tastyminerals

In your `predict` function find `z1 = X.dot(W1) + b1` and replace `X.dot` to `x.dot`.

It would be nice to evaluate **deeplearning4j**. They've actually done a [short review](http://deeplearning4j.org/compare-dl4j-torch7-pylearn.html) of popular deep learning frameworks vs deeplearning4j.

Nope, doesn't work. However, if I do the following: ```d private void selectText(CommandArgs args) { this._clientId.selectText; this._clientPass.selectText; //

I've only just started so there isn't much code, here is a test sample: ```d import std.stdio; import std.format; import tkd.tkdapplication; class App : TkdApplication { private Frame _loginFrame; private...

> I've taken a look and i'm stumped. It looks like the callback is firing correctly but the selection is failing for some reason. There also looks to be another...

I found my old Tkinter code that tackles this problem. I am going to try this with D tomorrow. ```python def ctrl_a(self, callback=False): """ Select all in entry or text...

Well, I was overly optimistic about it. It's not possible because `.bind` expects void function. I guess it's a bug then.

Same problem here :(

I managed to get it working by modifying the makefile though.

Actually just minor changes: ```diff [I] pavels@Pavels-MacBook-Pro ~/D/g/DCD (master)> git diff (base) diff --git a/makefile b/makefile index 0adeb83..b8e1af9 100644 --- a/makefile +++ b/makefile @@ -1,6 +1,6 @@ .PHONY: all -all:...