How to perform feed forward propagation in CNN using Keras?

Question

I want to perform feed forward propagation on CNN using Keras. I am trying to train CNN using my own optimizer, which I can't fit in the optimiser file of Keras. My optimiser in gradient free. I don't want any inbuilt to be used.

Gradient free means no gradients are calculated. Basically Evolutionary Algorithms(EAs) are considered to be gradient free. — Bhaskar Dhariyal, Nov 11 '17 at 08:58
OK. But how does it relate to convolutional neural networks? — Maxim, Nov 11 '17 at 09:01
I think that will be beyond the scope of question. I am open to discussion if you want to discuss personally. Please mail at bdhariyal94@gmail.com. — Bhaskar Dhariyal, Nov 11 '17 at 09:08
With respect to question, I don't want any optimiser to be involved now, I want to do it manually. — Bhaskar Dhariyal, Nov 11 '17 at 09:15

score 0 · Accepted Answer · answered Nov 13 '17 at 06:15

I found answer to this question. We just have to make model non trainable.

import numpy as np
import keras

x = keras.layers.Input(shape=(3,))
y = keras.layers.Dense(5)(x)

model = keras.models.Model(x, y)
model.trainable = False
model.compile(optimizer='rmsprop', loss='mse')

x = np.random.random((10, 3))
y = np.random.random((10, 5))
model.fit(x, y, epochs=10)

How to perform feed forward propagation in CNN using Keras?

1 Answers1

Linked