frugally-deep: ValueError: Refusing to perform an overparameterized separable convolution

Hi,

I try to use your awesome solution. few questions:

I have below error when I ran :

ValueError: Refusing to perform an overparameterized separable convolution: channel_multiplier * in_channels = 1 * 2 = 2 > 1 = out_channels

my training code: `from keras.preprocessing.image import ImageDataGenerator from keras.models import Sequential from keras.layers import Conv2D, MaxPooling2D from keras.layers import Activation, Dropout, Flatten, Dense from keras.callbacks import EarlyStopping, ModelCheckpoint from keras import backend as K from model import create_model from math import ceil

model = create_model() model.compile(loss=‘categorical_crossentropy’, optimizer=‘rmsprop’, metrics=[‘accuracy’])

`

is this output usable from your code or need to change ie:optimizer or not supported . ?

Best