Fractional max-pooling in keras

Question

The existing function in keras lib including max-pooling, average pooling, etc.

However, I would like to implement fractional max-pooling in keras based on the paper https://arxiv.org/abs/1412.6071.

My implementation are as follow:

model = Sequential()
......
model.add(Conv2D(32, (3, 3)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))

So, instead of model.add(MaxPooling2D(pool_size=(2, 2))), I would like to implement something like the following:

model.add(fractionalMaxpool2D(..............))

Is it possible? I am currently using keras as backend in tensorflow.

Appreciate if someone would provide the algorithm/code.

I am quite new to this as I didn't wrote any custom layer before so could anyone kindly help out? Thanks!

You need to be more specific. Do you want people to implement the algorithm for you? — Pablo, Jun 29 '17 at 01:15
Yes! That would be nice. Alright I will edit the post to be more specific! thanks — Jacob, Jun 29 '17 at 01:20
your questions is answered [here](https://stackoverflow.com/questions/44991470/using-tensorflow-layers-in-keras). — cLottzen, Jul 17 '17 at 06:19
Not ideal, but you can alternate between upscale and downscale to effectively get 2/3 and 3/2. — ldmtwo, Oct 28 '18 at 17:08

score 0 · Answer 1 · answered Jun 29 '17 at 06:29

In my opinion, you can do that by implementing your custom layer

class FractionalMaxpool2D(Layer):
    def __init__(self, output_dim):
        super(FractionalMaxpool2D, self).__init__()
        self.output_dim = output_dim
    def build(self, input_shape):
        # Create a trainable weight variable for this layer.
        # This kind of layer doesn't have any variable
        pass
    def call(self, x):
        # Handle you algorithm here
        return ....  
    def compute_output_shape(self, input_shape):
        # return the output shape
        return (input_shape[0], self.output_dim)

The problem is it's difficult to implement the core function for the Fractional max pooling that uses GPU. Please check this discussion from Keras's Github.

Alright thank for the advice! As i am new to it, would you kindly explain what's going on with the code? thanks — Jacob, Jun 29 '17 at 12:55
I means that you should create the custom layer for your purpose. The custom layer is extended from Layer class. Some important method in your layer is: - build: create the trainable variables if it is needed. - call: perform your calculation to get the output - compute_output_shape: return the output shape (what is the shape of data after it goes through your layer) — An Phú, Jun 30 '17 at 03:14

score 0 · Answer 2 · answered Mar 22 '20 at 18:29

You Can Use Keras Lambda Layer to Wrap tf.nn.fractional_max_pool, like

FMP = Lambda(lambda img, pool_size: tf.nn.fractional_max_pool(img, pool_size))

Now You can Use FMP in your Keras Code like other layers with Two Arguments

Img: with dimensions like [batch, height, width, channels]
pool_size: [1.0, pool_size_you_want, pool_size_you_want, 1.0]

The first and last are 1.0, which is because tf doesnot perform pooling on batch_size and channels, it performs on height and width

Fractional max-pooling in keras

2 Answers2