Courses by Software
Courses by Semester
Courses by Domain
Tool-focused Courses
Machine learning
POPULAR COURSES
Post Graduate Program in Hybrid Electric Vehicle Design and AnalysisPost Graduate Program in Computational Fluid DynamicsPost Graduate Program in CADPost Graduate Program in CAEPost Graduate Program in Manufacturing DesignPost Graduate Program in Computational Design and Pre-processingPost Graduate Program in Complete Passenger Car Design & Product DevelopmentSuccess Stories
Supervised Learning - Classification Week 7 Challenge
1) Pros and cons of SVM SVM 1) SVM stands for support vector machines.SVM is simple and provides good accuracy with less computational power, SVM is also used for regression but is widely applied for classification projects. There are several hyperplanes possible to classify the data points, but the…
Sushant Ovhal
updated on 12 Oct 2022
1) Pros and cons of SVM
SVM
1) SVM stands for support vector machines.SVM is simple and provides good accuracy with less computational power, SVM is also used for regression but is widely applied for classification projects. There are several hyperplanes possible to classify the data points, but the optimal hyperplane is the one with the maximum margin.
2) The name support vector machines is because the data points closest to the hyperplane are called support vectors as these points decide the optimal position of hyperplanes. changing the position of the support vector will change the position orientation of the hyperplane and also these support vectors are used to maximize the margin.
1)Pros:
1)It works really well with a clear margin of separation
2)It is effective in high-dimensional spaces.
3) It is effective in cases where the number of dimensions is greater than the number of samples.
4) It uses a subset of training points in the decision function (called support vectors), so it is also memory efficient
2)Cons:
1) It doesn’t perform well when we have large data set because the required training time is higher
2) It also doesn’t perform very well when the data set has more noise i.e. target classes are overlapping
3)SVM doesn’t directly provide probability estimates, these are calculated using an expensive five-fold cross-validation. It is included in, the related SVC method of the Python scikit-learn library.
4) SVM algorithm is not suitable for the large data set
5) As the support vector classifier works by putting data points, above and below the classifying hyperplane there is no probabilistic, explanation for the classification.
2) Apply SVM to the “Surface defects in stainless steel plates” dataset and evaluate it
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sn
import scipy.stats as stats
steel = pd.read_csv( 'faults.csv')
steel.hist(figsize=(30,30))
plt.show()
print('steel head command = ', steel.head())
print('steel describe command = ',steel.describe())
# heatmap Visualization
sns.set(rc={'figure.figsize':(12,10)})
corr = steel.corr()
sns.heatmap(corr, xticklabels=corr.columns.values,yticklabels=corr.columns.values)
plt.show()
# Divide the dataset into features and faults
fault = steel.values
fault_data=steel[["Pastry","Z_Scratch","K_Scatch","Stains","Dirtiness","Bumps","Other_Faults"]]
features = fault[:,0:27]
x=pd.DataFrame(features)
print(fault_data)
from sklearn import svm
model=svm.SVC(C=1,gamma="auto",kernel="linear")
# Convert to single column
faults_single = []
for i in range(fault_data.shape[0]):
if fault_data["Pastry"].values[i]==1:
faults_single.append("Pastry")
elif fault_data["Z_Scratch"].values[i] == 1:
faults_single.append("Z_Scratch")
elif fault_data["K_Scatch"].values[i] == 1:
faults_single.append("K_Scatch")
elif fault_data["Stains"].values[i] == 1:
faults_single.append("Stains")
elif fault_data["Dirtiness"].values[i] == 1:
faults_single.append("Dirtiness")
elif fault_data["Bumps"].values[i] == 1:
faults_single.append("Bumps")
else:
faults_single.append("Other_fault")
# heatmap visualization
sns.set(rc={'figure.figsize':(12,10)})
corr = fault_data.corr()
sns.heatmap(corr,xticklabels=corr.columns.values,yticklabels =corr.columns.values)
plt.show()
# Covert to single column with transpose
fault_single_col = []
for i in range(fault_dataframe.shape[0]):
if fault_dataframe["Pastry"].values[i] == 1:
fault_single_col.append("Pastry")
elif fault_dataframe["Z_Scratch"].values[i] == 1:
fault_single_col.append("Z_Scratch")
elif fault_dataframe["K_Scatch"].values[i] == 1:
fault_single_col.append("K_Scatch")
elif fault_dataframe["Stains"].values[i] == 1:
fault_single_col.append("Stains")
elif fault_dataframe["Dirtiness"].values[i] == 1:
fault_single_col.append("Dirtiness")
elif fault_dataframe["Bumps"].values[i] == 1:
fault_single_col.append("Bumps")
else:
fault_single_col.append("Other_Faults")
# heatmap Visualization
sns.set(rc={'figure.figsize':(12,10)})
corr = fault_dataframe.corr()
sns.heatmap(corr, xticklabels=corr.columns.values,yticklabels=corr.columns.values)
plt.show()
fault_single =np.array(faults_single)
print(fault_single.shape)
faultstype = pd.DataFrame({'faults':fault_single})
print(faultstype)
# countplot visualization
fig, ax=plt.subplots(1,2,figsize=(20,8))
faultstype['faults'].value_counts().plot.pie(ax=ax[0])
sns.countplot(x='faults',data=faultstype, ax=ax[1])
plt.show()
sc=StandardScaler()
X=sc.fit_transform(x)
x_train, x_test,y_train,y_test =train_test_split(X, faultstype, test_size =0.40, random_state =1000)
from sklearn.svm import SVC
model =SVC(C=1,gamma ="auto",kernel ="linear")
print(model.fit(x_train,y_train))
print(model.score(x_test,y_test))
y_predict =model.predict(x_test)
from sklearn import metrics
cm=metrics.confusion_matrix(y_test,y_predict)
print(cm)
# heatmap visualization
sns.set(rc={'figure.figsize':(12,10)})
sns.heatmap(cm,annot=True)
plt.xlabel("predicted")
plt.ylabel("Actual")
plt.show()
Leave a comment
Thanks for choosing to leave a comment. Please keep in mind that all the comments are moderated as per our comment policy, and your email will not be published for privacy reasons. Please leave a personal & meaningful conversation.
Other comments...
Be the first to add a comment
Read more Projects by Sushant Ovhal (22)
Project 1 - Analyzing the Education trends in Tamilnadu
This dashboard empowers mission driven organizations to harness the power of data visualization for social change. Women are tracked away from science and mathematics throughout their education, limiting their training and options to go into these fields as adults. The data set contains the data of women graduated by years,…
14 Nov 2023 01:32 PM IST
Project 1 - English Dictionary App & Library Book Management System
Project 1) English dictionary app and Library Book Management system
06 Nov 2023 04:04 PM IST
Project 1 - Implement and deploy CNN model in real-time using python on Fashion MNIST dataset
Implement and deploy CNN model in real-time using python on Fashion MNIST dataset
20 Dec 2022 07:04 AM IST
Project 2
Project 2
30 Nov 2022 11:41 AM IST
Related Courses
0 Hours of Content
Skill-Lync offers industry relevant advanced engineering courses for engineering students by partnering with industry experts.
Our Company
4th Floor, BLOCK-B, Velachery - Tambaram Main Rd, Ram Nagar South, Madipakkam, Chennai, Tamil Nadu 600042.
Top Individual Courses
Top PG Programs
Skill-Lync Plus
Trending Blogs
© 2025 Skill-Lync Inc. All Rights Reserved.