-
current reference GEMM implementation does not support multi-threading, cache optimization techniques, SIMD and many other optimization techniques used in efficient impl. of GEMM.
There are multip…
-
Current implementation of in compute-engine is as following:
- im2col
- bitpack the im2col matrix and filter matrix
- BGEMM
However, it makes sense to bitpack the input matrix first along the …
-
Support for many of the interesting instructions like popcnt (technically SSE4a) could be an interesting addition and prove to be useful to avoid using unmanaged code in certain performance sensitive …
-
It might make sense to have a `packed` datatype, or maybe `packed8`, `packed32` etc which is really just `uint8` / `uint32` under the hood.
This way, we can have the bitpacking op that sends `float` …
-
When running `cargo install --no-default-features --features postgres
`
```
error: failed to compile `plume v0.3.0 (/home/plume/Plume)`, intermediate artifacts can be found at `/home/plume/Plum…
-
We should prepare for GSoC 2020 soon - registration of projects under the PSF umbrella [closes soon (2nd of Feb)](https://python-gsoc.org/index.html#apply). We should review and update [our ideas page…
-
A representation like the following would be smaller and eliminate the need for `BoardVisitor` and perhaps also `GroupVisitor`.
```C++
class Position {
public:
// Returns the color of the cha…
-
Below is the first draft of an IPLD Schema for a UnixFSv2 that we could build and ship this year.
The implementation of the advanced layouts is yet to be defined.
If this seems reasonable I’ll t…
-
-
There seems to be a copy-past error here, the index should be different from one another https://github.com/tantivy-search/bitpacking/blob/8cc0a1700e690de97b0bb4e0fb205024c0deaf3c/src/bitpacker8x.rs#…