How to create and use a convolutional neural net?

[Booley]

Hi, I'm interested in using synaptic for a project but I'm having a tough time figuring out how to use a convolutional neural net. I'm using the following code do a simple create and test:

(def cnn (neural-net [{:type :input :fieldsize [28 28]}
                      {:type :convolution :act-fn :hyperbolic-tangent
                       :feature-map {:size [9 9] :k 3}
                       :pool {:kind :max :size [2 2]}}
                      {:type :fully-connected :act-fn :sigmoid :n 10}]
                     (training :backprop {:learning-rate {:epsilon 0.01}})))

(def row (vec (repeat (* 28 28) 1)))
(def output (vec (repeat 10 1)))

(def trset (training-set [row row] [output output]))
(train cnn trset 3)
@*1
(-> @cnn :training :stats)

However, the stats always say 0 epochs have been run (no matter how long I wait) and no exception is thrown. What's the proper way of creating a training set and using the convolutional neural net?