-
Can wasmedge now support model training?
-
Looking at the Markdown/HTML produced for wasi-nn, this tool seems to confuse the Markdown parser whether it should convert several lines to Markdown or print them as HTML: [`ml.md`](https://github.co…
-
How can I obtain inference output as streaming tokens so that I can pipe the output stream to the user and also calculate the number of tokens per second afterward? I can only find examples that retur…
-
### Summary
in my summarization task, I need to set the max_token parameter so that llm can stop generation at the length I need. Now llm seems continue generating to the n_predict length, which is…
-
It sounds obvious that we can compile ggml into wasm with WebGPU or Webgl2 support.
Has anyone done it and written about how to do it?
Alternatively, one can assist with some idea how to do it.
-
### Summary
I have encountered a bug in one of WasmEdge's example plugins, specifically in the "get-string" **(WasmEdge/examples/plugin/get-string/)** example. While trying to test my new plugin for …
-
Currently the wasi-nn API only allows loading ML models from their byte-serialized format (i.e., using `load`). This can be problematic for several reasons:
- models may be large — "gigabytes" …
-
I notice that the wasi-nn only run at host as described in `pytorch-mobilenet-image/README.md` without any description for docker. I wonder whether docker supports for wasi-nn with Pytorch backend. Do…
-
# Motivation
[Mediapipe](https://developers.google.com/mediapipe) is a collection of ML models for streaming data. The official website provides Python, iOS, Android, and TFLite-JS SDKs for using t…
-
## Motivation
The install script for OpenVINO is still using the 2021 edition. However, this year is 2023. We should update the version.
## Details
Please update https://github.com/WasmEdge/W…