I have tried to extract feature in the keras ( with tensorflow and as well as with theano background). But became not successful.
My code is :
model = Sequential()
model.add(Convolution2D(64, 5, 5, input_shape=(3,img_width, img_height)))
model.add(LeakyReLU(alpha=.01))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(64, 2, 2))
model.add(LeakyReLU(alpha=.01))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.1))
model.add(Flatten())
model.add(Dense(128))
model.add(Activation('relu'))
model.add(Dense(8))
model.add(ActivityRegularization(l2=0.01))
model.add(Activation('relu'))
model.add(Dropout(0.25))
model.add(Dense(1))
model.add(Activation('sigmoid'))
test_datagen = ImageDataGenerator(rescale=1./255)
test_generator = test_datagen.flow_from_directory(
test_data_dir,
target_size=(img_width, img_height),
batch_size=16,
class_mode='binary')
scores = model.evaluate_generator(test_generator, 237)
print("Accuracy = ", scores[1])
I have used fit_generator for train, validation and test cases.
My code for getting output from e.g., layer 3 was:
get_activations = theano.function([model.layers[0].input],
model.layers[3].output(train=False), allow_input_downcast=True)
activations = get_activations(test_generator)
But after excuting it I am getting an error :
File "test.py", line 96, in <module>
get_activations = theano.function([model.layers[0].input],
model.layers[3].output(train=False), allow_input_downcast=True)
TypeError: 'TensorVariable' object is not callable
How can I do it In theano or in the tensorflow mode ( any of them or both). I have used fit_generator in my code for image augmentation.
Please help.
First of all, I recommend you to switch to Keras-2.x for a better experience. Keras-2.x is more well-organized.
Secondly, using theano codes like theano.function is not recommended, you should use keras.backend.function instead.
The error is raised because model.layers[3].output is a tensor, you cannot call a tensor like model.layers[3].output(train=False).
My code to extract features usually looks like:
model = Sequential()
model.add(Convolution2D(64, 5, 5, input_shape=(3,img_width, img_height)))
model.add(Convolution2D(64, 2, 2))
# blablabla...
model.add(Dense(8, name='feature'))
model.add(Dropout(0.25))
model.add(Dense(1))
model.add(Activation('sigmoid'))
feature_network = Model(model.input, model.get_layer('feature').output)
feature = feature_network.predict(your_data_here)
I use feature to name the target feature layer and use the get_layer function to retrieve that layer.
Related
What would be the output of a network with 32 units in 1st convolutional layer followed by a pooling layer with stride=2 and k=2 followed by another convolutional layer with 64 units followed by another pooling layer with stride = 2 and k=2 , input image dimensions are 28*28*1
I will assume that you are using same padding in each convolution, then the output of you model will be of shape (7, 7, 64). You can use a code similar to this one in keras if you don't want to compute the size by hand:
from keras.models import Sequential
from keras.layers import Conv2D, MaxPooling2D
model = Sequential()
model.add(Conv2D(32, 3, padding="same", input_shape=(28, 28, 1)))
model.add(MaxPooling2D(2, 2))
model.add(Conv2D(64, 3, padding="same"))
model.add(MaxPooling2D(2, 2))
print(model.summary())
i'm trying to implement an A3C model with LSTM using keras, i beging with this version of A3C without LSTM: "https://github.com/coreylynch/async-rl", and try to modify only the network code, but i struggle to compile the whole model:
am i missing something ?
this is my model:
state = tf.placeholder("float", [None, agent_history_length, resized_width, resized_height])
vision_model = Sequential()
vision_model.add(Conv2D(activation="relu", filters=16, kernel_size=(8, 8), name="conv1", padding="same", strides=(4, 4),input_shape=(agent_history_length,resized_width, resized_height)))
vision_model.add(Conv2D(activation="relu", filters=32, kernel_size=(4, 4), name="conv2", padding="same", strides=(2, 2)))
vision_model.add(Flatten())
vision_model.add(Dense(activation="relu", units=256, name="h1"))
# Now let's get a tensor with the output of our vision model:
state_input = Input(shape=(1,agent_history_length,resized_width,resized_height))
encoded_frame_sequence = TimeDistributed(vision_model)(state_input)
encoded_video = LSTM(256)(encoded_frame_sequence) # the output will be a vector
action_probs = Dense(activation="softmax", units=4, name="p")(encoded_video)
state_value = Dense(activation="linear", units=1, name="v")(encoded_video)
policy_network = Model(inputs=state_input, outputs=action_probs)
value_network = Model(inputs=state_input, outputs=state_value)
p_params = policy_network.trainable_weights
v_params = value_network.trainable_weights
policy_network.summary()
value_network.summary()
p_out = policy_network(state_input)
v_out = value_network(state_input)
keras-rl examples lib does NOT support more than 2D input shape !
I want to train my data with a convolution neural network, I have reshaped my data:
Those are parameters that I have used:
'x_train.shape'=(500000, 3253)
'y_train.shape', (500000,)
'y_test.shape', (20000,)
'y_train[0]', 97
'y_test[0]', 99
'y_train.shape', (500000, 256)
'y_test.shape', (20000, 256)
This is how I define my model architecture:
# 3. Define model architecture
model = Sequential()
model.add(Conv1D(64, 8, strides=1, padding='valid',
dilation_rate=1, activation=None, use_bias=True, kernel_initializer='glorot_uniform',
bias_initializer='zeros', kernel_regularizer=None, bias_regularizer=None,
activity_regularizer=None, kernel_constraint=None, bias_constraint=None, input_shape=x_train.shape))
# input_traces=N_Features
# input_shape=(batch_size, trace_lenght,num_of_channels)
model.add(MaxPooling1D(pool_size=2,strides=None, padding='valid'))
model.add(Flatten())
model.add(Dropout(0.5))
model.add(Dense(1, activation='relu'))
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
print(model.summary())
model.fit(x_train, y_train, batch_size=100, epochs=500,verbose=2)
But i got two Errors :
1-
ValueError: Error when checking input: expected conv1d_1_input to have shape (None, 500000, 3253) but got array with shape (500000, 3253, 1)
2-
With model.fit()
How do I resolve this problem?
The input shape is wrong, it should be input_shape = (1, 3253) for Theano or (3253, 1) for TensorFlow. The input shape doesn't include the number of samples.
Then you need to reshape your data to include the channels axis:
x_train = x_train.reshape((500000, 1, 3253))
Or move the channels dimension to the end if you use TensorFlow. After these changes it should work.
input_shape = (3253, 1)
this must be Input_shape of first Convolution layer Conv1D
You got error with model.fit() Because you still don't build your model yet.
I have a csv file with 339732 rows and two columns :
the first being 29 feature values, i.e. X
the second being a binary label value, i.e. Y
dataframe = pd.read_csv("features.csv", header = None)
dataset = dataframe.values
X = dataset[:, 0:29].astype(float)
Y = dataset[:,29]
X_train, y_train, X_test, y_test = train_test_split(X,Y, random_state = 42)
I am trying to train it on a 1D convolutional layer:
model = Sequential()
model.add(Conv1D(64, 3, activation='relu', input_shape=(X_train.shape[0], 29)))
model.add(Conv1D(64, 3, activation='relu'))
model.add(MaxPooling1D(3))
model.add(Conv1D(128, 3, activation='relu'))
model.add(Conv1D(128, 3, activation='relu'))
model.add(GlobalAveragePooling1D())
model.add(Dropout(0.5))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='binary_crossentropy',
optimizer='rmsprop',
metrics=['accuracy'])
model.fit(X_train, y_train, batch_size=16, epochs=2)
score = model.evaluate(X_test, y_test, batch_size=16)
Since, the Conv1D layer expects a 3-D input, I transformed my input as follows:
X_train = np.reshape(X_train, (1, X_train.shape[0], X_train.shape[1]))
X_test = np.reshape(X_test, (1, X_test.shape[0], X_test.shape[1]))
However, this still throws error:
ValueError: Negative dimension size caused by subtracting 3 from 1 for 'conv1d_1/convolution/Conv2D' (op: 'Conv2D') with input shapes: [?,1,1,29], [1,3,29,64].
Is there any way to feed my input correctly?
As far as I know 1D Convolution layer accepts inputs of the form Batchsize x Width x Channels. You are reshaping with
X_train = np.reshape(X_train, (1, X_train.shape[0], X_train.shape[1]))
But X_train.shape[0] is your batchsize I guess.I think the problem is somewhere here. Can you please tell what is the shape of X_train before reshape?
You have to think about if your data have some progression relation between the 339732 entries or the 29 features, this means if the order matters. If not I don't think that CNN is suitable for this case.
If the 29 features "indicates the progression of something":
X_train = X_train.reshape((X_train.shape[0], X_train.shape[1],1))
If the 29 features are independent, then is like the channels on the image, but doesn't make sense convolute with only 1.
X_train = X_train.reshape((X_train.shape[0],1, X_train.shape[1]))
If you want to pick the 339732 entries like in blocks where the order matters (clip the 339732 or add zero padding in order to be divisible by timesteps):
X_train = X_train.reshape((int(X_train.shape[0]/timesteps),timesteps, X_train.shape[1],1))
I'm trying to build an autoencoder using Keras, based on [this example][1] from the docs. Because my data is large, I'd like to use a generator to avoid loading it into memory.
My model looks like:
model = Sequential()
model.add(Convolution2D(16, 3, 3, activation='relu', border_mode='same', input_shape=(3, 256, 256)))
model.add(MaxPooling2D((2, 2), border_mode='same'))
model.add(Convolution2D(8, 3, 3, activation='relu', border_mode='same'))
model.add(MaxPooling2D((2, 2), border_mode='same'))
model.add(Convolution2D(8, 3, 3, activation='relu', border_mode='same'))
model.add(MaxPooling2D((2, 2), border_mode='same'))
model.add(Convolution2D(8, 3, 3, activation='relu', border_mode='same'))
model.add(UpSampling2D((2, 2)))
model.add(Convolution2D(8, 3, 3, activation='relu', border_mode='same'))
model.add(UpSampling2D((2, 2)))
model.add(Convolution2D(16, 3, 3, activation='relu'))
model.add(UpSampling2D((2, 2)))
model.add(Convolution2D(1, 3, 3, activation='sigmoid', border_mode='same'))
model.compile(optimizer='adadelta', loss='binary_crossentropy')
My generator:
from keras.preprocessing.image import ImageDataGenerator
train_datagen = ImageDataGenerator(rescale=1./255)
train_generator = train_datagen.flow_from_directory('IMAGE DIRECTORY', color_mode='rgb', class_mode='binary', batch_size=32, target_size=(256, 256))
And then fitting the model:
model.fit_generator(
train_generator,
samples_per_epoch=1,
nb_epoch=1,
verbose=1,
)
I'm getting this error:
Exception: Error when checking model target: expected convolution2d_76 to have 4 dimensions, but got array with shape (32, 1)
That looks like the size of my batch rather than a sample. What am I doing wrong?
The error is most likely due to the class_mode='binary'. It makes the generator produce binary classes, so the output has shape (batch_size, 1), while your model produces a four dimensional output (since the last layer is a convolution).
I guess that you want your label to be the image itself. Based on the source of the flow_from_directory and the DirectoryIterator it uses, it is impossible to do by just changing the class_mode. A possible solution would be along the lines of:
train_generator_ = train_datagen.flow_from_directory('IMAGE DIRECTORY', color_mode='rgb', class_mode=None, batch_size=32, target_size=(256, 256))
def train_generator():
for x in train_iterator_:
yield x, x
Note that I set class_mode to None. It makes the generator to return just the image instead of tuple(image, label). I then define a new generator, that returns the image as both the input and the label.