Tianqi Chen comments

Results 637 comments of


                                            Tianqi Chen

stream_gpu-inl.h:65: Default GPU stream was used when MSHADOW_FORCE_STREAM was on

@antinucleon can you update the example?

python wrapper error

I think you are missing the input_shape parameters, currently, we require the user to specify the input shape of the data. Maybe you can checkout https://github.com/dmlc/cxxnet/blob/master/example/MNIST/mnist.py for more details

how to install memonger

You need to clone this repo and put it under your env variable python path

impracticable in SyncLastestModel(void) function

Oh, yes.. Can you open a PR on this ? Thanks

How to choose a GPU to run

Yes , this is done by Calling cuda set device before hanf

advices on how to understand the code

see https://github.com/dmlc/cxxnet/tree/master/src for a NOTE On code structure

advices on how to understand the code

Unfortunately there was no UML diagram. But I would suggest you to start from the interface headers (ends with .h), which are heavily documented

advices on how to understand the code

If you have specific questions on code structure or want to contribute on the code note, I would be happy to answer the questions

Shuffle in groups for IndexedRecordIO

@piiswrong merge this?

Is it easy to implement a RNN?

RNN was not yet implemented in cxxnet. However, it could be implemented with mshadow easily with matrix operation. A notable thing about V2 is that we enables natural layer weight...