signal_dct_02.append( dct ( signal[window*WINDOW_SIZE-WINDOW_DELAY:window*WINDOW_SIZE] ) )
signal_pca = PCA ( dct ( signal[window*WINDOW_SIZE-WINDOW_DELAY:window*WINDOW_SIZE] ) )
for vector in zip ( signal_pca.a.T[0], signal_pca.a.T[1], signal_pca.a.T[2] ):
if activity == 1:
print >>ACTIVITY_01, vector[0], vector[1], vector[2]
else:
print >>ACTIVITY_02, vector[0], vector[1], vector[2]
print >>training_output_file, str(vector[0]) + '\t' + str(vector[1]) + '\t' + str(vector[2]) + '\t' + str(signal_class)
print >>testing_output_file, str(vector[0]) + '\t' + str(vector[1]) + '\t' + str(vector[2])
# FANCY STUFF
mlx, mly = mlpy.data_fromfile('data_from_analyzer_01/train_data1.dat');
svm = mlpy.Svm(kernel='gaussian', C= 0.9);
svm.compute(mlx, mly)
mltx, mlty = mlpy.data_fromfile('data_from_analyzer_01/train_data1.dat');
for vector in mltx:
if svm.predict(vector) == 1:
print >>SVM_PREDICT_01, vector[0], vector[1], vector[2]
else:
print >>SVM_PREDICT_02, vector[0], vector[1], vector[2]
for vector in signal_dct_01:
print >>SIGNAL_DCT_01, vector[0], vector[1], vector[2]
xtr = np.array([[7.0, 2.0, 3.0, 1.0], # first sample
[1.0, 2.0, 3.0, 2.0], # second sample
[2.0, 2.0, 2.0, 1.0], # third sample#
[2.0, 4.0, 2.0, 6.0],
[2.0, 2.0, 7.0, 9.0]])
print xtr
print np.size(xtr), np.shape(xtr), np.ndim(xtr), xtr.dtype
ytr = np.array([1, 2, 3, 1, 2]) # classes
print ytr
print np.size(ytr), np.shape(ytr), np.ndim(ytr), xtr.dtype
#Save and read data from disk
print mlpy.data_tofile('data_example.dat', xtr, ytr, sep=' ')
x, y = mlpy.data_fromfile('data_example.dat')
print x
print y
print "mlpy.data_normalize(x) = ", mlpy.data_normalize(x)
#mysvm = mlpy.Svm() # initialize Svm class
myknn = mlpy.Knn(k = 1) # initialize knn class
## initialize fda class
myfda = mlpy.Fda()
#print mysvm.compute(xtr, ytr) # compute SVM
print myknn.compute(xtr, ytr) # compute knn
print myfda.compute(xtr, ytr) # compute fda
print count
metadata = post['json']
data = json.loads(str(metadata))
newPost = getPost.parsePostData(data["records"][0])
negativity = float(values[3])
if negativity > 60.0:
g = open("trainingData.txt","a")
g.write(str(postid)+","+str(newPost.likeCount)+","+str(newPost.commentCount)+","+str(newPost.repostCount)+","+str("1")+"\n")
g.close()
else:
g = open("trainingData.txt","a")
g.write(str(postid)+","+str(newPost.likeCount)+","+str(newPost.commentCount)+","+str(newPost.repostCount)+","+str("0")+"\n")
g.close()
except Exception:
continue
g.close()
f.close()
#writePostIdsToFile()
#writePostInformationToTheFile("fortyPercentOrMorePosts.txt")
x, y = mlpy.data_fromfile('data.dat')
print "done"
[7.0, 2.0, 3.0, 1.0], # first sample
[1.0, 2.0, 3.0, 2.0], # second sample
[2.0, 2.0, 2.0, 1.0], # third sample#
[2.0, 4.0, 2.0, 6.0],
[2.0, 2.0, 7.0, 9.0]
])
print xtr
print np.size(xtr), np.shape(xtr), np.ndim(xtr), xtr.dtype
ytr = np.array([1, 2, 3, 1, 2]) # classes
print ytr
print np.size(ytr), np.shape(ytr), np.ndim(ytr), xtr.dtype
#Save and read data from disk
print mlpy.data_tofile('data_example.dat', xtr, ytr, sep=' ')
x, y = mlpy.data_fromfile('data_example.dat')
print x
print y
print "mlpy.data_normalize(x) = ", mlpy.data_normalize(x)
#mysvm = mlpy.Svm() # initialize Svm class
myknn = mlpy.Knn(k=1) # initialize knn class
## initialize fda class
myfda = mlpy.Fda()
#print mysvm.compute(xtr, ytr) # compute SVM
print myknn.compute(xtr, ytr) # compute knn
print myfda.compute(xtr, ytr) # compute fda
