Jinhua Wang

@MaartenGr What I don't understand is why the dimensionality of the UMAP matrix is `(11580087, 3402)`. I understand that `11580087` is the document size, but shouldn't the size of the...

也可以用Dijkstra算法求解： ```python from queue import PriorityQueue class Solution: def minPathSum(self, grid: List[List[int]]) -> int: sol = [ [0] * len(grid[0]) for _ in grid ] visited = set() queue =...

这个应该就是Kadane's algorithm

> Is there any good way to use dmtcp with software that outputs files to disk? In my recent tests, if I run a program that writes large files to...

Is Spectral Mixture Kernel more memory consuming? It seems that I will incur an out-of-memory warning with about 1000 data points and 54 features, where a RBF kernel works just...

同问

Or rather, is there a way to train with mini-batches?

Thanks @sschoenholz and @jaehlee ! I will explore and let you know.

@sschoenholz I assume that, even if I use GPyTorch to access the NT mini-batch kernel, I would still need 6TB memory space, as pointed out by @jaehlee ?

Thanks!