Create a Program to Implement Statistics Visualization in Python Assignment Solution.

# Data visualization for Flights dataset

### Load needed library

import pandas as pd

import numpy as np

import time

import warnings

warnings.filterwarnings('ignore')

### Import our Data

df = pd.read_csv('flights.csv')

# Data Overview

df.head(100)

df.shape

### know the number of columns and row of our dataset

df.info()

df.dtypes

### know the type of every column

df.describe()

### Descriptive stats for our data

df.isna().sum()

### Check if there is missing data

df.duplicated().sum()

### we do not have any deuplicated rows

df.nunique().to_frame().rename(columns={0:'Count'})

### check the uniqueness of our columns row

df['carrier'].unique()

### know how many carriers that we have in our dataset

df['year'].unique()

df.day.describe()

print(df['carrier'].value_counts())

print(['WN: Southwest Airlines', 'AA: American Airlines', 'MQ: American Eagle Airlines', 'UA: United Airlines',

'OO: Skywest Airlines','DL: Delta Airlines','US: US Airways',

'EV: Atlantic Southeast Airlines','FL: AirTran Airways','YV: Mesa Airlines',

'B6: JetBlue Airways','9E: Pinnacle Airlines','AS: Alaska Airlines','F9: Frontier Airlines',

'HA: Hawaiian Airlines'])

### every airline and its frequency journies

# Data cleaning

### see the missing data

missing_data = df.isnull().sum(axis=0).reset_index()

missing_data.columns = ['variable', 'missing values']

missing_data['filling factor (%)']=(df.shape[0]-missing_data['missing values'])/df.shape[0]*100

missing_data.sort_values('filling factor (%)').reset_index(drop = True)

### we see that those columns (air_time, arr_delay, arr_time, dep_time,dep_delay) have some missing data that we must handel

df=df.dropna()

### remove any row with at least one missing value

df.isna().sum()

### now no nulls

cols=["day","month","year"]

df['date'] = df[cols].apply(lambda x: '-'.join(x.values.astype(str)), axis="columns")

### join the coulmns day month year to one colmun to be the date of our data

### Setting the Frequency

df.set_index("date", inplace=True)

df.head()

### our data now sorted by the date

# Exploratory Data Analysis

import matplotlib.pyplot as plt

import seaborn as sns

import warnings

warnings.filterwarnings('ignore')

#correlation matrix

corrmat = df.corr()

f, ax = plt.subplots(figsize=(12, 9))

sns.heatmap(corrmat, vmax=.8, square=True);

plt.show()

### this heatmap give us an intuation for the corrolation between our dataset columns which indicates that so coluns have very strong correlation like arrival time and departure time, other have so low correlation like arrival delay and distance.

delay_type = lambda x:((0,1)[x > 5],2)[x > 45]

fig = plt.figure(1, figsize=(10,7))

ax = sns.countplot(y="carrier", hue='year', data=df)

# Setting Labels

plt.setp(ax.get_xticklabels(), fontsize=12, weight = 'normal', rotation = 0);

plt.setp(ax.get_yticklabels(), fontsize=12, weight = 'bold', rotation = 0);

ax.yaxis.label.set_visible(False)

plt.xlabel('Flight count', fontsize=16, weight = 'bold', labelpad=10)

### this count plot give us an observiation on which is the most airlines have made flights in 2013

### we find that it is UA and B6 have most flights for 2014 where is very low flights for OO(SKYWEST)

#Status on time (0),

#slightly delayed (1),

#highly delayed (2),

for dataset in df:

df.loc[df['arr_delay'] <= 10, 'Status'] = 0

df.loc[df['arr_delay'] >= 10, 'Status'] = 1

df.loc[df['arr_delay'] >= 30, 'Status'] = 2

f,ax=plt.subplots(1,2,figsize=(20,8))

df['Status'].value_counts().plot.pie(autopct='%1.1f%%',ax=ax[0],shadow=True)

ax[0].set_title('Status')

ax[0].set_ylabel('')

sns.countplot('Status',order = df['Status'].value_counts().index, data=df,ax=ax[1])

ax[1].set_title('Status')

plt.show()

### In 2013, a 71% of flights were delayed by more than 10 minutes. 12.9% of flights had delays of more than 10 min and less than half hour.On the other hand, 16.6% above hour

delay = df[(df.Status >= 1) &(df.Status < 3)]

#histogram

sns.distplot(delay['arr_delay'])

plt.show()

### It can be seen that delays are mostly located on the left side of the graph,The most of delays are short, and unusual we have very large delay

fig = plt.figure(figsize=(20,8))

delay[['month','arr_delay']].groupby(['month']).mean().plot()

plt.show()

### Delays focused on February, June and December, might the cause of the sumer and winter holidays

fig = plt.figure(figsize=(20,8))

delay[['hour','arr_delay']].groupby(['hour']).mean().plot()

plt.show()

### it is clear the the delays is rush between the 17:21 hour

carrier_delay = df[['hour','carrier']].groupby(['carrier']).head()

carrier_delay

df.arr_delay.plot(figsize=(20,5))

plt.title("delays over 2013", size = 24)

plt.ylim(0,1400)

plt.show()

f,ax=plt.subplots(1,figsize=(20,8))

sns.barplot('carrier','arr_delay', data=delay,ax=ax,

order=['WN', 'AA','B6','AS', 'MQ',

'UA','OO','DL','US','EV','FL',

'YV', '9E','F9','HA'])