我先说：

我知道这是一个经常被问到的问题。我读了其他答案，排除了：

我没有用+=来做作业

我尝试在函数中显式地分配每个变量，以确保它们不是空的，以防函数执行的其他工作失败

它们是不是全局变量，我不希望它们是——它们只是内部变量，我用它们来计算我最终返回的是什么。在## Gets the data from external website - refreshes whenever the programme is called.

## Urllib2 required module

## csv to make life easier handling the data

import urllib2

import csv

import sys

import math

# import sqlite3 #don't need this just now, will probably run Django with MySQL when it comes to it

# import MySQLdb Likewise, don't need this just now.

#python3

import atexit

from time import time

from datetime import timedelta

def secondsToStr(t):

return str(timedelta(seconds=t))

line = "="*40

def log(s, elapsed=None):

print(line)

print(secondsToStr(time()), '-', s)

if elapsed:

print("Elapsed time:", elapsed)

print(line)

print()

def endlog():

end = time()

elapsed = end-start

log("End Program", secondsToStr(elapsed))

def now():

return secondsToStr(time())

start = time()

atexit.register(endlog)

log("Start Program")

def open_external_source():

# Checks if the core file's been modified since the last time we used it - if it hasn't, then we skip all of the file reading stuff.

#need to change this to just pull the headers the first time.

master_data_file = urllib2.urlopen("http://www.football-data.co.uk/mmz4281/1213/E0.csv", "GET")

print master_data_file

headers = master_data_file.info()

last_mod = headers["last-modified"]

settings = open ("settings.csv","r+")

historic_last_mod = settings.readline() #this only works when the setting is a 1 line file

print "Local file version: " + historic_last_mod

print "Server file version: " +last_mod

if last_mod == historic_last_mod :

print "It's the same, file not loaded"

return true

else :

return false

settings.close()

#the if statement's commented out because it was messing up the variables into the function

#if open_external_source == False:

master_data_file = urllib2.urlopen("http://www.football-data.co.uk/mmz4281/1213/E0.csv", "GET")

data = list(tuple(rec) for rec in csv.reader(master_data_file, delimiter=','))

print len(data)

print "printing full file"

print data

league_list = ["Arsenal", "Chelsea", "Liverpool", "Man City", "Man United", "Newcastle", "Newcastle", "Norwich","Reading","Southampton", "Stoke", "Sunderland", "Swansea", "Tottenham", "West Brom", "West Ham", "Wigan"]

league_stats = league_list

#for teams in league_list: - come back to this, will do this as a split and append.

#call the next set of functions to skip the data reading stuff

#This is the data reading section, that puts the data into our system

#If we do proceed, then we redo all of the calculations, and read the data file in again, in case of any corrections, etc.

#Column references:

#Home Goals 4

#Away Goals 5

#Full Time Result 6

#Home Shots 10

#Away Shots 11

#Home Shots on Target 12

#Away Shots on Target 13

#Calculates the average for a given team at home, columns are 4 Home Goals, 5 Away Goa

def CalcAverageHome(team, column, data):

total = 0

count = 0

n=0

for row in data:

if data[count][2] == team:

total += int(data[count][column])

n+=1

count += 1

try:

average = float(total) / n

except ZeroDivisionError:

average = 'Not played'

return average

def CalcAverageAway(team, column, data):

total = 0

count = 0

n=0

for row in data:

if data[count][3] == team:

total += int(data[count][column])

n+=1

count += 1

try:

average = float(total) / n

except ZeroDivisionError:

average = 'Not played'

return average

home_team = "Chelsea"

away_team = "Newcastle"

print "Here's the Average number of goals scored Home"

home_goals = CalcAverageHome(home_team, 4, data)

away_goals = CalcAverageAway(home_team, 5, data)

home_conceded = CalcAverageHome(home_team, 5, data)

away_conceded = CalcAverageAway(away_team, 4, data)

adjusted_home = home_goals * away_conceded

adjusted_away = away_goals * home_conceded

print home_team, home_goals, home_conceded, adjusted_home

print away_team, away_goals, away_conceded, adjusted_away

print "starting to try and work the league averages out here."

def poisson_probability(actual, mean):

# naive: math.exp(-mean) * mean**actual / factorial(actual)

# iterative, to keep the components from getting too large or small:

p = math.exp(-mean)

for i in xrange(actual):

p *= mean

p /= i+1

return p

for i in range (10):

print str((100*poisson_probability(i,adjusted_home)))+"%"

league_list = ["Arsenal", "Chelsea", "Liverpool", "Man City", "Man United", "Newcastle", "Newcastle", "Norwich","Reading","Southampton", "Stoke", "Sunderland", "Swansea", "Tottenham", "West Brom", "West Ham", "Wigan"]

# just assign the league list to the stats for now -

# eventually each team entry will become the first column of a new sublist

def LeagueAverages(data,column):

total = 0

n = 0

for row in data :

string = row[column]

if string.isdigit() == True:

total = total + int(row[column])

n += 1

league_average = float(total) / n

return league_average

print "League home goals average is:", LeagueAverages(data, 4)

print "League away goals average is:", LeagueAverages(data, 5)

print "finished that loop..."

league_stats = []

test_team = "Arsenal"

# Function iterates through the league teams and calculates the averages

# and places them in one long list.

for team in league_list:

league_stats.append(team)

league_stats.append(CalcAverageHome(team, 4, data))

print CalcAverageHome(team, 4, data)

league_stats.append(CalcAverageHome(team, 5, data))

CalcAverageHome(team, 5, data)

league_stats.append(CalcAverageHome(team, 7, data))

CalcAverageHome(team, 7, data)

league_stats.append(CalcAverageHome(team, 8, data))

CalcAverageHome(team, 8, data)

league_stats.append(CalcAverageHome(team, 10, data))

CalcAverageHome(team, 10, data)

league_stats.append(CalcAverageHome(team, 11, data))

CalcAverageHome(team, 11, data)

league_stats.append(CalcAverageHome(team, 12, data))

CalcAverageHome(team, 12, data)

league_stats.append(CalcAverageHome(team, 13, data))

CalcAverageHome(team, 13, data)

# This function should chunk the 'file', as when we run the above code,

# we'll end up with one incredibly long list that contains every team on the same line

def chunker(seq, size):

return (seq[pos:pos + size] for pos in xrange(0, len(seq), size))

chunker (league_stats, 9)

final_stats = []

for group in chunker(league_stats, 9):

print repr(group)

final_stats.append(repr(group))

#retrieve a particular value from the final stats array

"""

for row in final_stats:

if data[count][2] == team:

total += int(data[count][column])

n+=1

count += 1

"""

def create_probability_table(hometeam, awayteam, final_stats):

#reads in the home and away sides, calculates their performance adjusted

#ratings and then calculates the likelihood of each team scoring a particular

#number of goals (from 0-10)

#those likelihoods are then combined to provide an 11x11 matrix of probabilities

poisson_array = []

poisson_list_home = []

poisson_list_away = []

goals_home = 0

conceded_home = 0

goals_away = 0

conceded_away = 0

for team in final_stats:

if team == hometeam:

goals_home = team[1]

conceded_home = team [3]

print "home Goals, Home Conceded"

print goals_home, conceded_home

elif team == awayteam:

goals_away = team[2]

conceded_away = team[4]

print "Away Goals, Away Conceded"

print goals_away, conceded_away,

else:

pass

adjusted_goals_home = goals_home * conceded_away

adjusted_goals_away = goals_away * conceded_home

#this section creates the two probability lists for home and away num goals scored

for i in range (10):

poisson_list_home.append = (100*poisson_probability(i,adjusted_goals_home))

poisson_list_away.append = (100*poisson_probability(i,adjusted_goals_away))

print poisson_list_home

print poisson_list_away

for number in poisson_list_home:

for number in poisson_list_away:

probability_table.append(poisson_list_home[number] * poisson_list_away[number])

return probability_table

create_probability_table("Arsenal", "Chelsea", final_stats)

#and this section cross multiplies them into a new list

# for i in range (10):

# print data_frame [0:100] prints to console to provide visual check

master_data_file.close()

当我运行它时，它会抛出一个

^{pr2}$

错误-我不明白为什么！它是在函数开始时定义和分配的。这不是全球性的。在