Chuck Hastings
Chuck Hastings
There is a `transpose` function in the C++ API that will do this efficiently, it is supported by SG and MG. What we haven't done is discuss what the API...
Add support for weights in the new implementation per #2319
For integer vertex columns we can address this in the new C API.
I believe that all of the plumbing exists in C++ to handle this properly for int32 and int64 types. IMO, there are two options here: * The C API *already*...
isolated vertices are not being captured properly when initializing a Graph with an adjacency matrix
The new graph object construction code can handle this properly as long as the python code calls the construction properly. The legacy graph object construction code in C++ should handle...
NOTE, CI discovered a failure in the `RANDOM_WALKS_TEST`, it is being disabled in another PR. When this implementation is addressed, be sure to enable the test and try and isolate...
While doing this, it would be helpful to reference: https://github.com/NVIDIA/thrust and https://github.com/thrust/cub which are the latest code bases for thrust and cub.
nvcc -std=c++11 -o test_hits test_hits.cu -I ${GUNROCK_INSTALL}/include -I /usr/local/cuda-10.1/include --expt-extended-lambda -L ${GUNROCK_INSTALL}/lib -lgunrock -lgomp
rerun tests