-
On an i7-8550u, OpenBLAS is defaulting to 8 threads. I was comparing to RecursiveFactorizations.jl, and saw the performance is like:
```julia
using BenchmarkTools
import LinearAlgebra, RecursiveF…
-
I guess that in order to call pin()/unpin() in a control block, the latch must be held, otherwise there is no synchronization with other parts of the code that manually set the pin_cnt of a control bl…
-
A follow-up on the issues raised in https://github.com/PointCloudLibrary/pcl/pull/2171#issuecomment-364926779 .
## Ideas
- Start boilerplating on `pcl::experimental::VoxelGrid` so that we can sp…
-
The various [AVX FPU](https://de.wikipedia.org/wiki/Advanced_Vector_Extensions) extensions for x86 CPUs can be used for various media centered and/or in general mathematical optimized work load (besid…
-
Reference:
https://github.com/orientechnologies/orientdb-labs/blob/master/OEP_9.md
**Summary:**
Use 16k pages for next versions of OrientDB by default
**Goals:**
Improve speed of read/writes ope…
-
### Summary
Currently slashing makes two state reads per validator, always one write, sometimes two writes, every block. This causes excess time for:
- Fetching this data
- IAVL committing to thi…
-
Most of the code is generic enough to run on a different framework other than Scalding. However, there is some dependency on the new Execution module of Scalding that I couldn't completely get around.…
-
Currently, users should install the native Menoh Core library manually in addition to menoh-java itself. Providing supplemental JARs including the precompiled native libraries for each platform (Linux…
-
-
## NMF
NMF demonstrates his implicit incrementalization abilities to automatically create a binary decision tree from a truth table.
- Only looks at the BDT structure and creates this ones
- No…