162 lines
5.5 KiB
Python
Executable file
162 lines
5.5 KiB
Python
Executable file
#!/usr/bin/env python3
|
|
# -*- coding: utf-8 -*-
|
|
"""
|
|
Created on Thu May 26 05:19:25 2022
|
|
|
|
@author: tanu
|
|
"""
|
|
#%% https://www.kite.com/blog/python/smote-python-imbalanced-learn-for-oversampling/
|
|
import matplotlib.pyplot as plt
|
|
import numpy as np
|
|
import pandas as pd
|
|
from sklearn.svm import SVC
|
|
from imblearn.over_sampling import SMOTE
|
|
#%%############################################################################
|
|
|
|
def train_SVM(df):
|
|
# select the feature columns
|
|
X = df.loc[:, df.columns != 'label']
|
|
# select the label column
|
|
y = df.label
|
|
|
|
# train an SVM with linear kernel
|
|
clf = SVC(kernel='linear')
|
|
clf.fit(X, y)
|
|
|
|
return clf
|
|
|
|
|
|
def plot_svm_boundary(clf, df, title):
|
|
fig, ax = plt.subplots()
|
|
X0, X1 = df.iloc[:, 0], df.iloc[:, 1]
|
|
|
|
x_min, x_max = X0.min() - 1, X0.max() + 1
|
|
y_min, y_max = X1.min() - 1, X1.max() + 1
|
|
xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.02), np.arange(y_min, y_max, 0.02))
|
|
|
|
Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])
|
|
Z = Z.reshape(xx.shape)
|
|
out = ax.contourf(xx, yy, Z, cmap=plt.cm.coolwarm, alpha=0.8)
|
|
|
|
ax.scatter(X0, X1, c=df.label, cmap=plt.cm.coolwarm, s=20, edgecolors='k')
|
|
ax.set_ylabel('y')
|
|
ax.set_xlabel('x')
|
|
ax.set_title(title)
|
|
plt.show()
|
|
#%% SIMPLE RESAMPLING
|
|
############################################################################
|
|
# RESAMPLING
|
|
###############################################################################
|
|
#------------------------------
|
|
# Simple Random oversampling
|
|
# [Numerical + catgeorical]
|
|
#------------------------------
|
|
oversample = RandomOverSampler(sampling_strategy='minority')
|
|
X_ros, y_ros = oversample.fit_resample(X, y)
|
|
print(X_ros.shape) #228
|
|
|
|
#------------------------------
|
|
# Simple Random undersampling
|
|
# [Numerical + catgeorical]
|
|
#------------------------------
|
|
undersample = RandomUnderSampler(sampling_strategy='majority')
|
|
X_rus, y_rus = undersample.fit_resample(X, y)
|
|
print(X_rus.shape) #142
|
|
|
|
#------------------------------
|
|
# Simple ROS and RUS
|
|
# [Numerical + catgeorical]
|
|
#------------------------------
|
|
oversample = RandomOverSampler(sampling_strategy='minority')
|
|
X_ros, y_ros = oversample.fit_resample(X, y)
|
|
undersample = RandomUnderSampler(sampling_strategy='majority')
|
|
X_rouC, y_rouC = undersample.fit_resample(X_ros, y_ros)
|
|
print(X_rouC.shape) #228
|
|
###############################################################################
|
|
#%% SMOETE RESAMPLING
|
|
#------------------------------
|
|
# SMOTE: Oversampling
|
|
# [Numerical ONLY]
|
|
#------------------------------
|
|
k_sm = 1
|
|
sm = SMOTE(sampling_strategy = 'auto', k_neighbors = k_sm, **rs)
|
|
X_sm, y_sm = sm.fit_resample(X, y)
|
|
print(len(X_sm)) #228
|
|
#print(Counter(y))
|
|
y_sm_df = y_sm.to_frame()
|
|
y_sm_df.value_counts().plot(kind = 'bar')
|
|
|
|
#------------------------------
|
|
# SMOTE: Over + Undersampling COMBINED
|
|
# [Numerical ONLY]
|
|
#-----------------------------
|
|
sm_enn = SMOTEENN(enn=EditedNearestNeighbours(sampling_strategy='all', **rs, **njobs ))
|
|
X_enn, y_enn = sm_enn.fit_resample(X, y)
|
|
print(len(X_enn)) #53
|
|
|
|
#TO TRY
|
|
# sm_etomek = SMOTETomek(*, sampling_strategy='all',tomek = None), **njobs )
|
|
# X_etomek , y_etomek = sm_etomek.fit_resample(X, y)
|
|
# print(len(X_etomek))
|
|
|
|
# k_renn = 3 # deafult======= BAAADDD
|
|
# sm_renn = RepeatedEditedNearestNeighbours(sampling_strategy='all'
|
|
# , n_neighbors = k_renn
|
|
# , max_iter =100
|
|
# , **njobs )
|
|
# X_renn, y_renn = sm_renn.fit_resample(X, y)
|
|
# print(len(X_renn)) #22
|
|
|
|
# check:
|
|
# SMOTEhttps://imbalanced-learn.org/dev/references/generated/imblearn.under_sampling.AllKNN.html#imblearn.under_sampling.AllKNN
|
|
###############################################################################
|
|
#------------------------------
|
|
# SMOTE_NC: oversampling
|
|
# [numerical + categorical]
|
|
#https://stackoverflow.com/questions/47655813/oversampling-smote-for-binary-and-categorical-data-in-python
|
|
|
|
#------------------------------
|
|
# Determine categorical and numerical features
|
|
numerical_ix = X.select_dtypes(include=['int64', 'float64']).columns
|
|
numerical_ix
|
|
num_featuresL = list(numerical_ix)
|
|
numerical_colind = X.columns.get_indexer(list(numerical_ix) )
|
|
numerical_colind
|
|
|
|
categorical_ix = X.select_dtypes(include=['object', 'bool']).columns
|
|
categorical_ix
|
|
categorical_colind = X.columns.get_indexer(list(categorical_ix))
|
|
categorical_colind
|
|
|
|
k_sm = 5 # 5 is deafult
|
|
sm_nc = SMOTENC(categorical_features=categorical_colind, k_neighbors = k_sm, **rs, **njobs)
|
|
X_smnc, y_smnc = sm_nc.fit_resample(X, y)
|
|
print(len(X_smnc)) #228
|
|
|
|
#%%############################################################################
|
|
# FIXME : if that!
|
|
# SMOTE: Over+undersampling + [num+categorical]
|
|
# THIS WILL ONLY work if you handle the ohe separately for categ and find a way
|
|
# of inverse transforming it!
|
|
# t = [ ('cat', OneHotEncoder(), categorical_ix) ]
|
|
|
|
# col_transform = ColumnTransformer(transformers = t
|
|
# , remainder='passthrough')
|
|
# Xm = col_transform.fit_transform(X)
|
|
# Xm_colnames = col_transform.get_feature_names_out()
|
|
# Xmcolnames = pd.Index(Xm_colnames)
|
|
# Xmcolnames
|
|
|
|
# sm_ennC = SMOTEENN(enn=EditedNearestNeighbours(sampling_strategy='all'), **rs, **njobs )
|
|
# X_ennC, y_ennC = sm_ennC.fit_resample(Xm, y)
|
|
# print(X_ennC.shape)
|
|
|
|
# ohe = OneHotEncoder()
|
|
# ohe.fit(X)
|
|
|
|
# Xm = ohe.fit_transform(X[categorical_ix])
|
|
# print(Xm.shape)
|
|
|
|
# XmDF = pd.DataFrame(Xm.toarray())
|
|
#%%############################################################################
|
|
# TODO: Find over and undersampling JUST for categorical data
|