DNN for MNIST dataset with random graph architecture
Usage:
-
step1: python gengraph,oy <hidden layer 0> <hidden layer 1> ...
-
step2: python gennn.py graph_<hidden layer 0><hidden layer 1> ... .dat
-
step3: python3 randnn__.py
Example:
% python3 gengraph.py 64 32
% python3 python3 gennn.py graph_64_32.dat
% python3 randnn_96_52544.py