d180cf

Hi Andrew, Do you think it's the best approach to construct and train a simple DCNN in tensorflow and then run the trained NN with convnetjs in a JS app?...

Hello eidogo developers, I'm writing a tsumego solving js library and found that eidogo can simplify my debugging experience: eidogo can render the search tree and help me understand where...

Switching between several variations.

3

Here is a simple tsumego with 3 variations: ``` sgf (;FF[4]SZ[7]MA[ac] C[B => W][B+1 => W][W => W+aa,ab,ba,bb,ca,cb,da,db] AB[fa][fb][fc][ad][bd][cd][dd][ed][fd] AW[ea][eb][ac][bc][cc][dc][ec] SL[aa][ab][ba][bb][ca][cb][da][db] (;AB[aa][ca][cb][bb] AW[ab] C[B => W][B+1 => B+da,db][W+1 => W])...

The .innerSGF property.

1

As we know, in DOM every element has dual nature: 1. On one side it's a root of its DOM tree: we can traverse the subnodes without parsing anything, we...

Currently in the edit mode wgo can only add b/w stones and doesn't even let add a few b stones in a row. It would be more convenient if wgo...

Here is an interesting [paper](http://www.cs.cityu.edu.hk/~hwchun/research/PDF/Julian%20WONG%20-%20CCCT%202004%20a.pdf) that explains how to construct a simple NN to guess the correct move in 97% of cases. If mixed with a shallow 4-5 moves deep...

So the search could be stopped in the middle and the solver would still return something useful.