def Model(weights_path=None):
model = Sequential()
model.add(GaussianNoise(0.01, input_shape=(1, img_rows, img_cols)))
model.add(Convolution2D(nb_filters[0], nb_conv, nb_conv,
border_mode='valid'))
model.add(Activation('relu'))
model.add(Convolution2D(nb_filters[1], nb_conv, nb_conv))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(nb_pool, nb_pool)))
model.add(Dropout(0.25))
model.add(Convolution2D(nb_filters[2], nb_conv, nb_conv))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(nb_pool, nb_pool)))
model.add(Dropout(0.25))
model.add(Convolution2D(nb_filters[3], nb_conv, nb_conv))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(nb_pool, nb_pool)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))
if weights_path:
model.load_weights(weights_path)
return model
def CIFAR_10(img_rows, img_cols, img_channels=3, nb_classes=5, weights_path=None):
model = Sequential()
model.add(Convolution2D(32, 3, 3, border_mode='same',
input_shape=(img_channels, img_rows, img_cols)))
model.add(Activation('relu'))
model.add(Convolution2D(32, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Convolution2D(64, 3, 3, border_mode='same'))
model.add(Activation('relu'))
model.add(Convolution2D(64, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(512))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))
if weights_path:
model.load_weights(weights_path)
return model
def test_sequential_model_saving():
model = Sequential()
model.add(Dense(2, input_dim=3))
model.add(Dense(3))
model.compile(loss='mse', optimizer='rmsprop', metrics=['acc'])
x = np.random.random((1, 3))
y = np.random.random((1, 3))
model.train_on_batch(x, y)
out = model.predict(x)
fname = 'tmp_' + str(np.random.randint(10000)) + '.h5'
save_model(model, fname)
new_model = load_model(fname)
out2 = new_model.predict(x)
assert_allclose(out, out2, atol=1e-05)
# test that new updates are the same with both models
x = np.random.random((1, 3))
y = np.random.random((1, 3))
model.train_on_batch(x, y)
new_model.train_on_batch(x, y)
out = model.predict(x)
out2 = new_model.predict(x)
assert_allclose(out, out2, atol=1e-05)
# test load_weights on model file
model.load_weights(fname)
os.remove(fname)
def initial_num_char_phase1():
"""
识别二值化图像的字符个数
:param bw: 二值图像
:return:
"""
# 加载模型
model = Sequential()
model.add(Convolution2D(4, 5, 5, input_shape=(1, 30, 40), border_mode='valid'))
model.add(Activation('tanh'))
model.add(Convolution2D(8, 5, 5, input_shape=(1, 26, 36), border_mode='valid'))
model.add(Activation('tanh'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.55))
model.add(Convolution2D(16, 4, 4, input_shape=(1, 11, 16), border_mode='valid'))
model.add(Activation('tanh'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.60))
model.add(Flatten())
model.add(Dense(input_dim=16*4*6, output_dim=256, init='glorot_uniform'))
model.add(Activation('tanh'))
model.add(Dense(input_dim=256, output_dim=2, init='glorot_uniform'))
model.add(Activation('softmax'))
# 加载权值
model.load_weights('model/train_len_size1.d5')
sgd = SGD(l2=0.0, lr=0.05, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy', optimizer=sgd, class_mode="categorical")
return model
def test_merge_overlap():
left = Sequential()
left.add(Dense(nb_hidden, input_shape=(input_dim,)))
left.add(Activation('relu'))
model = Sequential()
model.add(Merge([left, left], mode='sum'))
model.add(Dense(nb_class))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=nb_epoch, show_accuracy=True, verbose=1, validation_data=(X_test, y_test))
model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=nb_epoch, show_accuracy=False, verbose=2, validation_data=(X_test, y_test))
model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=nb_epoch, show_accuracy=True, verbose=2, validation_split=0.1)
model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=nb_epoch, show_accuracy=False, verbose=1, validation_split=0.1)
model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=nb_epoch, verbose=0)
model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=nb_epoch, verbose=1, shuffle=False)
model.train_on_batch(X_train[:32], y_train[:32])
loss = model.evaluate(X_train, y_train, verbose=0)
assert(loss < 0.7)
model.predict(X_test, verbose=0)
model.predict_classes(X_test, verbose=0)
model.predict_proba(X_test, verbose=0)
model.get_config(verbose=0)
fname = 'test_merge_overlap_temp.h5'
model.save_weights(fname, overwrite=True)
model.load_weights(fname)
os.remove(fname)
nloss = model.evaluate(X_train, y_train, verbose=0)
assert(loss == nloss)
def __init__(self, restore=None, session=None, Dropout=Dropout, num_labels=10):
self.num_channels = 1
self.image_size = 28
self.num_labels = num_labels
model = Sequential()
nb_filters = 64
layers = [Conv2D(nb_filters, (5, 5), strides=(2, 2), padding="same",
input_shape=(28, 28, 1)),
Activation('relu'),
Conv2D(nb_filters, (3, 3), strides=(2, 2), padding="valid"),
Activation('relu'),
Conv2D(nb_filters, (3, 3), strides=(1, 1), padding="valid"),
Activation('relu'),
Flatten(),
Dense(32),
Activation('relu'),
Dropout(.5),
Dense(num_labels)]
for layer in layers:
model.add(layer)
if restore != None:
model.load_weights(restore)
self.model = model
def test_nested_sequential(in_tmpdir):
(x_train, y_train), (x_test, y_test) = _get_test_data()
inner = Sequential()
inner.add(Dense(num_hidden, input_shape=(input_dim,)))
inner.add(Activation('relu'))
inner.add(Dense(num_class))
middle = Sequential()
middle.add(inner)
model = Sequential()
model.add(middle)
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, verbose=1, validation_data=(x_test, y_test))
model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, verbose=2, validation_split=0.1)
model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, verbose=0)
model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, verbose=1, shuffle=False)
model.train_on_batch(x_train[:32], y_train[:32])
loss = model.evaluate(x_test, y_test, verbose=0)
model.predict(x_test, verbose=0)
model.predict_classes(x_test, verbose=0)
model.predict_proba(x_test, verbose=0)
fname = 'test_nested_sequential_temp.h5'
model.save_weights(fname, overwrite=True)
inner = Sequential()
inner.add(Dense(num_hidden, input_shape=(input_dim,)))
inner.add(Activation('relu'))
inner.add(Dense(num_class))
middle = Sequential()
middle.add(inner)
model = Sequential()
model.add(middle)
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
model.load_weights(fname)
os.remove(fname)
nloss = model.evaluate(x_test, y_test, verbose=0)
assert(loss == nloss)
# test serialization
config = model.get_config()
Sequential.from_config(config)
model.summary()
json_str = model.to_json()
model_from_json(json_str)
yaml_str = model.to_yaml()
model_from_yaml(yaml_str)
def get_conv(input_shape=(48, 48, 1), filename=None):
model = Sequential()
# model.add(Lambda(lambda x: (x-np.mean(x))/np.std(x),input_shape=input_shape, output_shape=input_shape))
model.add(Conv2D(32, (3, 3), activation='relu', name='conv1',
input_shape=input_shape, padding="same"))
model.add(Conv2D(32, (3, 3), activation='relu',
name='conv2', padding="same"))
model.add(MaxPooling2D(pool_size=(4, 4)))
model.add(Dropout(0.25))
model.add(Conv2D(64, (3, 3), name='conv3', padding='same'))
model.add(Activation('relu'))
model.add(Conv2D(64, (3, 3), name='conv4', padding='same'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(4, 4)))
model.add(Dropout(0.25))
model.add(Conv2D(256, (3, 3), activation="relu",
name="dense1")) # This was Dense(128)
model.add(Dropout(0.5))
# This was Dense(1)
model.add(Conv2D(1, (1, 1), name="dense2", activation="tanh"))
if filename:
model.load_weights(filename)
return model
def neural_net(num_sensors, params, load=''):
model = Sequential()
# First layer.
model.add(Dense(
params[0], init='lecun_uniform', input_shape=(num_sensors,)
))
model.add(Activation('relu'))
model.add(Dropout(0.2))
# Second layer.
model.add(Dense(params[1], init='lecun_uniform'))
model.add(Activation('relu'))
model.add(Dropout(0.2))
# Output layer.
model.add(Dense(3, init='lecun_uniform'))
model.add(Activation('linear'))
rms = RMSprop()
model.compile(loss='mse', optimizer=rms)
if load:
model.load_weights(load)
return model
def part4(weights):
global X_train, Y_train, X_test, Y_test
size = 3
model = Sequential()
model.add(Convolution2D(32, size, size, border_mode="same", input_shape=(img_channels, img_rows, img_cols)))
model.add(Activation("relu"))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.5))
model = add_top_layer(model)
model.load_weights(weights)
model = copy_freeze_model(model, 4)
# add top dense layer
model.add(Flatten())
model.add(Dense(512))
model.add(Activation("relu"))
model.add(Dropout(0.5))
model.add(Dense(10))
model.add(Activation("softmax"))
X_train, Y_train, X_test, Y_test = load_data(10)
model = train(model, auto=False)
print("Classification rate %02.5f" % (model.evaluate(X_test, Y_test, show_accuracy=True)[1]))
def getmodel():
nb_classes = 2
model = Sequential()
model.add(Convolution2D(32, 3, 3, border_mode='same',
input_shape=(1, RECEP_HEI, RECEP_WEI)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(32, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())
model.add(Dense(128))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))
sgd = SGD(lr=0.1, decay=1e-6, momentum=0.9, nesterov=True)
model.load_weights('weights.hdf5')
model.compile(loss='categorical_crossentropy',
optimizer=sgd,
metrics=['accuracy'])
# model.fit(X_train, Y_train, batch_size=32, nb_epoch=1,
# verbose=1, shuffle = True ,validation_split=0.25)
return model
def vgg_train(weights=None):
print "Compiling VGG Model..."
model = Sequential()
# input: 64x64 images with 3 channels -> (3, 64, 64) tensors.
# this applies 32 convolution filters of size 3x3 each.
model.add(Convolution2D(32, 3, 3, border_mode='valid', input_shape=(3,64,64)))
model.add(Activation('relu'))
model.add(Convolution2D(32, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Convolution2D(64, 3, 3, border_mode='valid'))
model.add(Activation('relu'))
model.add(Convolution2D(64, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.50))
model.add(Flatten())
# Note: Keras does automatic shape inference.
model.add(Dense(256,W_regularizer=WeightRegularizer(l1=1e-6,l2=1e-6)))
model.add(Activation('relu'))
model.add(Dropout(0.50))
model.add(Dense(200,W_regularizer=WeightRegularizer(l1=1e-5,l2=1e-5)))
model.add(Activation('softmax'))
if weights!=None:
model.load_weights(weights)
return model
def inference_dense(input_dim, class_num, optimizer='sgd', weights_file=''):
model = Sequential()
model.add(Dense(2048, input_dim=input_dim))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(1024))
model.add(Activation('relu'))
model.add(Dropout(0.5))
# model.add(Dense(256))
# model.add(Activation('relu'))
# model.add(Dropout(0.5))
model.add(Dense(class_num))
model.add(Activation('softmax'))
if weights_file:
model.load_weights(weights_file)
# adadelta = Adadelta(lr=1.0, rho=0.95, epsilon=1e-06)
if optimizer == 'sgd':
opt = SGD(lr=1e-4, decay=1e-6, momentum=0.9, nesterov=True)
elif optimizer == 'adam':
opt = Adam(lr=0.0001, beta_1=0.9, beta_2=0.999, epsilon=1e-08)
elif optimizer == 'adagrad':
opt = Adagrad(lr=0.01, epsilon=1e-08)
elif optimizer == 'adadelta':
opt = Adadelta(lr=1.0, rho=0.95, epsilon=1e-08)
elif optimizer == 'rmsprop':
opt = RMSprop(lr=0.001, rho=0.9, epsilon=1e-08)
print('compiling model....')
model.compile(loss='categorical_crossentropy', optimizer=opt, metrics=['accuracy'])
return model
def get_nn_model(token_dict_size):
_logger.info('Initializing NN model with the following params:')
_logger.info('Input dimension: %s (token vector size)' % TOKEN_REPRESENTATION_SIZE)
_logger.info('Hidden dimension: %s' % HIDDEN_LAYER_DIMENSION)
_logger.info('Output dimension: %s (token dict size)' % token_dict_size)
_logger.info('Input seq length: %s ' % INPUT_SEQUENCE_LENGTH)
_logger.info('Output seq length: %s ' % ANSWER_MAX_TOKEN_LENGTH)
_logger.info('Batch size: %s' % SAMPLES_BATCH_SIZE)
model = Sequential()
seq2seq = SimpleSeq2seq(
input_dim=TOKEN_REPRESENTATION_SIZE,
input_length=INPUT_SEQUENCE_LENGTH,
hidden_dim=HIDDEN_LAYER_DIMENSION,
output_dim=token_dict_size,
output_length=ANSWER_MAX_TOKEN_LENGTH,
depth=1
)
model.add(seq2seq)
model.compile(loss='mse', optimizer='rmsprop')
model.save_weights(NN_MODEL_PATH)
# use previously saved model if it exists
_logger.info('Looking for a model %s' % NN_MODEL_PATH)
if os.path.isfile(NN_MODEL_PATH):
_logger.info('Loading previously calculated weights...')
model.load_weights(NN_MODEL_PATH)
_logger.info('Model is built')
return model
def get_model(self):
classes = 36
# data = np.empty((57218, 1, 24, 24), dtype="float32")
model = Sequential()
model.add(Convolution2D(4, 5, 5, border_mode='valid', input_shape=(1, 24, 24)))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Convolution2D(8, 3, 3, border_mode='valid'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(16, 3, 3, border_mode='valid'))
model.add(BatchNormalization())
model.add(Activation('relu'))
# model.add(Dropout(0.5))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())
model.add(Dense(128, init='normal'))
model.add(BatchNormalization())
model.add(Activation('tanh'))
model.add(Dense(classes, init='normal'))
model.add(Activation('softmax'))
sgd = SGD(l2=0.0, lr=0.05, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy', optimizer=sgd, class_mode="categorical")
model.load_weights("eduLogin/captcha/tmp/weights.11-0.05.h5")
return model
def build(width, height, depth, num_classes, weights_path=None, dropout=True):
model = Sequential()
# First set of CONV => RELU => POOL.
model.add(Convolution2D(20, 5, 5, border_mode='same',
input_shape=(height, width, depth)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
# Second set of CONV => RELU => POOL.
model.add(Convolution2D(50, 5, 5, border_mode='same'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
# Last set of FC => RELU layers
model.add(Flatten())
model.add(Dense(500))
model.add(Activation('relu'))
if dropout:
model.add(Dropout(.5))
# softmax classifier
model.add(Dense(num_classes))
model.add(Activation('softmax'))
# If a weights path is supplied (inicating that the model was
# pre-trained), then load the weights
if weights_path is not None:
model.load_weights(weights_path)
# return the constructed network architecture
return model
def predict_ranking(evalFile, outFile):
X, qids, pids = load_data(evalFile)
input_dim = X[0].shape[1]
assert len(pids[0]) == len(X[0])
model = Sequential()
model.add(Dense(64, input_dim=input_dim, init='uniform', activation='relu'))
model.add(Dense(64, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='binary_crossentropy', optimizer='rmsprop', metrics=['accuracy'])
weightsFile = '../model/weights.hdf5'
model.load_weights(weightsFile)
Y_p = []
for x in X:
Y_p.append(model.predict(x))
f = open(outFile, 'w')
for n, qid in enumerate(qids):
tupes = zip(Y_p[n], pids[n])
sortedTupes = sorted(tupes, key=lambda x: x[0], reverse=True)
for n, (y, pid) in enumerate(sortedTupes):
f.write('{}\tITER\t{}\t{}\t{}\tSOMEID\n'.format(qid, pid, n, 1001-n))
def spatialNet(weights=None):
model = Sequential()
leakyrelu = ELU()
model.add(Convolution3D(30, 10, 17, 17, subsample=(2,2,2), trainable=False, input_shape=(1, 60,32,32)))
model.add(Activation(leakyrelu))
model.add(BatchNormalization(mode=2))
model.add(MaxPooling3D(pool_size=(13, 2, 2),trainable=False, strides=(13,2, 2)))
model.add(Reshape((60, 4, 4)))
model.add(Convolution2D(100, 3, 3, trainable=False))
model.add(Activation(leakyrelu))
model.add(BatchNormalization(mode=2))
model.add(MaxPooling2D((2, 2), strides=(2, 2), trainable=False))
model.add(Flatten())
model.add(Dense(400, trainable=False))
model.add(Activation(leakyrelu))
model.add(BatchNormalization(mode=2))
model.add(Dense(50))
model.add(Activation(leakyrelu))
model.add(BatchNormalization(mode=2))
model.add(Dense(1, activation='sigmoid'))
model.add(BatchNormalization(mode=2))
if weights:
model.load_weights(weights)
return model
def home_made_convnet(Xshape, Yshape):
print 'initializing model...'
model = Sequential()
model.add(ZeroPadding2D((2, 2), input_shape=(Xshape))) # 0
model.add(Convolution2D(64, 5, 5, activation='relu')) # 1
model.add(Convolution2D(64, 5, 5, activation='relu')) # 3
model.add(MaxPooling2D((2, 2), strides=(2, 2))) # 4
model.add(Convolution2D(128, 3, 3, activation='relu')) # 6
model.add(Convolution2D(128, 3, 3, activation='relu')) # 8
model.add(MaxPooling2D((2, 2), strides=(2, 2))) # 9
model.add(Flatten()) # 31
model.add(Dense(256, activation='relu')) # 32
model.add(Dropout(0.5)) # 33
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(Yshape, activation='softmax'))
model.load_weights('weights_f.h5')
sgd = SGD(lr=0.001, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(optimizer=sgd,
loss='categorical_crossentropy',
metrics=['accuracy'])
return model
class Model: def __init__(self, init='normal', activation='relu', batch_size=32, lr=1e-2, load=None): self.batch_size = batch_size self.model = Sequential() self.model.add(Convolution1D(8, 1, input_shape=(2, 4), init=init)) self.model.add(Activation(activation)) self.model.add(Flatten()) self.model.add(Dense(8, init=init)) self.model.add(BatchNormalization()) self.model.add(Activation(activation)) self.model.add(Dense(8, init=init)) self.model.add(BatchNormalization()) self.model.add(Activation(activation)) self.model.add(Dense(2, init=init)) if load != None: self.model.load_weights(load) self.model.compile(SGD(lr=lr), loss='mse') def predict(self, X): return self.model.predict(X.reshape((1,) + X.shape))[0] def learn(self, X, y): return self.model.fit(X, y, nb_epoch=1, batch_size=self.batch_size, shuffle=True, verbose=0) def save(self, filename): return self.model.save_weights(filename, overwrite=True)
def temporalNet(weights=None):
model = Sequential()
model.add(Convolution3D(30, 20, 17, 17, activation='relu', trainable=False, subsample=(4,2,2), input_shape=(1, 120,32,32)))
model.add(MaxPooling3D(pool_size=(13, 2, 2), strides=(13,2, 2), trainable=False))
model.add(Reshape((60, 4, 4)))
model.add(Convolution2D(100, 3, 3, activation='relu', trainable=False))
model.add(MaxPooling2D((2, 2), strides=(2, 2), trainable=False))
model.add(Flatten())
model.add(Dense(400, activation='relu', trainable=False))
model.add(Dense(50, activation='relu'))
#model.add(Dense(14, activation='relu'))
model.add(Dense(1, activation='relu'))
if weights:
model.load_weights(weights)
return model
def M8(weights_path=None, input_shape=(1, 64, 64), n_output=None):
model = Sequential()
model.add(Convolution2D(
32, 3, 3, border_mode='same', input_shape=input_shape))
model.add(Activation('relu'))
model.add(Convolution2D(32, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.5))
model.add(Convolution2D(64, 3, 3, border_mode='same'))
model.add(Activation('relu'))
model.add(Convolution2D(64, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.5))
model.add(Convolution2D(128, 3, 3, border_mode='same'))
model.add(Activation('relu'))
model.add(Convolution2D(128, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.5))
model.add(Flatten())
model.add(Dense(1024))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(n_output))
model.add(Activation('softmax'))
if weights_path:
model.load_weights(weights_path)
return model
def Colorize(weights_path=None):
model = Sequential()
# input: 100x100 images with 3 channels -> (3, 100, 100) tensors.
# this applies 32 convolution filters of size 3x3 each.
model.add(Convolution2D(512, 1, 1, border_mode='valid',input_shape=(960,224,224)))
model.add(Activation('relu'))
model.add(normalization.BatchNormalization())
model.add(Convolution2D(256, 1, 1, border_mode='valid'))
model.add(Activation('relu'))
model.add(normalization.BatchNormalization())
model.add(Convolution2D(112, 1, 1, border_mode='valid'))
model.add(Activation('relu'))
model.add(normalization.BatchNormalization())
print "output shape: ",model.output_shape
#softmax
model.add(Reshape((112,224*224)))
print "output_shape after reshaped: ",model.output_shape
model.add(Activation('softmax'))
if weights_path:
model.load_weights(weights_path)
return model
def make_model_full(inshape, num_classes, weights_file=None):
model = Sequential()
model.add(KL.InputLayer(input_shape=inshape[1:]))
# model.add(KL.Conv2D(32, (3, 3), padding='same', input_shape=inshape[1:]))
model.add(KL.Conv2D(32, (3, 3), padding='same'))
model.add(KL.Activation('relu'))
model.add(KL.Conv2D(32, (3, 3)))
model.add(KL.Activation('relu'))
model.add(KL.MaxPooling2D(pool_size=(2, 2)))
model.add(KL.Dropout(0.25))
model.add(KL.Conv2D(64, (3, 3), padding='same'))
model.add(KL.Activation('relu'))
model.add(KL.Conv2D(64, (3, 3)))
model.add(KL.Activation('relu'))
model.add(KL.MaxPooling2D(pool_size=(2, 2)))
model.add(KL.Dropout(0.25))
model.add(KL.Flatten())
model.add(KL.Dense(512))
model.add(KL.Activation('relu'))
model.add(KL.Dropout(0.5))
model.add(KL.Dense(num_classes))
model.add(KL.Activation('softmax'))
if weights_file is not None and os.path.exists(weights_file):
model.load_weights(weights_file)
return model
def temporalNet(weights=None):
model = Sequential()
#3D convolutional layer with 32x32 optical flow as input
model.add(Convolution3D(30, 20, 17, 17, subsample=(4,2,2), input_shape=(1, 120,32,32)))
model.add(Activation(LeakyReLU()))
model.add(BatchNormalization())
model.add(MaxPooling3D(pool_size=(13, 2, 2), strides=(13,2, 2)))
model.add(Reshape((60, 4, 4)))
model.add(Convolution2D(100, 3, 3))
model.add(Activation(LeakyReLU()))
model.add(BatchNormalization())
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(Flatten())
model.add(Dense(400))
model.add(Activation(LeakyReLU()))
model.add(Dropout(0.2))
model.add(Dense(50))
model.add(Activation(LeakyReLU()))
model.add(BatchNormalization())
model.add(Dense(4, activation='softmax'))
if weights:
model.load_weights(weights)
return model
def vgg_basic(img_size, weights_path = None, lr = 0.001):
'''
INPUT: img_size = size of images to train/ model was trained on
weights_path = path to get weights of trained model
OUTPUT: the fitted/unfitted model depending on if a weights path was
specified
A basic convolutional neural net. I found this one to have the best results.
'''
model = Sequential()
model.add(ZeroPadding2D((1,1),input_shape=(3, img_size, img_size)))
model.add(Convolution2D(64, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(64, 3, 3, activation='relu'))
model.add(MaxPooling2D((2,2), strides=(2,2)))
model.add(Flatten())
model.add(Dense(1024, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(5, activation='softmax'))
if weights_path:
model.load_weights(weights_path)
adam = Adam(lr = lr)
model.compile(optimizer=adam,
loss='categorical_crossentropy')
return model
def create_model(load_weights=False):
nn = Sequential()
nn.add(Convolution2D(32, 1, 3, 3, border_mode='same', activation='relu'))
nn.add(Convolution2D(32, 32, 3, 3, border_mode='same', activation='relu'))
nn.add(MaxPooling2D(poolsize=(2,2)))
nn.add(Dropout(0.25))
nn.add(Convolution2D(64, 32, 3, 3, border_mode='same', activation='relu'))
nn.add(Convolution2D(64, 64, 3, 3, border_mode='same', activation='relu'))
nn.add(MaxPooling2D(poolsize=(2,2)))
nn.add(Dropout(0.25))
nn.add(Flatten())
nn.add(Dense(64*7*7, 256, activation='relu'))
nn.add(Dropout(0.5))
nn.add(Dense(256,10, activation='softmax'))
sgd = SGD(lr=0.1, decay=1e-6, momentum=0.9, nesterov=True)
nn.compile(loss='categorical_crossentropy', optimizer=sgd)
if load_weights:
nn.load_weights('cnn_weights.hdf5')
return nn
def model(df, parent_id, go_id):
# Training
batch_size = 64
nb_epoch = 64
# Split pandas DataFrame
n = len(df)
split = 0.8
m = int(n * split)
train, test = df[:m], df[m:]
# train, test = train_test_split(
# labels, data, batch_size=batch_size)
train_label, train_data = train['labels'], train['data']
if len(train_data) < 100:
raise Exception("No training data for " + go_id)
test_label, test_data = test['labels'], test['data']
test_label_rep = test_label
train_data = train_data.as_matrix()
test_data = test_data.as_matrix()
train_data = numpy.hstack(train_data).reshape(train_data.shape[0], 8000)
test_data = numpy.hstack(test_data).reshape(test_data.shape[0], 8000)
shape = numpy.shape(train_data)
print('X_train shape: ', shape)
print('X_test shape: ', test_data.shape)
model = Sequential()
model.add(Dense(8000, activation='relu', input_dim=8000))
model.add(Highway())
model.add(Dense(1, activation='sigmoid'))
model.compile(
loss='binary_crossentropy', optimizer='rmsprop', class_mode='binary')
model_path = DATA_ROOT + parent_id + '/' + go_id + '.hdf5'
checkpointer = ModelCheckpoint(
filepath=model_path, verbose=1, save_best_only=True)
earlystopper = EarlyStopping(monitor='val_loss', patience=7, verbose=1)
model.fit(
X=train_data, y=train_label,
batch_size=batch_size, nb_epoch=nb_epoch,
show_accuracy=True, verbose=1,
validation_split=0.2,
callbacks=[checkpointer, earlystopper])
# Loading saved weights
print 'Loading weights'
model.load_weights(model_path)
pred_data = model.predict_classes(
test_data, batch_size=batch_size)
return classification_report(list(test_label_rep), pred_data)
def logistic_regression(model_folder, layer, dimension, number_of_feature,
cost="binary_crossentropy", learning_rate=1e-6, dropout_rate=0.5, nepoch=10, activation="relu"):
model = Sequential()
model.add(Dense(dimension, input_dim=number_of_feature, init="uniform", activation=activation))
model.add(Dropout(dropout_rate))
for idx in range(0, layer-2, 1):
model.add(Dense(dimension, input_dim=dimension, init="uniform", activation=activation))
model.add(Dropout(dropout_rate))
model.add(Dense(1, init="uniform", activation="sigmoid"))
optimizer = RMSprop(lr=learning_rate, rho=0.9, epsilon=1e-06)
model.compile(loss=cost, optimizer=optimizer, metrics=['accuracy'])
filepath_model = get_newest_model(model_folder)
if filepath_model:
model.load_weights(filepath_model)
log("Load weights from {}".format(filepath_model), INFO)
else:
log("A new one model, {}".format(model_folder), INFO)
return model
def C3D_Sports1M(weights_path=None):
model = Sequential()
# 1st layer group
model.add(Convolution3D(64, 3, 3, 3, activation='relu',
border_mode='same', name='conv1',
subsample=(1, 1, 1),
input_shape=(3, 16, 112, 112)))
model.add(MaxPooling3D(pool_size=(1, 2, 2), strides=(1, 2, 2),
border_mode='valid', name='pool1'))
# 2nd layer group
model.add(Convolution3D(128, 3, 3, 3, activation='relu',
border_mode='same', name='conv2',
subsample=(1, 1, 1)))
model.add(MaxPooling3D(pool_size=(2, 2, 2), strides=(2, 2, 2),
border_mode='valid', name='pool2'))
# 3rd layer group
model.add(Convolution3D(256, 3, 3, 3, activation='relu',
border_mode='same', name='conv3a',
subsample=(1, 1, 1)))
model.add(Convolution3D(256, 3, 3, 3, activation='relu',
border_mode='same', name='conv3b',
subsample=(1, 1, 1)))
model.add(MaxPooling3D(pool_size=(2, 2, 2), strides=(2, 2, 2),
border_mode='valid', name='pool3'))
# 4th layer group
model.add(Convolution3D(512, 3, 3, 3, activation='relu',
border_mode='same', name='conv4a',
subsample=(1, 1, 1)))
model.add(Convolution3D(512, 3, 3, 3, activation='relu',
border_mode='same', name='conv4b',
subsample=(1, 1, 1)))
model.add(MaxPooling3D(pool_size=(2, 2, 2), strides=(2, 2, 2),
border_mode='valid', name='pool4'))
# 5th layer group
model.add(Convolution3D(512, 3, 3, 3, activation='relu',
border_mode='same', name='conv5a',
subsample=(1, 1, 1)))
model.add(Convolution3D(512, 3, 3, 3, activation='relu',
border_mode='same', name='conv5b',
subsample=(1, 1, 1)))
model.add(ZeroPadding3D(padding=(0, 1, 1)))
model.add(MaxPooling3D(pool_size=(2, 2, 2), strides=(2, 2, 2),
border_mode='valid', name='pool5'))
model.add(Flatten())
# FC layers group
model.add(Dense(4096, activation='relu', name='fc6'))
model.add(Dropout(.5))
model.add(Dense(4096, activation='relu', name='fc7'))
model.add(Dropout(.5))
model.add(Dense(487, activation='softmax', name='fc8'))
if weights_path:
model.load_weights(weights_path)
return model
activation='relu',
padding='same',
input_shape=input_shape)) # (?, 28, 28, 32)
auto_encoder.add(UpSampling2D((2, 2))) # (?, 28, 28, 32)
auto_encoder.add(Conv2D(15, (3, 3), activation='relu',
padding='same')) # (?, 28, 28, 32)
auto_encoder.add(Conv2D(1, (3, 3), activation='sigmoid',
padding='same')) # (?, 28, 28, 1)
###############################################################
#model compile: optimizer, loss function
auto_encoder.compile(optimizer='sgd', loss='mean_squared_error')
#model summary
auto_encoder.summary()
auto_encoder.load_weights('denoise_epoch_3_5.h5')
#train the model
auto_encoder.fit(x_train_noisy,
x_train,
epochs=2,
batch_size=128,
shuffle=True,
validation_data=(x_test_noisy, x_test))
#obtain the model predictions in testing dataset
decoded_imgs_noise = auto_encoder.predict(x_test_noisy)
decoded_imgs = auto_encoder.predict(x_test)
#save the model
auto_encoder.save('denoise_epoch_3_7.h5')
n = 10
def build_model(args):
"""
NVIDIA model used
Image normalization to avoid saturation and make gradients work better.
Convolution: 5x5, filter: 24, strides: 2x2, activation: RELU
Convolution: 5x5, filter: 36, strides: 2x2, activation: RELU
Convolution: 5x5, filter: 48, strides: 2x2, activation: RELU
Convolution: 3x3, filter: 64, strides: 1x1, activation: RELU
Convolution: 3x3, filter: 64, strides: 1x1, activation: RELU
Drop out (0.5)
Fully connected: neurons: 100, activation:RELU
Fully connected: neurons: 50, activation: RELU
Fully connected: neurons: 10, activation: RELU
Fully connected: neurons: 1 (output)
# the convolution layers are meant to handle feature engineering
the fully connected layer for predicting the steering angle.
dropout avoids overfitting
"""
model = Sequential()
model.add(Lambda(lambda x: x / 255.0 - 0.5, input_shape=INPUT_SHAPE))
model.add(
Conv2D(24,
5,
5,
activation='relu',
subsample=(2, 2),
W_regularizer=l2(0.001)))
model.add(BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))
model.add(
Conv2D(36,
5,
5,
activation='relu',
subsample=(2, 2),
W_regularizer=l2(0.001)))
model.add(BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))
model.add(
Conv2D(48,
5,
5,
activation='relu',
subsample=(2, 2),
W_regularizer=l2(0.001)))
model.add(BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))
model.add(Conv2D(64, 3, 3, activation='relu', W_regularizer=l2(0.001)))
# model.add(BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))
model.add(Conv2D(64, 3, 3, activation='relu', W_regularizer=l2(0.001)))
# model.add(BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))
# model.add(Dropout(args.keep_prob))
model.add(Flatten())
model.add(Dense(1164, activation='relu', W_regularizer=l2(0.001)))
model.add(Dropout(0.5))
model.add(Dense(100, activation='relu', W_regularizer=l2(0.001)))
model.add(Dropout(0.5))
model.add(Dense(50, activation='relu', W_regularizer=l2(0.001)))
model.add(Dropout(0.4))
model.add(Dense(10, activation='relu', W_regularizer=l2(0.001)))
model.add(Dropout(0.4))
model.add(Dense(1))
model.summary()
model.load_weights("model-ghosts.h5")
return model
def vgg16_model(img_rows, img_cols, channel=1, num_classes=None):
"""VGG 16 Model for Keras
Model Schema is based on
https://gist.github.com/baraldilorenzo/07d7802847aaad0a35d3
ImageNet Pretrained Weights
https://drive.google.com/file/d/0Bz7KyqmuGsilT0J5dmRCM0ROVHc/view?usp=sharing
Parameters:
img_rows, img_cols - resolution of inputs
channel - 1 for grayscale, 3 for color
num_classes - number of categories for our classification task
"""
model = Sequential()
model.add(ZeroPadding2D((1, 1), input_shape=(channel, img_rows, img_cols)))
model.add(Convolution2D(64, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(64, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(128, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(128, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
# Add Fully Connected Layer
model.add(Flatten())
model.add(Dense(4096, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(4096, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(1000, activation='softmax'))
# Loads ImageNet pre-trained data
model.load_weights('D:\Projeto\Datase\vgg16_weights.h5')
# Truncate and replace softmax layer for transfer learning
model.layers.pop()
model.outputs = [model.layers[-1].output]
model.layers[-1].outbound_nodes = []
model.add(Dense(num_classes, activation='softmax'))
# Uncomment below to set the first 10 layers to non-trainable (weights will not be updated)
#for layer in model.layers[:10]:
# layer.trainable = False
# Learning rate is changed to 0.001
sgd = SGD(lr=1e-3, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(optimizer=sgd,
loss='categorical_crossentropy',
metrics=['accuracy'])
return model
class QNetwork(tf.keras.Model):
# This class essentially defines the network architecture.
# The network should take in state of the world as an input,
# and output Q values of the actions available to the agent as the output.
def __init__(self,
environment_name,
obs_space,
action_space,
lr,
save_weights_path=None):
# Define your network architecture here. It is also a good idea to define any training operations
# and optimizers here, initialize your variables, or alternately compile your model here.
#pdb.set_trace()
super(QNetwork, self).__init__()
decay_rate = lr / 10000000
self.model = Sequential()
self.save_weights_path = save_weights_path
# pdb.set_trace()
input_dim = obs_space.shape[0]
if (environment_name == "CartPole-v0"):
self.model.add(Dense(30, input_dim=input_dim,
activation='tanh')) #30 or 50
# self.model.add(Dense(32,activation='relu'))
self.model.add(Dense(action_space.n))
else:
self.model.add(Dense(96, input_dim=input_dim, activation='relu'))
self.model.add(Dense(96, activation='relu'))
self.model.add(Dense(96, activation='tanh'))
self.model.add(Dense(action_space.n))
adam = optimizers.Adam(lr=lr)
self.model.compile(optimizer=adam, loss='mean_squared_error')
def save_model_weights(self, suffix):
# Helper function to save your model / weights.
file_path = os.path.join(self.save_weights_path, suffix)
self.model.save_weights(file_path)
def load_model(self, model_file):
# Helper function to load an existing model.
# e.g.: torch.save(self.model.state_dict(), model_file)
self.model = tf.keras.load_model(
os.path.join(self.save_weights_path, model_file))
def load_model_weights(self, weight_file):
# Helper funciton to load model weights.
# e.g.: self.model.load_state_dict(torch.load(model_file))
self.model.load_weights(
os.path.join(self.save_weights_path, weight_file))
pass
def custom_mse_loss(y_true, y_pred):
## y_true: actual q_value of the state, chosen action
## y_pred: q_values for all the functions for the corresponding state
y_out = y_pred[0, actions]
loss = keras.losses.mean_squared_loss(y_true, y_out)
return loss
model.add(Dropout(0.2))
model.add(LSTM(256))
model.add(Dropout(0.2))
model.add(Dense(y.shape[1], activation='softmax'))
model.compile(loss='categorical_crossentropy', optimizer='adam')
#filepath="weights-improvement-{epoch:02d}-{loss:.4f}.hdf5"
filepath="weights-improvement-{epoch:02d}-{loss:.4f}-bigger.hdf5"
checkpoint = ModelCheckpoint(filepath, monitor='loss', verbose=1, save_best_only=True, mode='min')
callbacks_list = [checkpoint]
model.fit(X, y, epochs=50, batch_size=64, callbacks=callbacks_list)
'''
filename = "FILENAME"
model.load_weights(filename)
model.compile(loss='categorical_crossentropy', optimizer='adam')
int_to_char = dict((i, c) for i, c in enumerate(chars))
start = np.random.randint(0, len(dataX)-1)
pattern = dataX[start]
print("Seed: ")
print("\"", ''.join([int_to_char[value] for value in pattern]), "\"")
for i in range(1500):
x = np.reshape(pattern, (1, len(pattern), 1))
x = x / float(vocabulary)
prediction = model.predict(x, verbose=0)
index = np.argmax(prediction)
validation_data=(x_val, y_val),
epochs=4,
callbacks=[checkpointer],
batch_size=64)
del data
del texts
del Y_train
del labels
print("Creating average feature vecs for test reviews..")
texts_test = []
for review in test["review"]:
texts_test.append(review_to_wordlist(review, remove_stopwords=True))
sequences_test = tokenizer.texts_to_sequences(texts_test)
X_test = pad_sequences(sequences_test, maxlen=MAX_SEQUENCE_LENGTH)
model.load_weights('weights.hdf5')
result = model.predict(X_test)
result[result > 0.5] = 1
result[result < 0.5] = 0
result = result.astype(int)
result = np.squeeze(result)
id_t = test["id"]
output = pd.DataFrame(data={"id": id_t, "sentiment": result})
output.to_csv("Word2Vec_CNNVectors_300_pool_epoch4_lemma.csv",
index=False,
quoting=3)
model.add(Dense(128))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Dropout(0.75))
model.add(Dense(64))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(4))
model.add(Activation('softmax'))
model.load_weights(weights)
def precision(cl, row):
return row[cl] / sum(row)
def recall(cl, col):
return col[cl] / sum(col)
def test_cls(cls_name, cls_num, res):
files = [f for f in glob.glob(data + '/' + cls_name + '/*.bmp')]
for f in files:
img = image.load_img(f, target_size=(img_size, img_size))
img = image.img_to_array(img)
class DogBreedClassifier():
def __init__(self):
self._faceDetector = HumanFaceDetector()
self._dogDetector = DogDetector()
with open('data/dog_names.json', 'r') as f:
self._dog_names = json.load(f)
f.close()
# ResNet-50 model for dog breed classification
with session.graph.as_default() and session.session.as_default():
self._dogBreedCNN = Sequential()
self._dogBreedCNN.add(
GlobalAveragePooling2D(input_shape=(7, 7, 2048)))
self._dogBreedCNN.add(Dense(133, activation='softmax'))
self._dogBreedCNN.load_weights(
'data/model_weights_best_Resnet50.hdf5')
# ResNet-50 model for feature extration
with session.graph.as_default() and session.session.as_default():
self._featureExtractor = ResNet50(weights='imagenet',
include_top=False)
# Helper function to generate bottleneck features from CNN
def _extract_bottleneck_features(self, tensor):
'''
INPUT:
Image tensor
OUTPUT:
bottleneck features
Description:
Extract bottleneck features
'''
return self._featureExtractor.predict(preprocess_input(tensor))
# Classification of dog breed
def predict_breed(self, img_path):
'''
INPUT:
img_path path of image
OUTPUT:
dog breed
Description:
Takes a path to an image as input and returns the dog breed that is predicted by the model.
'''
features = self._extract_bottleneck_features(
convert_path_to_tensor(img_path))
# obtain predicted vector
predicted_vector = self._dogBreedCNN.predict(features)
# return dog breed that is predicted by the model
return self._dog_names[np.argmax(predicted_vector)]
# Algorithm to obtain dog breed from dog images or resembling dog breed from human face
def classify_dog_breed(self, img_path):
'''
INPUT:
img_path path of image
OUTPUT:
dog breed
Description:
Takes a path to an image and check if a dog or a human face is in the image.
If True, then, predicts the dog breed (resembling dog breed for human face).
If neither a dog or human face in the image, return (-1,-1)
'''
if self._dogDetector.detect_dog(img_path):
# image contains a dog
breed_id = self.predict_breed(img_path)
isDog = 1
else:
faces = self._faceDetector.detect_faces(img_path)
if len(faces) and faces.shape[0] == 1:
# image contains at least a human face
breed_id = self.predict_breed(img_path)
isDog = 0
else:
# Image does not contain a dog nor a human face
breed_id = -1
isDog = -1
return (isDog, breed_id)
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes, activation='softmax'))
model.compile(loss=keras.losses.categorical_crossentropy,
optimizer=keras.optimizers.Adadelta(),
metrics=['accuracy'])
epochs = 50
batch_size = 128
tbCallBack = keras.callbacks.TensorBoard(log_dir='./tensorboard',
histogram_freq=0,
write_graph=True,
write_images=True)
model.load_weights('model.hdf5')
model.fit(train_X,
train_y,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_data=(test_X, test_y),
callbacks=[tbCallBack])
scores = model.evaluate(train_X, train_y)
print("\n%s: %.2f%%" % (model.metrics_names[1], scores[1] * 100))
scores = model.evaluate(test_X, test_y)
print("\nTEST %s: %.2f%%" % (model.metrics_names[1], scores[1] * 100))
# coding: utf-8
from keras.models import Sequential
from keras.layers import Dense
import numpy as np
import gym
env = gym.make('CartPole-v0')
brain = Sequential([
Dense(16, activation='relu', input_shape=(4,)),
Dense(2, activation='softmax')
])
brain.load_weights('./best.h5')
def test(env, brain):
done = False
score = 0
state = env.reset()
while not done:
# env.render()
state = np.expand_dims(state, axis=0)
action = brain.predict(state).argmax()
state, reward, done, _ = env.step(action)
score += reward
return score
scores = []
for i in range(100):
score = test(env, brain)
save_best_only=True)
early_stopping = EarlyStopping(monitor='val_loss', patience=pat)
opt = Adam(lr=learnrate,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-08,
decay=0.0)
model.compile(loss="exp_xsec", optimizer=opt)
model.fit(X_AL,
y_AL,
validation_data=(X_train, y_train),
nb_epoch=epochs,
batch_size=bat,
verbose=2,
callbacks=[history, checkpointer, early_stopping])
model.load_weights("27_general_gpu_5_" + str(j) + ".hdf5")
learnrate /= 2
j = j + 1
#evaluation
y_pred = model.predict(X_test)
print(np.shape(y_pred))
y_pred = np.exp(y_pred - 14.92)
print(np.shape(y_pred))
y_true = np.exp(y_test - 14.92)
print(np.shape(y_true))
y_true = np.reshape(y_true, (-1, 1))
y_true = np.array(y_true)
y_pred = np.array(y_pred)
#true vs. prediction figure for the 10k sample test set
ideal = np.arange(np.min(y_true), np.max(y_true), 0.005)
fig3 = plt.figure(figsize=(10, 10))
### Dog race model ###
# Transfer learning
print('Loading inception_resnet_v2')
model_cropped = InceptionResNetV2(weights='imagenet', include_top=False)
# End layers
doggo_model = Sequential()
# Input shape from the output of model_cropped using 224x224 images
doggo_model.add(GlobalAveragePooling2D(input_shape=(5, 5, 1536)))
doggo_model.add(Dense(133))
doggo_model.add(Dropout(0.4))
doggo_model.add(Dense(133))
doggo_model.add(Dropout(0.4))
doggo_model.add(Dense(133, activation='softmax'))
print('Loading end-layer weights')
doggo_model.load_weights('models/weights.best.doggo_model.hdf5')
### Human detector
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('models/haarcascade_frontalface_alt.xml')
# index webpage displays cool visuals and receives user input text for model
@app.route('/')
@app.route('/index', methods=['POST'])
def index():
return render_template('master.html') #, ids=ids, graphJSON=graphJSON)
@app.route('/predict', methods=['POST'])
class Biometric:
def __init__(self, window):
#Initializing the GUI
self.window = window
self.window.wm_title("Face Biometric")
l1 = Label(window, text = "Enter the roll no.")
l1.grid(row = 0, column = 0)
self.roll = StringVar()
self.e1 = Entry(window, textvariable = self.roll)
self.e1.grid(row = 0, column = 2, rowspan = 2)
b1 = Button(window, text = "Create profile", width = 12, command = self.create_directory)
b1.grid(row = 4, column = 4)
b2 = Button(window, text = "Capture face", width = 12, command = self.capture_face)
b2.grid(row = 5, column = 4)
b3 = Button(window, text = "Train", width = 12, command = self.train)
b3.grid(row = 6, column = 4)
b4 = Button(window, text = "load model", width = 12, command = self.load)
b4.grid(row = 7, column = 4)
b5 = Button(window, text = "Close", width = 12, command = self.window.destroy)
b5.grid(row = 8, column = 4)
b6 = Button(window, text = "Creator", width = 12, command = self.creator)
b6.grid(row = 8, column = 1)
b7 = Button(window, text = "Detect", width = 12, command = self.detect)
b7.grid(row = 8, column = 2)
b8 = Button(window, text = "Save model", width = 12, command = self.save)
b8.grid(row = 8, column = 3)
self.list1 = Listbox(window, height = 3, width = 25)
self.list1.grid(row = 6, column = 1, columnspan = 3)
#Neural Network Structure initialization
# Initializing the CNN
self.classifier = Sequential()
# Step 1 - Convolution
self.classifier.add(Conv2D(32, (3, 3), input_shape = (256, 256, 3), activation = 'relu'))
# Step 2 - Pooling
self.classifier.add(MaxPooling2D(pool_size = (2, 2)))
# Adding a second convolutional layer
self.classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
self.classifier.add(MaxPooling2D(pool_size = (2, 2)))
# Step 3 - Flattening
self.classifier.add(Flatten())
# Step 4 - Full connection
self.classifier.add(Dense(units = 128, activation = 'relu'))
self.classifier.add(Dropout(0.1))
self.classifier.add(Dense(units = 128, activation = 'relu'))
self.classifier.add(Dropout(0.1))
self.classifier.add(Dense(units = self.compute_files(), activation = 'sigmoid'))
# Compiling the CNN
self.classifier.compile(optimizer = 'adam', loss = 'sparse_categorical_crossentropy', metrics = ['accuracy'])
def create_directory(self):
os.mkdir('dataset/training_set/' + self.roll.get())
os.mkdir('dataset/test_set/' + self.roll.get())
def capture_face(self):
video = cv2.VideoCapture(0)
frame_no = 0
print("Capturing ")
while True:
_,frame = video.read() #1st is the boolean, 2nd array of image
cv2.imshow("Capturing",frame)
key = cv2.waitKey(1)
if frame_no == 200:
break
if frame_no % 10 == 8 or frame_no % 10 == 9:
cv2.imwrite('dataset/' + '/test_set/' + self.roll.get()+'/' + str(frame_no)+'.jpg', frame)
else:
cv2.imwrite('dataset/' + '/training_set/' +self.roll.get()+'/'+str(frame_no)+'.jpg', frame)
frame_no = frame_no + 1
self.list1.delete(0,END)
self.list1.insert(END, "Frames captured:"+str(frame_no))
video.release()
cv2.destroyAllWindows()
def load(self):
self.classifier.load_weights('weight_catdog.hdf5')
def save(self):
self.classifier.save_weights('weight_students.hdf5', overwrite = True)
def train(self):
train_datagen = ImageDataGenerator(rescale = 1./255,
shear_range = 0.2,
zoom_range = 0.2,
horizontal_flip = True)
test_datagen = ImageDataGenerator(rescale = 1./255)
training_set = train_datagen.flow_from_directory('dataset/training_set',
target_size = (256, 256),
batch_size = 32,
class_mode = 'sparse')
test_set = test_datagen.flow_from_directory('dataset/test_set',
target_size = (256, 256),
batch_size = 32,
class_mode = 'sparse')
#with tf.device('/CPU:2'):
self.classifier.fit_generator(training_set,
steps_per_epoch = self.compute_files()*160,
epochs = 10,
validation_data = test_set,
validation_steps = self.compute_files()*40)
save()
def detect(self):
video = cv2.VideoCapture(0)
_, test_img = video.read()
test_img = image.load_img(test_img, target_size = (256,256))
test_img = image.img_to_array(test_img)
test_img = np.expand_dims(test_img, axis = 0)
result.append(classifier.predict(test_img))
result = classifier.predict(test_img)
categories = training_set.class_indices
value = categories.values()
key = categories.keys()
c = {}
for i,j in zip(value,key):
c[i]=j
self.list1.delete(0,END)
self.list1.append(END,c[int(r)])
def compute_files(self):
cwd = os.getcwd()
os.chdir(cwd + '\\dataset')
l = os.listdir()
os.chdir(cwd)
return len(l)
def creator(self):
self.list1.delete(0,END)
self.list1.insert(END, "Created by Shubham Banerjee!")
def rnn_train(x_train, y_train, batch_size=128, epochs=10, pretrained=None):
with open('text_tokenizer.pkl', 'rb') as f:
tokenizer = pickle.load(f)
x_train = sequence.pad_sequences(x_train)
# Split validation data
x_train, x_val, y_train, y_val = train_test_split(x_train,
y_train,
test_size=0.1,
random_state=6)
word_index = tokenizer.word_index
emb = utils.load_embedding(100, pretrain='glove')
# Define RNN model
rnn = Sequential()
rnn.add(
Embedding(len(word_index) + 1,
100,
weights=[emb],
input_length=x_train.shape[1],
trainable=False))
#rnn.add(Flatten())
rnn.add(GRU(256, dropout=0.6))
rnn.add(Dense(256, activation='relu'))
rnn.add(Dropout(0.2))
rnn.add(Dense(128, activation='relu'))
rnn.add(Dropout(0.2))
rnn.add(Dense(128, activation='relu'))
rnn.add(Dropout(0.2))
rnn.add(Dense(64, activation='relu'))
rnn.add(Dropout(0.2))
rnn.add(Dense(64, activation='relu'))
rnn.add(Dropout(0.2))
rnn.add(Dense(y_train.shape[1], activation='sigmoid'))
# Compile & print model summary
rnn.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=[f1_score])
print(rnn.summary())
model_json = rnn.to_json()
with open("models/rnn_model.json", "w") as json_file:
json_file.write(model_json)
plot_model(rnn, to_file='rnn_model.png', show_shapes=True)
#history = History()
checkpointer = ModelCheckpoint(filepath="./models/rnn.h5",
verbose=1,
save_best_only=True,
monitor='val_f1_score',
mode='max')
earlystopping = EarlyStopping(monitor='val_f1_score',
patience=10,
verbose=1,
mode='max')
if pretrained != None:
rnn.load_weights(pretrained)
print('Continue Training.')
rnn.fit(x_train,
y_train,
batch_size=batch_size,
initial_epoch=20,
verbose=1,
epochs=epochs,
validation_data=(x_val, y_val),
callbacks=[earlystopping, checkpointer])
else:
# Train model
print('Start training from scratch.')
rnn.fit(x_train,
y_train,
batch_size=batch_size,
verbose=1,
epochs=epochs,
validation_data=(x_val, y_val),
callbacks=[earlystopping, checkpointer])
def func():
"""
Best performing model till now. Added layers to the webcamemocognizer one.
"""
global probdislike, problike, count, stop
model = Sequential()
model.add(
Convolution2D(32, (3, 3), padding='valid', input_shape=(48, 48, 1)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(64, (3, 3), padding='valid', activation='relu'))
#model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(128, (3, 3), padding='valid'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(256, (3, 3), padding='valid'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())
model.add(Dense(128, kernel_initializer="lecun_uniform"))
#model.add(Dense(128))
model.add(Activation('relu'))
model.add(Dense(2))
model.add(Activation('softmax'))
#model = model_from_json(open('./models/Face_model_architecture.json').read())
#model.load_weights('_model_weights.h5')
model.load_weights('deep_model_weights_binary_best.h5py')
#optimizer =Adam(lr=0.001, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.0)
sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='binary_crossentropy', optimizer=sgd)
cascPath = "haarcascade_frontalface_default.xml"
faceCascade = cv2.CascadeClassifier(cascPath)
video_capture = cv2.VideoCapture(0)
while True:
# Capture frame-by-frame
# sleep(0.8)
ret, frame = video_capture.read()
# detect faces
gray, detected_faces = detect_face(frame)
face_index = 0
# predict output
for face in detected_faces:
(x, y, w, h) = face
if w > 100:
# draw rectangle around face
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
# extract features
extracted_face = extract_face_features(
gray, face, (0.075, 0.05)) #(0.075, 0.05)
# predict smile
prediction_result = model.predict_classes(
extracted_face.reshape(1, 48, 48, 1))
pn = model.predict(extracted_face.reshape(1, 48, 48, 1))
# draw extracted face in the top right corner
frame[face_index * 48:(face_index + 1) * 48,
-49:-1, :] = cv2.cvtColor(extracted_face * 255,
cv2.COLOR_GRAY2RGB)
# annotate main image with a label
if prediction_result == 1:
cv2.putText(frame, "like!!", (x, y), cv2.FONT_ITALIC, 2,
155, 10)
#print("Like")
problike = problike + pn[0][1]
probdislike = probdislike + pn[0][0]
count = count + 1
elif prediction_result == 0:
cv2.putText(frame, "dislike", (x, y),
cv2.FONT_HERSHEY_SIMPLEX, 2, 155, 10)
#print("DisLike")
problike = problike + pn[0][1]
probdislike = probdislike + pn[0][0]
count = count + 1
# increment counter
face_index += 1
# Display the resulting frame
#cv2.imshow('Video', frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
if stop == 1:
video_capture.release()
while True:
if stop == 0:
video_capture = cv2.VideoCapture(0)
break
# When everything is done, release the capture
video_capture.release()
cv2.destroyAllWindows()
opt = keras.optimizers.rmsprop(lr=0.0001, decay=1e-6)
model.compile(loss='categorical_crossentropy',
optimizer=opt,
metrics=['accuracy'])
orig_data = 'data/keras_dataset_300k_{}.pkl' # 0-9
import_batch_dset1 = 'data/import_batch_20170813_{}.pkl' # 0-9
savefile = 'net/keras_net_v0_2017aug7.h5'
filelist1 = [import_batch_dset1.format(i) for i in range(9)]
filelist2 = [orig_data.format(i) for i in range(10)]
testfile = 'data/import_batch_20170813_9.pkl' # 0-9
filelist = filelist1[:4] + filelist2[:5] + filelist1[4:] + filelist2[5:]
model.load_weights(savefile)
def train_net(load=savefile):
'''train_net loads parameters from savefile by default,
and then trains 100 epochs on multi-file dataset
specified by filelist'''
if load:
try:
model.load_weights(load)
except:
input('WARNING: failed to load weights but will overwrite file {}'
.format(savefile))
else:
print('WARNING: not saving progress.')
#train model
model.fit(resh_x, resh_y, epochs=40)
#, validation_split=0.2)
#, callbacks=[early_stopping_monitor])
# train model before starting the game
createModel()
if TRAIN:
preTrainModel()
model.save("savedmodel")
sys.exit(0)
# else
model.load_weights("savedmodel")
healths = [-1]*2
def printStats(update, attack):
if update[0] != -1:
healths[0] = update[0]
if update[1] != -1:
healths[1] = update[1]
dr_att = "missed"
if update[3] != -1:
dr_att = update[3]
h_att = "missed"
if update[2] != -1:
h_att = update[2]
LSTM(HIDDEN_DIM, input_shape=(None, VOCAB_SIZE), return_sequences=True))
for i in range(LAYER_NUM - 1):
model.add(LSTM(HIDDEN_DIM, return_sequences=True))
model.add(TimeDistributed(Dense(VOCAB_SIZE)))
model.add(Activation('softmax'))
print("Compiling the network")
model.compile(loss="categorical_crossentropy", optimizer="rmsprop")
print("Generating a sample")
generate_text(model, args['generate_length'], VOCAB_SIZE, ix_to_char)
if not WEIGHTS == '':
print("Loading weights")
model.load_weights(WEIGHTS)
nb_epoch = int(WEIGHTS[WEIGHTS.rfind('_') + 1:WEIGHTS.find('.')])
else:
nb_epoch = 0
# Training if there is no trained weights specified
if args['mode'] == 'train' or WEIGHTS == '':
while True:
print('\n\nEpoch: {}\n'.format(nb_epoch))
model.fit(X, y, batch_size=BATCH_SIZE, verbose=1, epochs=1)
nb_epoch += 1
generate_text(model, GENERATE_LENGTH, VOCAB_SIZE, ix_to_char)
if nb_epoch % 10 == 0:
model.save_weights(
'checkpoint_layer_{}_hidden_{}_epoch_{}.hdf5'.format(
LAYER_NUM, HIDDEN_DIM, nb_epoch))
width = 224
height = 224
model = Sequential()
model.add(Conv2D(64, (4,4), input_shape=(width,height,3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(32, (3,3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(16, (2,2), activation='relu'))
model.add(Flatten())
model.add(Dense(256, activation='relu'))
model.add(Dense(128, activation='relu'))
model.add(Dense(128, activation='relu'))
model.add(Dense(2, activation='softmax'))
model.load_weights("eighth_try.h5")
model.compile(loss='categorical_crossentropy',
optimizer='adadelta',
metrics=['accuracy'])
def save_img(path, savepath, origimg, typeimg, layeridx):
img = load_img(path, target_size=(224,224))
x = img_to_array(img) #numpy array
x = x.reshape(x.shape) #adds on dimension for keras
kernel_regularizer=keras.regularizers.l2(weight_decay),
kernel_initializer=he_normal(),
name='fc2'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Dropout(dropout))
model.add(
Dense(10,
kernel_regularizer=keras.regularizers.l2(weight_decay),
kernel_initializer=he_normal(),
name='predictions_cifa10'))
model.add(BatchNormalization())
model.add(Activation('softmax'))
# 加载预训练的VGG19参数
model.load_weights(filepath, by_name=True)
# -------- 优化器设置 -------- #
sgd = optimizers.SGD(lr=.1, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy',
optimizer=sgd,
metrics=['accuracy'])
tb_cb = TensorBoard(log_dir=log_filepath, histogram_freq=0)
change_lr = LearningRateScheduler(scheduler)
cbks = [change_lr, tb_cb]
print('Using real-time data augmentation.')
datagen = ImageDataGenerator(horizontal_flip=True,
width_shift_range=0.125,
height_shift_range=0.125,
def test_nested_sequential(in_tmpdir):
(x_train, y_train), (x_test, y_test) = _get_test_data()
inner = Sequential()
inner.add(Dense(num_hidden, input_shape=(input_dim, )))
inner.add(Activation('relu'))
inner.add(Dense(num_class))
middle = Sequential()
middle.add(inner)
model = Sequential()
model.add(middle)
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_data=(x_test, y_test))
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
verbose=2,
validation_split=0.1)
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
verbose=0)
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
shuffle=False)
model.train_on_batch(x_train[:32], y_train[:32])
loss = model.evaluate(x_test, y_test, verbose=0)
model.predict(x_test, verbose=0)
model.predict_classes(x_test, verbose=0)
model.predict_proba(x_test, verbose=0)
fname = 'test_nested_sequential_temp.h5'
model.save_weights(fname, overwrite=True)
inner = Sequential()
inner.add(Dense(num_hidden, input_shape=(input_dim, )))
inner.add(Activation('relu'))
inner.add(Dense(num_class))
middle = Sequential()
middle.add(inner)
model = Sequential()
model.add(middle)
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
model.load_weights(fname)
os.remove(fname)
nloss = model.evaluate(x_test, y_test, verbose=0)
assert (loss == nloss)
# test serialization
config = model.get_config()
Sequential.from_config(config)
model.summary()
json_str = model.to_json()
model_from_json(json_str)
yaml_str = model.to_yaml()
model_from_yaml(yaml_str)
def Run_CNN(test_dir):
face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
IMG_SIZE = 350
PretrainedVGG = VGG16(weights='imagenet', include_top=False, input_shape=(IMG_SIZE,IMG_SIZE,3))
model = Sequential()
model.add(PretrainedVGG)
model.add(Dense(512, activation='relu'))
model.add(Dropout(0.1))
model.add(Dense(256, activation='relu'))
model.add(Dense(128, activation='relu'))
model.add(BatchNormalization())
model.add(Dense(64, activation='relu'))
model.add(Flatten())
model.add(Dense(7, activation='softmax'))
model.load_weights('CNNModel1.h5')
GenderModel = load_model('GenderDetectionModel.model')
font = cv2.FONT_HERSHEY_SIMPLEX
fontScale = 0.5
fontColor = (0,0,255)
lineType = 2
expressionLabels = ['Neutral','Anger','Disgust','Fear','Happy','Sadness','Surprise']
GenderLabels = ['Male','Female']
hogX = cv2.HOGDescriptor()
cap = cv2.VideoCapture(test_dir)
frameRate = cap.get(5)
preds, gpreds = [0]*7, [0,0]
while 1:
frameId = cap.get(1)
ret, frame = cap.read()
if not ret: break
if frameId%floor(frameRate): continue
faces = face_cascade.detectMultiScale(frame, 1.3,5)
for (x,y,w,h) in faces:
if w<100 or h<100: continue
cv2.rectangle(frame,(x,y),(x+w,y+h),(0,255,0),4)
roi_gray = frame[y:y+h, x:x+w]
genderPic = roi_gray.copy()
roi_gray = cv2.cvtColor(roi_gray,cv2.COLOR_BGR2GRAY)
roi_gray = cv2.resize(roi_gray, (IMG_SIZE, IMG_SIZE), interpolation = cv2.INTER_AREA)
roi_gray = np.repeat(roi_gray,3,-1)
roi_gray = roi_gray.reshape(-1,IMG_SIZE,IMG_SIZE,3)
prede = np.argmax(model.predict(roi_gray)[0])
genderPic = genderPic / 255.0
genderPic = cv2.resize(genderPic, (96, 96), interpolation = cv2.INTER_AREA)
genderPic = genderPic.reshape(-1,96,96,3)
predg = np.argmax(GenderModel.predict(genderPic)[0])
preds[prede]+=1
gpreds[predg]+=1
cv2.putText(frame,expressionLabels[prede]+' '+GenderLabels[predg],
(x,y-10),font,fontScale,
fontColor,lineType)
cv2.imshow('image',frame)
k = cv2.waitKey(60) & 0xff
if k == 27: break
cap.release()
cv2.destroyAllWindows()
totalPreds = sum(preds)
for i in range(7): print('{}: {} %'.format(expressionLabels[i],round(preds[i]/totalPreds,4)*100))
mxIndx = np.argmax(np.array(preds))
ExDict = dict(zip(expressionLabels,list(range(7))))
Exy_true = ExDict[test_dir.split('\\')[-2]]
if preds[Exy_true]==preds[mxIndx] and Exy_true!=mxIndx: mxIndx=Exy_true
print('CNN Winning expression: {} with Accuracy: {} %'.format(expressionLabels[mxIndx],round(preds[mxIndx]/totalPreds,4)*100))
totalX = sum(gpreds)
mxIndx2 = np.argmax(np.array(gpreds))
GenDict = dict(zip(GenderLabels,[0,1]))
Geny_true = GenDict[test_dir.split('_')[-2]]
if gpreds[Geny_true]==gpreds[mxIndx2] and Geny_true!=mxIndx2: mxIndx2=Geny_true
print('Winning Gender: {} with Accuracy: {} %'.format(GenderLabels[mxIndx2],round(gpreds[mxIndx2]/totalX,4)*100))
return (preds,gpreds)
batch_size=batch_size,
nb_epoch=BIG_epochs,
verbose=1,
validation_data=(x_test, y_test))
print("Evaluating model ..")
score = model.evaluate(x_test, y_test, verbose=1)
print('Test loss:', score[0])
print('Test accuracy:', score[1])
print("Savin Model weights")
model.save_weights("new_stockweighs.h5")
print("Model Weights Saved")
else:
print("loading weights TRAIN_BIG FLAG set as FALSE")
model.load_weights('stockweighs.h5')
print("Evaluating Initial Model ...\n")
score = model.evaluate(x_test, y_test, verbose=1)
print('Test loss:', score[0])
print('Test accuracy:', score[1])
print('-' * 30)
print("Parameter Size of Initial Model and Memory Footprint ")
trainable_params = model.count_params()
footprint = trainable_params * 4
print("Memory footprint per Image Feed Forward ~= ",
footprint / 1024.0 / 1024.0, "Mb") # 2x Backprop
print('-' * 30)
## Obtaining the Output of the last Convolutional Layer After Flatten. Caruana et. al
test_dir = os.path.join(imdb_dir, 'test')
labels = []
texts = []
for label_type in ['neg', 'pos']:
dir_name = os.path.join(test_dir, label_type)
for fname in sorted(os.listdir(dir_name)):
if fname[-4:] == '.txt':
f = open(os.path.join(dir_name, fname))
texts.append(f.read())
f.close()
if label_type == 'neg':
labels.append(0)
else:
labels.append(1)
sequences = tokenizer.texts_to_sequences(texts)
x_test = pad_sequences(sequences, maxlen=maxlen)
y_test = np.asarray(labels)
# And let's load and evaluate the first model:
# In[25]:
model.load_weights('pre_trained_glove_model.h5')
model.evaluate(x_test, y_test)
# We get an appalling test accuracy of 54%. Working with just a handful of training samples is hard!
print(p)
target_names = ['CLASS0(CRACK)', 'CLASS1(PATCH)', 'CLASS2(ROAD)']
#
print(classification_report(np.argmax(y_test,axis=1),y_pred,target_names=target_names))
#diagonal values
print(confusion_matrix(np.argmax(y_test,axis=1),y_pred))
#%% saving weights
fname= "pav_dis_cnn.hdf5"
model.save_weights(fname,overwrite=True)
#%%loading weights
fname="pav_dis_cnn.hdf5"
model.load_weights(fname)
#%%
score = model.evaluate(x_test, y_test, verbose=0)
print('Test score', score)
#print('Test accuracy',score)
predict= model.predict_classes(x_test[1:15])
print(y_test[1:15])
#%%
#from keras.utils.np_utils import to_categorical
#a= np.categorical_crossentropy(x_test, y_pred)
#%%
def test_sequential(in_tmpdir):
(x_train, y_train), (x_test, y_test) = _get_test_data()
# TODO: factor out
def data_generator(x, y, batch_size=50):
index_array = np.arange(len(x))
while 1:
batches = _make_batches(len(x_test), batch_size)
for batch_index, (batch_start, batch_end) in enumerate(batches):
batch_ids = index_array[batch_start:batch_end]
x_batch = x[batch_ids]
y_batch = y[batch_ids]
yield (x_batch, y_batch)
model = Sequential()
model.add(Dense(num_hidden, input_shape=(input_dim, )))
model.add(Activation('relu'))
model.add(Dense(num_class))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_data=(x_test, y_test))
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
verbose=2,
validation_split=0.1)
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
verbose=0)
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
shuffle=False)
model.train_on_batch(x_train[:32], y_train[:32])
loss = model.evaluate(x_test, y_test)
prediction = model.predict_generator(data_generator(x_test, y_test),
1,
max_queue_size=2,
verbose=1)
gen_loss = model.evaluate_generator(data_generator(x_test, y_test, 50),
1,
max_queue_size=2)
pred_loss = K.eval(
K.mean(
losses.get(model.loss)(K.variable(y_test),
K.variable(prediction))))
assert (np.isclose(pred_loss, loss))
assert (np.isclose(gen_loss, loss))
model.predict(x_test, verbose=0)
model.predict_classes(x_test, verbose=0)
model.predict_proba(x_test, verbose=0)
fname = 'test_sequential_temp.h5'
model.save_weights(fname, overwrite=True)
model = Sequential()
model.add(Dense(num_hidden, input_shape=(input_dim, )))
model.add(Activation('relu'))
model.add(Dense(num_class))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
model.load_weights(fname)
os.remove(fname)
nloss = model.evaluate(x_test, y_test, verbose=0)
assert (loss == nloss)
# test serialization
config = model.get_config()
Sequential.from_config(config)
model.summary()
json_str = model.to_json()
model_from_json(json_str)
yaml_str = model.to_yaml()
model_from_yaml(yaml_str)
recurrent_activation='sigmoid',
activity_regularizer=regularizers.l2(
0.01))) # activity_regularizer=regularizers.l2(0.01)
model.add(BatchNormalization())
model.add(Dropout(0.5))
model.add(Dense(10))
model.add(BatchNormalization())
model.add(LeakyReLU())
model.add(Dropout(0.5))
model.add(Dense(1))
opt = Nadam(lr=0.005) # 0.005, 0.002, 0.001, 0.01
model.compile(loss='mse', optimizer=opt)
model.load_weights(MODEL_FILE_PATH)
predictions = model.predict(test_X)
score = model.evaluate(test_X, test_Y, verbose=0)
print('Test loss:', score) # this is mean_squared_error
###########################################################
df_result = pd.DataFrame(test_Y, columns=['real'])
df_result['pred'] = pd.Series(predictions.reshape(-1))
yhat = ta.DEMA(np.array(df_result['pred'].astype(float)),
timeperiod=3) # 디노이징 이동평균
df_result['pred_dn'] = pd.Series(yhat, index=df_result.index)
df_result['pred_change'] = (df_result['pred_dn'] -
df_result['pred_dn'].shift(3))
df_result['classify'] = df_result['pred_change'].transform(
classify_trinary)
write_graph=True,
write_grads=False,
write_images=False,
embeddings_freq=0,
embeddings_layer_names=None,
embeddings_metadata=None)
callbacks_list = [checkpoint, early_stopping, tensor_board]
print('Train...')
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
validation_data=[x_val, y_val],
callbacks=callbacks_list)
model.load_weights(filepath)
score, acc = model.evaluate(x_test, y_test)
print(score, acc)
score, acc = model.evaluate(x_test, y_test)
print(score, acc)
# mosei = MOSI()
# embeddings = mosei.embeddings()
# sentiments = mosei.sentiments()
# train_ids = mosei.train()
# valid_ids = mosei.valid()
# embeddings = mosei.embeddings()
# train_set_ids = []
# for vid in train_ids:
# for sid in embeddings['embeddings'][vid].keys():
# if embeddings['embeddings'][vid][sid]:
freq = int(parts[1])
if freq >= UNAMBIG_FREQ_THRESHOLD:
word2tag[parts[0]] = freq
with open(os.path.join(model_folder, 'm1_chars_postagger_net.config'),
'rt') as cfg:
params = json.load(cfg)
WINDOW = int(params['WINDOW'])
bits_per_char = int(params['bits_per_char'])
bits_per_word = int(params['bits_per_word'])
output_size = int(params['output_size'])
INVERT = bool(params['invert'])
model = model_from_json(
open(os.path.join(model_folder, 'm1_chars_postagger_net.arch')).read())
model.load_weights(
os.path.join(model_folder, 'm1_chars_postagger_net.model'))
rdr_corpus = codecs.open(input_path, 'r', 'utf-8')
wrt_result = codecs.open(result_path, 'w', 'utf-8')
winspan = int((WINDOW - 1) / 2)
processed_sent_count = 0
total_word_count = 0
incorrect_pos_count = 0
incorrect_tagset_count = 0
sent = []
good = True
line_num = 0
for line0 in rdr_corpus:
# verbose=0, loss='euclidean',
# mean_y_train=mean_y_train, std_y_train=std_y_train)
# model.fit(X_train, Y_train, batch_size=batch_size, epochs=5,
# callbacks=[modeltest_1, modeltest_2])
model.fit(X_train, Y_train, batch_size=batch_size, epochs=50)
# Potentially save weights
# model.save_weights("path", overwrite=True)
final_time = time.time()
time_spent_printer(start_time, final_time)
else:
print('loading model ..')
# print('loading model from %s' % (folder + filename + ".hdf5"))
model.load_weights(folder + filename + ".hdf5")
# --
print("Test...")
# Evaluate model
# Dropout approximation for training data:
standard_prob = model.predict(X_train, batch_size=500, verbose=1)
print(
np.mean(((mean_y_train + std_y_train * np.atleast_2d(Y_train).T) -
(mean_y_train + std_y_train * standard_prob))**2, 0)**0.5)
# --
# Dropout approximation for test data:
