ShrashtiSinghal

S.No.,	  Group Name,	                    Group Link,                                              Number of Members in K*,	 Social Media,	 Average Rating
3,	      Data Science - R & Python,	    https://www.facebook.com/groups/AnalyticsEdge/,                                                                      217,	  Facebook,	   88.7
5,	      Data Science,	                       https://www.facebook.com/groups/DataScienceGroup/,                                                               137,	 Facebook	,   88.6
15,	    Data Science with Python,	     https://www.facebook.com/groups/1006538092836222/,                                                                     56.5,	Facebook,	84.5
6,	      Machine Learning and Data Science,	   https://www.linkedin.com/groups/4298680/,                                                               128,	Linkedin,	68.34
13,	    Python Data Science and Machine Learning,	    https://www.linkedin.com/groups/4388870/,                                                      63,	L

ShrashtiSinghal

#Cross Validation

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import RepeatedKFold
from sklearn.model_selection import cross_val_score
import seaborn as sns
from sklearn.linear_model import LinearRegression, Lasso,ElasticNet, Ridge, MultiTaskLasso, LassoLars, OrthogonalMatchingPursuit
from sklearn.model_selection import train_test_split

ShrashtiSinghal

#Regularization

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

from sklearn.linear_model import LinearRegression, Lasso,ElasticNet, Ridge, MultiTaskLasso, LassoLars, OrthogonalMatchingPursuit
from sklearn.model_selection import train_test_split
from sklearn import metrics

ShrashtiSinghal

#Test Train Split

import numpy as np
import pandas as pd
from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
from sklearn import metrics

class predit:
    def bestFitLine(self):

ShrashtiSinghal

#Log Transformation to remove skewness
y =  np.log1p(y)
for col in x.columns:
    if np.abs(x[col].skew()) > 0.3:
        x[col] = np.log1p(x[col])

ShrashtiSinghal

#Scale all the attributes except RAD, CHAS, ZN
from sklearn import preprocessing
min_max_scaler = preprocessing.MinMaxScaler()
header_new = ['CRIM', 'INDUS', 'NOX', 'RM', 'AGE','DIS','TAX', 'PTRATIO','B','LSTAT']
x = data.loc[:,header_new]
y = data['MEDV']
x = pd.DataFrame(data=min_max_scaler.fit_transform(x), columns=header_new)

ShrashtiSinghal

   #Find Percentage of outliers in every column
   for k, v in data.items():
        q1 = v.quantile(0.25)
        q3 = v.quantile(0.75)
        IQR = q3 - q1
        v_col = v[(v <= q1 - 1.5 * IQR) | (v >= q3 + 1.5 * IQR)]
        percentage = np.shape(v_col)[0] * 100.0 / np.shape(data)[0]
        print("Column %s outliers = %.2f%%" % (k, percentage))

ShrashtiSinghal

#HeatMap to find the corelated features
plt.figure(figsize=(25, 10))
sns.heatmap(data.corr().abs(),  annot=True)

ShrashtiSinghal

#Visualze Scale of the data
import seaborn as sns
import matplotlib.pyplot as plt

fig,ax = plt.subplots(figsize=(15, 7))
ax.set_xscale("log")
for k,v in data.items():
    sns.boxplot(data=df, orient='h')
plt.grid()
plt.tight_layout(pad=0.5, w_pad=0.7, h_pad=5.0)

ShrashtiSinghal

#Visualize Box plots for outliers
fig, axs = plt.subplots(ncols=5, nrows=3, figsize=(20, 10))
index = 0
axs = axs.flatten()
for k,v in data.items():
    sns.boxplot(x=k, data=data, ax=axs[index], color="orange")
    index += 1
plt.tight_layout(pad=0.4, w_pad=0.5, h_pad=5.0)
data = data[~(data['MEDV'] >= 50.0)]
print(np.shape(data))