	2017/2018	14277	0	63	Sub	Konyaspor	2223	away	27
3539	376	19	TurSL	0	90	2506732	5.0	2017-09-18	P. Eze	316804	2017/2018	14277	0	-1	Unused Sub	Konyaspor	2223	away	0

155124 rows × 19 columns

In [28]:

lst_competitions = list(df_mins['competition'].unique())

In [31]:

for i, g in df_mins.groupby('competition'):
    g.to_csv(data_dir_stratabet + '/raw/mins_played/individual_competitions/stratabet_events_mins_played_{}.csv'.format(i), header=True, index_label=True)

In [50]:

df_mins.to_csv(data_dir_stratabet + '/raw/mins_played/' + 'stratabet_mins_played_all.csv', index=None, header=True)

Key Entries¶

In [32]:

# Combine individual csv files to form one DataFrame, df_raw, using glob
all_files = glob.glob(data_dir_stratabet + '/downloaded/collection/Key Entries' + "/*.csv")    # Creates a list of all csv files

li = []    # pd.concat takes a list of DataFrames as an argument

for filename in all_files:
    df_raw_temp = pd.read_csv(filename, index_col=None, header=0)
    li.append(df_raw_temp)

df_key_entries = pd.concat(li, axis=0, ignore_index=True)    # ignore_index=True as we don't want pandas to try an align row indexes

In [34]:

df_key_entries

Out[34]:

	index	competition	gsm_id	kickoffDate	kickoffTime	hometeam_team1	awayteam_team2	team	keyentryArea	keyentryType
0	0	AusAL	2239006	2017-04-09	07:00:00	Central Coast Mariners	Newcastle Jets	Central Coast Mariners	Right	Pass
1	1	AusAL	2239006	2017-04-09	07:00:00	Central Coast Mariners	Newcastle Jets	Central Coast Mariners	Left	Run
2	2	AusAL	2239006	2017-04-09	07:00:00	Central Coast Mariners	Newcastle Jets	Central Coast Mariners	Left	Run
3	3	AusAL	2239006	2017-04-09	07:00:00	Central Coast Mariners	Newcastle Jets	Newcastle Jets	Box	Run
4	4	AusAL	2239006	2017-04-09	07:00:00	Central Coast Mariners	Newcastle Jets	Central Coast Mariners	Box	Pass
...	...	...	...	...	...	...	...	...	...	...
195682	2178	GerBL2	2255467	2017-04-16	11:30:00	Bochum	Greuther Furth	Bochum	Right	Pass
195683	2179	GerBL2	2255467	2017-04-16	11:30:00	Bochum	Greuther Furth	Bochum	Box	Pass
195684	2180	GerBL2	2255467	2017-04-16	11:30:00	Bochum	Greuther Furth	Greuther Furth	Right	Run
195685	2181	GerBL2	2255467	2017-04-16	11:30:00	Bochum	Greuther Furth	Greuther Furth	Box	Pass
195686	2182	GerBL2	2255467	2017-04-16	11:30:00	Bochum	Greuther Furth	Bochum	Right	Pass

195687 rows × 10 columns

In [36]:

df_key_entries = df_key_entries.sort_values(['competition', 'kickoffDate', 'kickoffTime'], ascending=[True, True, True])

In [37]:

df_key_entries

Out[37]:

	index	competition	gsm_id	kickoffDate	kickoffTime	hometeam_team1	awayteam_team2	team	keyentryArea	keyentryType
6550	5788	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	Melbourne Victory	Left	Run
6551	5789	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	Melbourne Victory	Box	Pass
6552	5790	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	Brisbane Roar	Left	Pass
6553	5791	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	Brisbane Roar	Box	Run
6554	5792	AusAL	2238877	2016-10-07	08:50:00	Brisbane Roar	Melbourne Victory	Melbourne Victory	Left	Pass
...	...	...	...	...	...	...	...	...	...	...
78488	210	TurSL	2291441	2017-06-03	14:00:00	Trabzonspor	Bursaspor	Bursaspor	Left	Pass
78489	211	TurSL	2291441	2017-06-03	14:00:00	Trabzonspor	Bursaspor	Trabzonspor	Right	Pass
78490	212	TurSL	2291441	2017-06-03	14:00:00	Trabzonspor	Bursaspor	Trabzonspor	Right	Pass
78491	213	TurSL	2291441	2017-06-03	14:00:00	Trabzonspor	Bursaspor	Trabzonspor	Right	Pass
78492	214	TurSL	2291441	2017-06-03	14:00:00	Trabzonspor	Bursaspor	Trabzonspor	Left	Pass

195687 rows × 10 columns

In [39]:

lst_competitions = list(df_key_entries['competition'].unique())

In [40]:

for i, g in df_key_entries.groupby('competition'):
    g.to_csv(data_dir_stratabet + '/raw/key_entries/individual_competitions/stratabet_events_key_entries_{}.csv'.format(i), header=True, index_label=True)

In [49]:

df_key_entries.to_csv(data_dir_stratabet + '/raw/key_entries/' + 'stratabet_key_entries_all.csv', index=None, header=True)

Match Info¶

In [41]:

# Combine individual csv files to form one DataFrame, df_raw, using glob
all_files = glob.glob(data_dir_stratabet + '/downloaded/collection/Match Info' + "/*.csv")    # Creates a list of all csv files

li = []    # pd.concat takes a list of DataFrames as an argument

for filename in all_files:
    df_raw_temp = pd.read_csv(filename, index_col=None, header=0)
    li.append(df_raw_temp)

df_match_info = pd.concat(li, axis=0, ignore_index=True)    # ignore_index=True as we don't want pandas to try an align row indexes

In [42]:

df_match_info

Out[42]:

	index	competition	gsm_id	awayteam_team2	hometeam_team1	kickoffDate	kickoffTime	pitchType	weatherType	tempCelcius	tempType	formation_team1	formation_team2	formation_time
0	0	AusAL	2239006	Newcastle Jets	Central Coast Mariners	2017-04-09	07:00:00	Average	Cloud, Sun	23	-	4_1_4_1	4_2_3_1	83:35
1	1	AusAL	2239005	Brisbane Roar	Perth Glory	2017-04-08	12:00:00	Average	Clear	18	-	4_4_2	4_2_3_1	79:52
2	2	AusAL	2239004	Melbourne Victory	Western Sydney Wanderers	2017-04-08	09:50:00	Average	Cloud, Sun	22	Warm	4_2_3_1	4_2_3_1	79:18
3	3	AusAL	2239003	Sydney	Wellington Phoenix	2017-04-08	07:35:00	Average	Cloud	13	-	3_1_3_1_2	4_2_3_1	84:35
4	4	AusAL	2239003	Sydney	Wellington Phoenix	2017-04-08	07:35:00	Average	Cloud	13	-	4_1_2_1_2	4_2_3_1	65:16
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
7223	428	TurSL	2291147	Kasimpasa	Trabzonspor	2016-08-20	16:30:00	Average, Wet	Cloud, Light Rain	25	Warm	4_1_4_1	4_2_3_1	81:11
7224	429	TurSL	2291147	Kasimpasa	Trabzonspor	2016-08-20	16:30:00	Average, Wet	Cloud	25	Warm	4_1_4_1	4_2_3_1	81:11
7225	430	TurSL	2291151	Konyaspor	Rizespor	2016-08-20	16:30:00	Good	Clear	25	Warm	4_1_4_1	4_4_2	84:29
7226	431	TurSL	2291151	Konyaspor	Rizespor	2016-08-20	16:30:00	Good	Clear	25	Warm	4_2_3_1	4_4_2	68:11
7227	432	TurSL	2291149	Bursaspor	Adanaspor	2016-08-19	18:45:00	Average, Good, Patchy	Clear	26	-	4_2_3_1	4_2_3_1	83:04

7228 rows × 14 columns

In [43]:

df_match_info = df_match_info.sort_values(['competition', 'kickoffDate', 'kickoffTime'], ascending=[True, True, True])

In [44]:

df_match_info

Out[44]:

	index	competition	gsm_id	awayteam_team2	hometeam_team1	kickoffDate	kickoffTime	pitchType	weatherType	tempCelcius	tempType	formation_team1	formation_team2	formation_time
192	173	AusAL	2238877	Melbourne Victory	Brisbane Roar	2016-10-07	08:50:00	Excellent	Clear	21	Warm	4_3_3	4_3_3	88:06
190	171	AusAL	2238878	Melbourne City	Wellington Phoenix	2016-10-08	06:35:00	Good	Cloud	14	-	4_2_3_1	4_4_2	69:06
191	172	AusAL	2238878	Melbourne City	Wellington Phoenix	2016-10-08	06:35:00	Good	Cloud	14	-	4_2_3_1	3_5_2	61:19
188	169	AusAL	2238879	Sydney	Western Sydney Wanderers	2016-10-08	08:50:00	Excellent	Clear	17	Warm	4_3_3	4_2_3_1	83:12
189	170	AusAL	2238879	Sydney	Western Sydney Wanderers	2016-10-08	08:50:00	Excellent	Clear	17	Warm	4_3_3	5_4_1	91:47
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
6795	0	TurSL	2291442	Alanyaspor	Rizespor	2017-06-03	14:00:00	Good	Clear, Sun	19	Warm	3_5_2	4_1_4_1	77:00
6796	1	TurSL	2291449	Osmanlispor	Besiktas	2017-06-03	14:00:00	Excellent	Cloud, Sun	28	-	4_2_3_1	4_2_3_1	74:10
6797	2	TurSL	2291444	Fenerbahce	Adanaspor	2017-06-03	14:00:00	Good	Sun	28	Hot	4_5_1	4_2_3_1	80:52
6798	3	TurSL	2291443	Konyaspor	Galatasaray	2017-06-03	14:00:00	Average, Soft	Sun	27	Hot	4_2_3_1	4_2_2_2	80:14
6799	4	TurSL	2291441	Bursaspor	Trabzonspor	2017-06-03	14:00:00	Excellent	Cloud, Sun	22	Warm	4_2_3_1	4_2_3_1	73:59

7228 rows × 14 columns

In [45]:

lst_competitions = list(df_match_info['competition'].unique())

In [46]:

for i, g in df_match_info.groupby('competition'):
    g.to_csv(data_dir_stratabet + '/raw/match_info/individual_competitions/stratabet_match_info_{}.csv'.format(i), header=True, index_label=True)

In [47]:

df_match_info.to_csv(data_dir_stratabet + '/raw/match_info/' + 'stratabet_match_info_all.csv', index=None, header=True)