Max Graey

Improve ported implementation

11

I guess this wasm version is port of levenshtein-js. Wdyt about port [fast-levenshtein](https://github.com/hiddentao/fast-levenshtein) instead? It looks like much simpler and faster and don't require store full matrix.

Question about half/float16 format for tensors

6

Which 16-bit float format are preferable for tensor's type? 1) IEEE half-precision 16-bit float (5 bits exponent, 10 bits fraction) 2) bfloat16 (8 bits exponent, 7 bits fraction), (supports by...

AssemblyScript expose full compiler API now

1

AssemblyScript now expose full compiler api. See commits [#1](https://github.com/AssemblyScript/assemblyscript/commit/6ed45bee98e8b644b87d50b40983414351f62311) and [#2](https://github.com/AssemblyScript/assemblyscript/commit/aedc821c099104919500d7f36425616b9a85606e). So fork of AS is not necessary I guess for now cc @lrettig

New search engine?

1

It seems Algolia engine have some issues. What about switching to fully local engine like Flexsearch? Here plugin for vuepress: https://github.com/leo-buneev/vuepress-plugin-fulltext-search Benchmarks of Flexsearch: https://github.com/nextapps-de/flexsearch#benchmark-ranking cc @trusktr

No optimization for wasm?

2

Hello everyone. May I ask something? Why your scaffold tool don't use any optimization for AssemplyScript here: https://github.com/holochain/holochain-cmd/blob/master/src/cli/scaffold/assemblyscript.rs#L24 You could add optimization for speed (just add `-O3` flag to `asc`)...