Convolution exercise : dimension of the bias

amaas / stanford_dl_ex

Programming exercises for the Stanford Unsupervised Feature Learning and Deep Learning Tutorial

http://ufldl.stanford.edu/tutorial

MIT License

2.58k stars 1.6k forks source link

Closed tdeboissiere closed 8 years ago

tdeboissiere commented 8 years ago

Hi,

I think the initialisation of the bias in cnnExercise.m should b: b = rand(numFilters, 1);

instead of

b = rand(numFilters);

because in the cnnConvolve exercise, the bias should be of shape (numFilters,1)