#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Tue Mar 13 13:16:33 2018

@author: teacher
"""
import matplotlib.pyplot as plt
import pandas as pd
import math

# Define the data type for two of the columns to get rid of warning message.
dtypes = {'Landmark':'str','Facility Type':'str'}
# Read in the CSV file, telling Pandas to read the 'Created Date' column as datetime types.
potholes = pd.read_csv("311_Service_Requests_potholes.csv",parse_dates=['Created Date'],dtype = dtypes)

# We don't need this line, since we are now using the parse_dates option in read_csv() above
#potholes['Created Date'] = pd.to_datetime(potholes['Created Date'])

# Get the counts of how many pothole requests were made on each date
pothole_counts = potholes['Created Date'].dt.date.value_counts()

# Plot a line graph of the number of pothole requests each day.
# The option rot=90 rotates the x axis labels by 90 degrees.
pothole_counts.plot(rot=90)
plt.show()

# Plot a histogram of the number of pothole requests each day.
pothole_counts.plot.hist(bins = 20)
plt.show()

# Population = pothole counts for each day of the year
# Compute population mean
mu = pothole_counts.mean()
print("Population mean is",mu)
# Compute standard deviation of the population
# use option ddof = 0 to get population standard deviation
sigma = pothole_counts.std(ddof = 0)
# define n = sample size
n = 10000
# create an empty list for the sample means
sample_means_list = []
# start a loop
for i in range(10000):
# inside the loop:
    # sample from the population
    pothole_sample = pothole_counts.sample(n, replace=True)
    # compute mean of sample
    sample_mean = pothole_sample.mean()
    # store sample mean in list
    sample_means_list.append(sample_mean)

# convert list to Series
sample_means_series = pd.Series(sample_means_list)
# plot histogram of series
sample_means_series.plot.hist(bins =40)
# print the mean of the samples means (series)
print("Mean of sample means is",sample_means_series.mean())
# print the standard deviation of the sample means
print("Standard deviation of sample means is",sample_means_series.std(ddof=0))
# value of the standard deviation of the sample means predicted by the Central Limit Theorem:
print("sigma/sqrt(n) is",sigma/math.sqrt(n))