3.15.3 Can you answer these questions?

3.15.3.1 Help: Reclassify
3.15.3.2 Pattern matching problem in version 11.01
3.15.3.3 CUDADot runs slowly and writes "CUDALink experienced a kernel launch failure" with matrix size > 3300
3.15.3.1 Help: Reclassify

http://mathematica.stackexchange.com/questions/128194/help-reclassify

 
Is it possible to improve the current ClassifierFunction with more training data without running the previous training again? 
 
In this way i will be able to deal with very large datasets in chunks.
 
- asked by user13892 5votes
3.15.3.2 Pattern matching problem in version 11.01

http://mathematica.stackexchange.com/questions/128247/pattern-matching-problem-in-version-11-01

 
Consider this simple example: 
 
p1 = ListPlot3D[ 
Table[Sin[j^2 + i], {i, 0, Pi, Pi/5}, {j, 0, Pi, Pi/5}], 
Mesh -> None, Axes -> False, Boxed -> False, ColorFunction -> Hue]; 
p2 =  ...
                                                                                  

                                                                                  
 
- asked by xslittlegrass 8votes
3.15.3.3 CUDADot runs slowly and writes "CUDALink experienced a kernel launch failure" with matrix size > 3300

http://mathematica.stackexchange.com/questions/128124/cudadot-runs-slowly-and-writes-cudalink-experienced-a-kernel-launch-failure-wi

 
I am trying to learn CUDA and start from the help examples. It turnes out that CPU calculation is much faster for matrix sizes < 3200 than CUDA, while CUDADot gives the error for sizes > 3300. I  ...
 
- asked by luu 4votes