import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np

df = pd.read_csv('E:/videogame/Video_Games_Sales_as_at_22_Dec_2016.csv')
df.head(20)

df.shape #16719 rows, 16 columns. Get to know your data size for efficient analysis

df.columns  #get all the columns

#rename the columns to avoid confusion.
df.rename(columns={'EU_Sales' : 'Europe_Sales', 'NA_Sales' : 'N.America_Sales', 'JP_Sales' : 'Japan_Sales'},  inplace=True)

df.head(10) # take a look at the data.I have chosen first 7 rows

df.info()

df.describe() #very important step in data analysis. Gives you a clear idea of sales and numerical figures

#filter out the game with most Global sales
filterr = (df['Global_Sales']==df['Global_Sales'].max())
df['Name'][filterr]  #Wii Sports rules

df['Publisher'].value_counts()

Publisher_get = df.groupby(['Publisher'])
Publisher_get.get_group('Electronic Arts')
#For the sports lovers. the data below gives you look at all the games by EA_Sports

filt = (df['Japan_Sales']==df['Japan_Sales'].max())
df['Publisher'][filt]  #Nintendo company rules iN japan 
#In other regions Wii game has the highest sales

print(df['Japan_Sales'].isna().sum())
print(df['Publisher'].isna().sum()) #Publisher has a lot of empty values lets clean it up

type(np.nan) #Nan values create problems as they are float and rest of values are integers

plt.style.use('fivethirtyeight')
fig = plt.figure(figsize=(25,25))
plt.plot(df['Name'].head(10), df['N.America_Sales'].head(10), color='red', label='N.America_Sales')
plt.plot(df['Name'].head(10), df['Japan_Sales'].head(10), color ='pink', label='Japan_Sales')
plt.plot(df['Name'].head(10), df['Europe_Sales'].head(10), color='yellow', label='Europe_Sales')
plt.tight_layout()
plt.legend()
plt.xlabel('Famous Games')
plt.ylabel('Sales')
plt.title('Popularity of Famous Games')

df.Publisher = df.Publisher.fillna('')
df.Publisher.isna().sum() #Cleared Nan values with empty string

Nin = (df['Publisher']=='Nintendo')
EA = (df['Publisher']=='Electronic Arts')
#filtering out EA sports and Nintendo to compare which compay dominates

#Nintendo sales across various regions
print(df['Japan_Sales'][Nin].sum())
print(df['Europe_Sales'][Nin].sum())
print(df['N.America_Sales'][Nin].sum())
print(df['Global_Sales'][Nin].sum())

#EA sales across various regions
print(df['Japan_Sales'][EA].sum())
print(df['Europe_Sales'][EA].sum())
print(df['N.America_Sales'][EA].sum())
print(df['Global_Sales'][EA].sum())

#Lets analyse the Genre of Video Games which are popular
df['Genre'].value_counts()

df['Genre'].fillna('Unknown', inplace = True) #Nan values replaced with "Unknown string"

labels = ['Action', 'Sports', 'Misc', 'Role-Playing', 'Shooter', 'Adventure', 'Racing', 'Platform', 'Simulation', 'Fighting', 'Strategy', 'Puzzle', 'Unknown']            

#Most popular games produced are of Action Genre followed by Sports
plt.style.use('fivethirtyeight')
fig = plt.figure(figsize=(11,11))
plt.pie(list(df['Genre'].value_counts()), autopct='%1.1f%%', labels=labels, wedgeprops={'edgecolor':'black'})
plt.title('Most Popular Genres in Video Games')
plt.tight_layout()
plt.show()

t  = sns.pairplot(df)
t 
#analyse the paiplots

df['Critic_Score']

df['Critic_Score'].isna().sum() #lots of values in Critic score

df['Critic_Score'].median()

#Replace Nan values in Critic_score with median Values
df['Critic_Score'].fillna(71, inplace=True)

df['Critic_Score'].max()

filtera = (df['Critic_Score']==df['Critic_Score'].max())

df['Name'][filtera] #best games according to critic scores

"""So GTA, Tony Hawk's Pro Skater 2 and SoulCalibur dominate when it comes to Critics ratings.So these are the games you should
loook out for next time you go game shopping"""

df['Rating'].value_counts()

df['Rating'].isna().sum() #Nan values which create the problems

games_with_no_rating  = df['Rating'].isna()
df['Name'][games_with_no_rating] 
#the games with no ratings are mostly normal games which everyone can play and not adult games

#Replace Nan values with E Rating i.e Everyone can play it
df['Rating'].fillna('E', inplace=True)
df['Rating'].value_counts()
list(df['Rating'].value_counts())

from collections import Counter
a = list(df['Rating'])
letter_counts = Counter(a)
d = pd.DataFrame.from_dict(letter_counts, orient='index')


d.plot(kind='bar')

#In the above Barplot E rating dominates which signifies most games made in industries are for people of all ages.

df

df['Year_of_Release'] = pd.to_datetime(df['Year_of_Release'], format='%Y')

df['Year_of_Release'].min()

df['Year_of_Release'].max()