def testCNN1DKeras(mult_donnees,mesDonnees_test):
print("debut test conv1D")
def f1(y_true, y_pred):
def recall(y_true, y_pred):
"""Recall metric.
Only computes a batch-wise average of recall.
Computes the recall, a metric for multi-label classification of
how many relevant items are selected.
"""
true_positives = K.sum(K.round(K.clip(y_true * y_pred, 0, 1)))
possible_positives = K.sum(K.round(K.clip(y_true, 0, 1)))
recall = true_positives / (possible_positives + K.epsilon())
return recall
def precision(y_true, y_pred):
"""Precision metric.
Only computes a batch-wise average of precision.
Computes the precision, a metric for multi-label classification of
how many selected items are relevant.
"""
true_positives = K.sum(K.round(K.clip(y_true * y_pred, 0, 1)))
predicted_positives = K.sum(K.round(K.clip(y_pred, 0, 1)))
precision = true_positives / (predicted_positives + K.epsilon())
return precision
precision = precision(y_true, y_pred)
recall = recall(y_true, y_pred)
return 2*((precision*recall)/(precision+recall+K.epsilon()))
def create_model():
# create model
model = Sequential()
model.add(Conv1D(4, 10, strides=2, padding='same', activation='relu'))
model.add(Dropout(0.4))
model.add(MaxPooling1D(3))
model.add(Conv1D(40, 5, strides=2, padding='same', activation='relu'))
model.add(Flatten())
model.add(Dense(2, activation='softmax'))
# Compile model
model.compile(loss='binary_crossentropy', optimizer='SGD', metrics=['binary_accuracy'])
return model
result=[]
for r,donnees in mult_donnees.items():
print(donnees.data.shape)
print("r = "+str(r))
model = KerasClassifier(build_fn=create_model, epochs=1, batch_size=1000)
m = GenericModele(model,donnees)
s = m.f1Score(mode="proba")
result.append([r,s])
print("reactionTime,f1score : "+str(result))
return result
def testRiemannMDM(mult_donnees):
print("debut test riemann MDM")
result = []
for r, donnees in mult_donnees.items():
print("r = " + str(r))
m = GenericModele(pyriemann.classification.MDM(), donnees)
m.dataToCov()
s = m.f1Score()
result.append([r, s])
print("reactionTime,f1score : " + str(result))
return result
def testKNNBrut(mult_donnees):
print("debut test knn brut")
result = []
for r, donnees in mult_donnees.items():
print("r = " + str(r))
neigh = KNeighborsClassifier(n_neighbors=3)
m = GenericModele(neigh, donnees)
m.vectorize()
s = m.f1Score()
result.append([r, s])
print("reactionTime,f1score : " + str(result))
return result
def testSVMBrut(mult_donnees):
print("debut test SVM brut")
result = []
for r, donnees in mult_donnees.items():
print("r = " + str(r))
clf = SVC(gamma='auto', probability=True, max_iter=100, verbose=1)
m = GenericModele(clf, donnees)
m.vectorize()
s = m.f1Score()
result.append([r, s])
print("reactionTime,f1score : " + str(result))
return result
def testCovSVM(mult_donnees,mesDonnees_test):
print("debut test cov SVM")
result=[]
for r,donnees in mult_donnees.items():
print("r = "+str(r))
clf = SVC(gamma='auto',probability=True,max_iter=100,verbose=1)
m = GenericModele(clf,donnees)
m.dataToCov()
m.vectorize()
m.fit()
s = m.score(mesDonnees_test[r].data,mesDonnees_test[r].labels)
result.append([r,s])
print("reactionTime,f1score : "+str(result))
return result
def testKNNTF(mult_donnees,mesDonnees_test):
print("debut test knn tf")
result=[]
for r,donnees in mult_donnees.items():
print("r = "+str(r))
neigh = KNeighborsClassifier(n_neighbors=3)
m = GenericModele(neigh,donnees)
m.dataToTf()
m.vectorize(tf=True)
m.fit()
s = m.score(mesDonnees_test[r].data,mesDonnees_test[r].labels)
result.append([r,s])
print("reactionTime,f1score : "+str(result))
return result
def testKNNPaseBas(nbPoint,slider,mult_donnees,mesDonnees_test):
print("debut test passe bas knn")
result=[]
for r,donnees in mult_donnees.items():
print("r = "+str(r))
neigh = KNeighborsClassifier(n_neighbors=3)
m = GenericModele(neigh,donnees)
m.vectorize()
m.dataToMoy(nbPoint,slider)
m.fit()
s = m.score(mesDonnees_test[r].data,mesDonnees_test[r].labels)
result.append([r,s])
print("reactionTime,f1score : "+str(result))
return result
def testRiemannKNN(mult_donnees,mesDonnees_test):
print("debut test riemann KNN")
result=[]
for r,donnees in mult_donnees.items():
print("r = "+str(r))
m = GenericModele(pyriemann.classification.KNearestNeighbor(n_neighbors=3),donnees)
m.dataToCov()
m.fit()
s = m.score(mesDonnees_test[r].data,mesDonnees_test[r].labels)
result.append([r,s])
print("reactionTime,f1score : "+str(result))
return result
def testRiemannMDMPlusXdawn(mult_donnees,mesDonnees_test):
print("debut test riemann MDM")
result=[]
for r,donnees in mult_donnees.items():
print("r = "+str(r))
m = GenericModele(pyriemann.classification.MDM(),donnees)
m.dataToXdawnCov()
m.fit()
s = m.score(mesDonnees_test[r].data,mesDonnees_test[r].labels)
result.append([r,s])
print("reactionTime,f1score : "+str(result))
return result
def testSVMPaseBas(nbPoint, slider, mult_donnees):
print("debut test passe bas SVM")
result = []
for r, donnees in mult_donnees.items():
print("r = " + str(r))
clf = SVC(gamma='auto', probability=True, max_iter=100, verbose=1)
m = GenericModele(clf, donnees)
m.vectorize()
m.dataToMoy(nbPoint, slider)
s = m.f1Score()
result.append([r, s])
print("reactionTime,f1score : " + str(result))
return result
myFileList = ["data/subject1/Session1/1.gdf"]
reactTimeToTest = [1, 0.1, 0.04]
mesDonnees = dict()
for r in reactTimeToTest:
mesDonnees[r] = dataLoader.DataLoader(myFileList, r)
os.mkdir("resultats/TFKNN_valeurs_k")
for r, donnees in mesDonnees.items():
print(r)
result = []
for k in range(1, 20, 2):
neigh = KNeighborsClassifier(n_neighbors=k)
m = GenericModele(neigh, donnees)
m.dataToTf()
m.vectorize(tf=True)
m.dataToMoy(nbPoint, slider)
s = m.f1Score()
result.append([k, s])
plt.clf()
f = plt.figure()
plt.title("F1 score en fonction de la valeur de k")
plt.xlabel("Valeur de k")
plt.ylabel("F1 score")
plt.plot([re[0] for re in result], [re[1] for re in result])
name = "resultats/TFKNN_valeurs_k/KNN_tf_valeur_de_k_pour_time_" + str(r)
f.savefig(name + ".pdf", bbox_inches='tight')
saveResult(name + ".csv", result)
import pyriemann
import csv
def saveResult(name, result):
myfile = open(name, 'w')
wr = csv.writer(myfile, quoting=csv.QUOTE_ALL)
wr.writerow(["Time(second)", "F1 Score"])
for r in result:
wr.writerow(r)
reactTimeToTest = [1, 0.1, 0.04]
for r in reactTimeToTest:
print(r)
result = []
for k in range(1, 50, 2):
m = GenericModele(
r, pyriemann.classification.KNearestNeighbor(n_neighbors=k))
m.load_data_from_file("data/subject1/Session1/1.gdf")
m.dataToCov()
s = m.f1Score()
result.append([k, s])
plt.clf()
f = plt.figure()
plt.plot([re[0] for re in result], [re[1] for re in result])
name = "resultats/riemannKNN_valeur_de_k_pour_time_" + str(r)
f.savefig(name + ".pdf", bbox_inches='tight')
saveResult(name + ".csv", result)
wr.writerow(r)
myFileList = ["data/subject1/Session1/1.gdf"]
reactTimeToTest = [1, 0.1, 0.04]
mesDonnees = dict()
for r in reactTimeToTest:
mesDonnees[r] = dataLoader.DataLoader(myFileList, r)
os.mkdir("resultats/rieamannKNN_valeurs_k")
for r, donnees in mesDonnees.items():
print(r)
result = []
for k in range(1, 20, 2):
m = GenericModele(
pyriemann.classification.KNearestNeighbor(n_neighbors=k), donnees)
m.dataToCov()
s = m.f1Score()
print([k, s])
result.append([k, s])
plt.clf()
f = plt.figure()
plt.title("F1 score en fonction de la valeur de k")
plt.xlabel("Valeur de k")
plt.ylabel("F1 score")
plt.plot([re[0] for re in result], [re[1] for re in result])
name = "resultats/rieamannKNN_valeurs_k/riemannKNN_valeur_de_k_pour_time_" + str(
r)
f.savefig(name + ".pdf", bbox_inches='tight')
saveResult(name + ".csv", result)
wr.writerow(r)
myFileList = ["data/subject1/Session1/1.gdf"]
reactTimeToTest = [1, 0.1, 0.04]
mesDonnees = dict()
for r in reactTimeToTest:
mesDonnees[r] = dataLoader.DataLoader(myFileList, r)
os.mkdir("resultats/BrutKNN_valeurs_k")
for r, donnees in mesDonnees.items():
print(r)
result = []
for k in range(1, 20, 2):
neigh = KNeighborsClassifier(n_neighbors=k)
m = GenericModele(neigh, donnees)
m.vectorize()
s = m.f1Score()
result.append([k, s])
plt.clf()
f = plt.figure()
plt.title("F1 score en fonction de la valeur de k")
plt.xlabel("Valeur de k")
plt.ylabel("F1 score")
plt.plot([re[0] for re in result], [re[1] for re in result])
name = "resultats/BrutKNN_valeurs_k/KNN_brut_valeur_de_k_pour_time_" + str(
r)
f.savefig(name + ".pdf", bbox_inches='tight')
saveResult(name + ".csv", result)
