omar2205

So @nestarz created a Deno `onnx_runtime` here https://github.com/nestarz/onnx_runtime

Not sure how. For example, this is my script file ```ts import * as t from 'npm:@xenova/transformers' const pipeline = t.default.pipeline let pipe = await pipeline('sentiment-analysis') let out = await...

Hey, I just tried it, using npm: and esm.sh and both got an error. esm.sh: ``` Uncaught Error: Dynamic require of "../bin/napi-v3/linux/x64/onnxruntime_binding.node" is not supported ``` npm: ``` Uncaught Error:...

Thanks for the ping @xenova, I followed that and it worked, however, I'm met with `Invalid URL:` error. ```ts import { pipeline } from 'npm:@xenova/transformers'; /* warning napi_add_finalizer is not...

I tried to do that before, but it fails: ``` No model specified. Using default model: "Xenova/distilbert-base-uncased-finetuned-sst-2-english". Uncaught TypeError: Invalid URL: 'Xenova/distilbert-base-uncased-finetuned-sst-2-english/tokenizer.json' ```

Here you go ``` No model specified. Using default model: "Xenova/distilbert-base-uncased-finetuned-sst-2-english". Uncaught TypeError: Invalid URL: 'Xenova/distilbert-base-uncased-finetuned-sst-2-english/tokenizer.json' at getSerialization (ext:deno_url/00_url.js:88:11) at new URL (ext:deno_url/00_url.js:383:27) at new Request (ext:deno_fetch/23_request.js:297:25) at Cache.[[[matchAll]]] (ext:deno_cache/01_cache.js:230:11)...

That fixed it. But we got another error 😅 ``` > let out = await pipe('I love transformers!'); An error occurred during model execution: "TypeError: Tensor.data must be a typed...

Yeah looks like it. Is it possible to use this https://github.com/nestarz/onnx_runtime?

After second thought, I don't know how that would even work. It's written for Deno.

Trying using jsdelivr gives `Uncaught DOMException` error