import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
data = pd.read_csv("C:/Users/Rahul/Desktop/Capstone papers/Reviews.csv")
data.shape
data.head()
df_count_prcnt = data.Score.value_counts()
print(df_count_prcnt)
print(df_count_prcnt).sum()
def compute_percentage(x):
pct = float(x/df_count_prcnt.sum()) * 100
return round(pct, 2)
score_prnct = compute_percentage(df_count_prnct)
df_count_prcnt = data.Score.value_counts()
def compute_percentage(x):
pct = (x/df_count_prcnt.sum()) * 100
return pct
score_prcnt = compute_percentage(df_count_prcnt)
print(score_prcnt)
score_prcnt.plot(kind="bar", colormap='jet')
data['datetime'] = pd.to_datetime(data["Time"], unit='s')
data_grp = data.groupby([data.datetime.dt.year, data.datetime.dt.month, data.Score]).count()['ProductId'].unstack().fillna(0)
data_grp.plot(figsize=(20,10), rot=45, colormap='jet')
data_grp.plot(kind="bar",figsize=(30,10), stacked=True, colormap='jet')