what is the difference between Flatten() and GlobalAveragePooling2D() in keras

tensorflow machine-learning keras deep-learning keras-layer
user239457 picture user239457 · Mar 15, 2018 · Viewed 17.3k times · Source

I want to pass the output of ConvLSTM and Conv2D to a Dense Layer in Keras, what is the difference between using global average pooling and flatten Both is working in my case.

model.add(ConvLSTM2D(filters=256,kernel_size=(3,3)))
model.add(Flatten())
# or model.add(GlobalAveragePooling2D())
model.add(Dense(256,activation='relu'))

Answer

Dr. Snoopy picture Dr. Snoopy · Mar 15, 2018

That both seem to work doesn't mean they do the same.

Flatten will take a tensor of any shape and transform it into a one dimensional tensor (plus the samples dimension) but keeping all values in the tensor. For example a tensor (samples, 10, 20, 1) will be flattened to (samples, 10 * 20 * 1).

GlobalAveragePooling2D does something different. It applies average pooling on the spatial dimensions until each spatial dimension is one, and leaves other dimensions unchanged. In this case values are not kept as they are averaged. For example a tensor (samples, 10, 20, 1) would be output as (samples, 1, 1, 1), assuming the 2nd and 3rd dimensions were spatial (channels last).

Related questions

How do you make TensorFlow + Keras fast with a TFRecord dataset?

What is an example of how to use a TensorFlow TFRecord with a Keras Model and tf.session.run() while keeping the dataset in tensors w/ queue runners? Below is a snippet that works but it needs the following improvements: …

Read More
What is the role of "Flatten" in Keras?

I am trying to understand the role of the Flatten function in Keras. Below is my code, which is a simple two-layer network. It takes in 2-dimensional data of shape (3, 2), and outputs 1-dimensional data of shape (1, 4): model = Sequential() model.add(…

Read More
How to apply gradient clipping in TensorFlow?

Considering the example code. I would like to know How to apply gradient clipping on this network on the RNN where there is a possibility of exploding gradients. tf.clip_by_value(t, clip_value_min, clip_value_max, name=…

Read More