f.write(str(['Test loss: ', score[0]]))
f.write('\n')
f.write(str(['Test accuracy: ', score[1]]))
f.write('\n')
confusion = []
precision = []
recall = []
f1s = []
kappa = []
auc = []
roc = []
scores = np.array([np.argmax(t) for t in np.asarray(model.predict(X_testNP))])
predict = np.array([np.argmax(t) for t in np.round(np.asarray(model.predict(X_testNP)))])
targ = np.array([np.argmax(t) for t in Y_test])
confusion.append(sklm.confusion_matrix(targ.flatten(), predict.flatten()))
precision.append(sklm.precision_score(targ.flatten(), predict.flatten(), average = 'macro'))
recall.append(sklm.recall_score(targ.flatten(), predict.flatten(), average = 'macro'))
f1s.append(sklm.f1_score(targ.flatten(), predict.flatten(), average = 'macro'))
kappa.append(sklm.cohen_kappa_score(targ.flatten(), predict.flatten()))
confusion.append(sklm.confusion_matrix(targ.flatten(), predict.flatten()))
f.write(str(['Area Under ROC Curve (AUC): ', auc]))
f.write('\n')
f.write('Confusion: ')
f.write('\n')
if K.image_dim_ordering() == 'th':
# Transpose image dimensions (Theano uses the channels as the 1st dimension)
im = im.transpose((2, 0, 1))
# Use pre-trained weights for Theano backend
weights_path = '../imagenet_models/densenet169_weights_th.h5'
else:
# Use pre-trained weights for Tensorflow backend
weights_path = '../imagenet_models/densenet169_weights_tf.h5'
# Insert a new dimension for the batch_size
im = np.expand_dims(im, axis=0)
# Test pretrained model
model = DenseNet(reduction=0.5, classes=1000, weights_path=weights_path)
sgd = SGD(lr=1e-2, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(optimizer=sgd,
loss='categorical_crossentropy',
metrics=['accuracy'])
out = model.predict(im)
# Load ImageNet classes file
classes = []
with open('resources/classes.txt', 'r') as list_:
for line in list_:
classes.append(line.rstrip('\n'))
print('Prediction: ' + str(classes[np.argmax(out)]))
weights_path = 'imagenet_models/densenet169_weights_th.h5'
else:
# Use pre-trained weights for Tensorflow backend
weights_path = 'imagenet_models/densenet169_weights_tf.h5'
# Insert a new dimension for the batch_size
im = np.expand_dims(im, axis=0)
# Test pretrained model
model = DenseNet(reduction=0.5, classes=10, weights_path=weights_path)
# Learning rate is changed to 1e-3
sgd = SGD(lr=1e-2, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(optimizer=sgd, loss='categorical_crossentropy', metrics=['accuracy'])
out = model.predict(im)
# Load ImageNet classes file
classes = []
with open('resources/classes.txt', 'r') as list_:
for line in list_:
classes.append(line.rstrip('\n'))
print 'Prediction: '+str(classes[np.argmax(out)])
from keras.datasets import cifar10
(X_train, Y_train), (X_test, Y_test) = cifar10.load_data()
img_size = 32
img_chan = 3
n_classes = 10
f.write(str(['Test loss: ', score[0]]))
f.write('\n')
f.write(str(['Test accuracy: ', score[1]]))
f.write('\n')
confusion = []
precision = []
recall = []
f1s = []
kappa = []
auc = []
roc = []
scores = np.asarray(model.predict(X_test))
predict = np.round(np.asarray(model.predict(X_test)))
targ = Y_test
auc.append(sklm.roc_auc_score(targ.flatten(), scores.flatten()))
confusion.append(sklm.confusion_matrix(targ.flatten(), predict.flatten()))
precision.append(sklm.precision_score(targ.flatten(), predict.flatten()))
recall.append(sklm.recall_score(targ.flatten(), predict.flatten()))
f1s.append(sklm.f1_score(targ.flatten(), predict.flatten()))
kappa.append(sklm.cohen_kappa_score(targ.flatten(), predict.flatten()))
f.write(str(['Area Under ROC Curve (AUC): ', auc]))
f.write('\n')
f.write('Confusion: ')
f.write('\n')
f.write(str(np.array(confusion)))
f.write(str(['Test loss: ', score[0]]))
f.write('\n')
f.write(str(['Test accuracy: ', score[1]]))
f.write('\n')
confusion = []
precision = []
recall = []
f1s = []
kappa = []
auc = []
roc = []
scores = np.array(
[np.argmax(t) for t in np.asarray(model.predict(X_test))])
predict = np.array(
[np.argmax(t) for t in np.round(np.asarray(model.predict(X_test)))])
targ = np.array([np.argmax(t) for t in Y_test])
auc.append(sklm.roc_auc_score(targ.flatten(), scores.flatten()))
confusion.append(sklm.confusion_matrix(targ.flatten(), predict.flatten()))
precision.append(sklm.precision_score(targ.flatten(), predict.flatten()))
recall.append(sklm.recall_score(targ.flatten(), predict.flatten()))
f1s.append(sklm.f1_score(targ.flatten(), predict.flatten()))
kappa.append(sklm.cohen_kappa_score(targ.flatten(), predict.flatten()))
confusion.append(sklm.confusion_matrix(targ.flatten(), predict.flatten()))
f.write(str(['Area Under ROC Curve (AUC): ', auc]))
f.write('\n')