TO START
# IMPORT DATA LIBRARIES
import pandas as pd
import numpy as np
# IMPORT VIS LIBRARIES
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline
# IMPORT MODELLING LIBRARIES
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import classification_report
|
PRELIMINARY OPERATIONS
df.pd.read_csv('data.csv') |
read da |
df.head() |
check head df |
df.info() |
check info df |
df.describe() |
check stats df |
df.columns |
check col names |
VISUALISE DATA
sns.heatmap(df.isnull())* |
check null values |
sns.set_style('whitegrid') |
set different style |
sns.countplot('col',df) |
countplot |
sns.countplot('col',df,palette='') |
countplot |
sns.countplot('col',df,hue='',palette='') |
countplot |
sns.distplot(df['col'].dropna(),bins=30) |
distribution plot |
sns.heatmap(): can take more useful parameters;
yticklabels=False,cbar=False,cmap='viridis'
DATA CLEANING
create a personalise function* |
impute values |
apply the personalised function* |
apply function |
dummy_var =
pd.get_dummies(df['col'],drop_first=True)* |
convert categorical features |
df.drop(['old.col1',...'']) |
drop old columns |
df= pd.concat([dummy_var],axis=1) |
add the new dummy var into the df |
See imputing and apply section.
drop.first=True: without this command, we would have two specular columns, leading to issues of multicollinearity.
|
|
IMPUTING AND APPLY
# EXAMPLE OF A POSSIBLE FUNCTION TO IMPUTE MISSING VALUES
def impute_age(cols):
Age = cols[0]
Pclass = cols[1]
if pd.isnull(Age):
if Pclass == 1:
return 37
elif Pclass == 2:
return 29
else:
return 24
else:
return Age
# EXAMPLE OF HOW TO APPLY THE FUNCTION
train['Age'] = rain[['Age','Pclass']].apply(impute_age,axis=1)
|
TRAIN and EVALUATE MODEL
CREATE X and y |
X = df[['col1','col2',etc.]] |
create df features |
y = df['col'] |
create df var to predict |
SPLIT DATASET |
X_train, X_test, y_train, y_test =
train_test_split(
X,
y,
test_size=0.3) |
split df in train
and test df |
FIT THE MODEL |
log = LogisticRegression() |
instatiate model |
log.fit(X_train,y_train) |
train/fit the model |
MAKE PREDICTIONS |
predictions = log.predict(X_test) |
make predictions |
EVAUATE MODEL |
print(classification_report(y_test,predictions)) |
useful measures |
confusion_matrix(y_test, predictions) |
|
