In [34]:

    

from sklearn.svm import SVC
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score
from sklearn.preprocessing import normalize
import numpy as np
import pandas as pd


    

In [32]:

    

data=pd.read_csv("eeg_data.csv")
data.head()


    

    
Out[32]:






  

    

      

      
-0.16897227
      
-0.178321981
      
-0.095154471
      
-0.051145173
      
-0.024695237
      
0.006821866
      
0.146600514
      
0.049728133
      
-0.030760814
      
-0.030717532
      
...
      
0.784022832
      
0.506615997
      
0.657809798
      
0.726842459
      
0.044562985
      
0.169059196
      
0.267928391
      
0.438069021
      
-0.028649762
      
1
    
  
  

    

      
0
      
1.150855
      
1.023013
      
0.482823
      
1.808277
      
1.681753
      
0.369737
      
1.162185
      
2.948635
      
1.297460
      
1.556453
      
...
      
0.240170
      
0.612654
      
0.860099
      
0.984312
      
0.763681
      
1.023541
      
0.478984
      
0.591697
      
-0.018466
      
1
    
    

      
1
      
-0.239892
      
-0.286383
      
-0.117660
      
-0.041761
      
0.045529
      
0.350156
      
0.757991
      
0.672565
      
0.261040
      
0.249775
      
...
      
0.320247
      
1.314314
      
1.746596
      
1.972543
      
0.145621
      
0.192455
      
0.480254
      
1.275407
      
-0.019574
      
1
    
    

      
2
      
0.616870
      
0.935586
      
0.557368
      
0.506930
      
1.129597
      
0.015468
      
-0.019766
      
0.371247
      
0.159066
      
0.341238
      
...
      
2.490612
      
1.543079
      
1.936653
      
2.168959
      
1.326236
      
1.985246
      
1.321501
      
1.634900
      
-0.008216
      
1
    
    

      
3
      
-0.203458
      
-0.082669
      
0.101029
      
0.022260
      
0.554096
      
0.258584
      
0.295721
      
0.092840
      
0.010436
      
0.063246
      
...
      
1.729174
      
1.035423
      
1.461515
      
1.546399
      
0.365488
      
0.338945
      
1.001988
      
1.224796
      
-0.013884
      
1
    
    

      
4
      
0.594482
      
0.142026
      
0.020022
      
0.562168
      
0.201262
      
1.015226
      
1.649477
      
0.720609
      
0.267196
      
0.035037
      
...
      
0.831768
      
0.712272
      
0.949732
      
1.076258
      
0.202871
      
1.748063
      
0.576664
      
0.627897
      
-0.025558
      
1
    
  

5 rows × 1345 columns

In [41]:

    

Norm_feature=normalize(data.loc[:,data.columns != '1'], norm='l2', axis=1, copy=True)
labels=np.array(data.loc[:,data.columns == '1'])


    

In [123]:

    

print "Total feature ", len(Norm_feature[1])
print "Total data ", len(Norm_feature),len(labels)


    

    




Total feature  1344
Total data  2669 2669

In [95]:

    

train_x=np.concatenate((Norm_feature[:730],Norm_feature[800:2061],Norm_feature[2150:]),axis=0)
train_y=np.concatenate((labels[:730],labels[800:2061],labels[2150:]),axis=0)

test_x=np.concatenate((Norm_feature[700:801],Norm_feature[2060:2150]),axis=0)
test_y=np.concatenate((labels[700:801],labels[2060:2150]))


    

In [96]:

    

print "training data",len(train_x)
print "testing data",len(test_x)


    

    




training data 2510
testing data 191

In [97]:

    

# clsf = classifier
clsf=SVC(kernel='rbf',gamma=.8,C=4) #SVM Classier
svm={} #store accuracy data for plotting


    

In [98]:

    

#training
clsf.fit(train_x, train_y)


    

    
Out[98]:





SVC(C=4, cache_size=200, class_weight=None, coef0=0.0,
  decision_function_shape=None, degree=3, gamma=0.8, kernel='rbf',
  max_iter=-1, probability=False, random_state=None, shrinking=True,
  tol=0.001, verbose=False)

In [99]:

    

#Now testing for trainig data
#Now predict values for given classifier
prediction = clsf.predict(train_x)
svm["trainig_set"]=accuracy_score(prediction,train_y)*100
print 'Accuracy Check ',svm["trainig_set"],'%'


    

    




Accuracy Check  96.2549800797 %

In [100]:

    

prediction = clsf.predict(test_x)
svm["test_set"]=accuracy_score(prediction,test_y)*100
print 'Accuracy Check ',svm["test_set"],'%'


    

    




Accuracy Check  70.1570680628 %

In [101]:

    

svm


    

    
Out[101]:





{'test_set': 70.157068062827221, 'trainig_set': 96.254980079681275}

In [102]:

    

decision={}
clf=DecisionTreeClassifier(max_depth=10,min_samples_split=4)
clf


    

    
Out[102]:





DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=10,
            max_features=None, max_leaf_nodes=None, min_samples_leaf=1,
            min_samples_split=4, min_weight_fraction_leaf=0.0,
            presort=False, random_state=None, splitter='best')

In [103]:

    

#train
clf = clf.fit(train_x, train_y)


    

In [104]:

    

#Now testing for trainig data
#Now predict values for given classifier
prediction = clf.predict(train_x)
decision["trainig_set"]=accuracy_score(prediction,train_y)*100
print 'Accuracy Check ',decision["trainig_set"],'%'


    

    




Accuracy Check  96.4940239044 %

In [105]:

    

prediction = clf.predict(test_x)
decision["test_set"]=accuracy_score(prediction,test_y)*100
print 'Accuracy Check ',decision["test_set"],'%'


    

    




Accuracy Check  62.3036649215 %

In [106]:

    

decision


    

    
Out[106]:





{'test_set': 62.303664921465973, 'trainig_set': 96.494023904382473}

In [108]:

    

from sklearn.neural_network import MLPClassifier


    

In [109]:

    

neural=dict()
clf = MLPClassifier(algorithm='l-bfgs',hidden_layer_sizes=(50,), alpha=1e-5, random_state=1)
clf


    

    
Out[109]:





MLPClassifier(activation='relu', algorithm='l-bfgs', alpha=1e-05,
       batch_size=200, beta_1=0.9, beta_2=0.999, early_stopping=False,
       epsilon=1e-08, hidden_layer_sizes=(50,), learning_rate='constant',
       learning_rate_init=0.001, max_iter=200, momentum=0.9,
       nesterovs_momentum=True, power_t=0.5, random_state=1, shuffle=True,
       tol=0.0001, validation_fraction=0.1, verbose=False,
       warm_start=False)

In [110]:

    

#trainig Network
clf.fit(train_x, train_y)


    

    




/home/sudhanshu/anaconda2/lib/python2.7/site-packages/sklearn/neural_network/multilayer_perceptron.py:894: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().
  y = column_or_1d(y, warn=True)






    
Out[110]:





MLPClassifier(activation='relu', algorithm='l-bfgs', alpha=1e-05,
       batch_size=200, beta_1=0.9, beta_2=0.999, early_stopping=False,
       epsilon=1e-08, hidden_layer_sizes=(50,), learning_rate='constant',
       learning_rate_init=0.001, max_iter=200, momentum=0.9,
       nesterovs_momentum=True, power_t=0.5, random_state=1, shuffle=True,
       tol=0.0001, validation_fraction=0.1, verbose=False,
       warm_start=False)

In [112]:

    

#Now testing for trainig data
#Now predict values for given classifier
prediction = clf.predict(train_x)
neural["trainig_set"]=accuracy_score(prediction,train_y)*100
print 'Accuracy Check ',neural["trainig_set"],'%'


    

    




Accuracy Check  98.4860557769 %

In [114]:

    

prediction = clf.predict(test_x)
neural["test_set"]=accuracy_score(prediction,test_y)*100
print 'Accuracy Check ',neural["test_set"],'%'


    

    




Accuracy Check  71.7277486911 %

In [121]:

    

import pylab as plt
plt.rcParams['figure.figsize'] = 16, 12


    

In [122]:

    

#%pylab inline
classifier = [1,2,3]
accuracy_test_data=[svm["test_set"],decision["test_set"],neural["test_set"]]
LABELS=["SVM","Decision-Tree","Neural-net"]

plt.bar(classifier,accuracy_test_data,align='center')
plt.xticks(classifier, LABELS) #binding label with x axis data
plt.xlabel('Classifier')
plt.ylabel('Accuracy')
plt.title('Analysis with 2 Class')
plt.show()


    

In [ ]: