EBLearn printing outputs of all feature maps and kernels during forward processing of LeNet-5 MNIST

[999SSS]

Good morning, dear respective members of EBLearn!

I am writing to you regarding few questions about forward processing run of LeNet-5 on EBLearn software, if is it possible may I ask you few question about it, I would be very grateful to you for receive a reply! I have currently observed an EBLearn tutorial about how to launch classical LeNet-5 with output of trace window on forward processing steps here (http://eblearn.sourceforge.net/mnist.html), as I have understood this example shows the classical LeNet-5 architecture as shown here (http://eblearn.sourceforge.net/lib/exe/lenet5.png), except that it does not have F6 layer (RBF, Euclidean distance layer with loading of ideal samples of 12*7 images), only Output layer with 10 neurons which is an ordinary MLP. May I ask you how it would be possible to see the numerical representation of all generated kernels, feature maps after each forward processing of each sample MNIST input number image (E.g. Would it be possible to see what numbers contains in Layer C1 6 feature maps of size 28*28 and see what numbers contains in it's 6 kernels of size 5*5, see what numbers contains in S2 layer feature maps of size 14*14, see what numbers contains in C3 layer 16 feature maps of size 10*10 and see what numbers contains in their 60 kernels of 5*5 size, see what numbers contains in 16 feature maps of S4 layer of size 5*5, see what number contains in C5 120 feature maps of size 1*1 and in their 1920 kernels of size 5*5 and finally see what numerical outputs has the Output MLP layer), could you please say generally as all kernels and feature maps are an 2D Matrix Array, is it possible to set configuration file so on each forward processing (not training) of MNIST number to see in Console trace (print) of all this Array's with Output MLP layer? Again I would be very grateful to you for receive a reply! Thank you in advance!