Boxi Liu

> @PercyLau I am now aware of any discussions regarding the implementation of sparse-matrices in Numba. Perhaps it would make sense to trawl through our discourse: https://numba.discourse.group/ -- and perhaps...

> @PercyLau and there is also: > > https://sparse.pydata.org/en/stable/index.html > > Which, I believe, uses Numba under the hood. @esc. However, the most of the projects lacks of CSR/CSC sparse...

Hi, Paul Generally, the problem roots in the way that how tensorflow manages gpu/cpu resources. In particular, there are three issues. 1. Some envs only support virtual GPUs (vGPUs) for...

> Hi, > > Thanks for the wrapper info. > > 1. As Tensorflow by defaults eats up nearly all the GPU memory, we'd need to figure out some alternative...

One possible way may be to modify the constructor of gradient class therefore parameters can be passed into objective function type.