Search code examples
deep-learningbatch-processingshapeskeras-layer

Need of specifying the input shape in Keras

I am seeking help from put the right form of input_shape,

Let's say if I have the code like:

   G  = ImageGenerator(batch_size,args,****kwargs)

where G is a custom-made data_generator which specify the batch size and then fit the data generated from G into :

m.fit_generator(G ,#some other parameters)

I know at the fist layer of model ,it's necessary to give input_shape.

My question is :

In this situation, where batch size have been declared in generatorG,

Do I have to tell Keras(pass the arguments) when build the first layer :

batch_size=32,input_shape=(256,256)

or just

input_shape=(256,256)

Thanks in advance!

Solution

  • According to Matias Valdenegro :batch size is not an argument of any layer, only input_shape is necessary.