I want to make a copy of a Conv2D layer.
I tried this:
Edit: I've changed the example code to a mcve
Edit2: I've changed the code based on fuglede 's answer
from keras.models import Sequential
from keras.layers import Dense, Conv2D, Flatten
from keras.datasets import mnist
from keras.utils import to_categorical
import matplotlib.pyplot as plt
import numpy as np
import random
(X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(60000, 28, 28, 1)
X_test = X_test.reshape(10000, 28, 28, 1)
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)
model = Sequential()
model.add(Conv2D(random.randint(32, 64), kernel_size=random.randint(1, 3), activation='relu', input_shape=(28, 28, 1)))
model.add(Conv2D(32, kernel_size=3, activation='relu'))
model.add(Flatten())
model.add(Dense(10, activation='softmax'))
model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
other_model = Sequential()
layer = model.layers[1]
other_model.add(Conv2D(random.randint(32, 64), kernel_size=random.randint(1, 3), activation='relu', input_shape=(28, 28, 1)))
copy_layer = Conv2D(layer.filters, kernel_size=layer.kernel_size, activation='relu')
other_model.add(copy_layer)
copy_layer.set_weights(layer.get_weights())
But I'm getting this error:
ValueError: Layer weight shape (3L, 3L, 61L, 32L) not compatible with provided weight shape (3L, 3L, 40L, 32L)
Edit: The purpose of this is, I'm using a genetic algorithm to evolve/"train" a set a neural networks, and this a part of the crossover step.
This happens because the layer is only initialized once it's added to a model. If you swap the two last lines of your example, it should work as expected.
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.