Deploy Transformer Models in the Browser with

In this video we will demo how to use #ONNXRuntime web with a distilled BERT model to inference on device in the browser with #JavaScript. This demo is based on the amazing work of our community member Jo Kristian Bergum!
Connect with / jobergum
Source Link: github.com/job...
Blog: / moving-ml-inference-fr...
Model: huggingface.co...
Dataset: huggingface.co...
#machinelearning #transformers #pytorch #onnx #onnxruntime #JavaScript #web

[@seikatsu_ki]

This instructor is actually an angel, thank you Madam for the straightforward tutorial!

[@santoshmohanram536]

Really helpful thanks.

[@jn-zr2ki]

very useful and clear

[@ONNXRuntime]

Thank you!!

[@nemines001]

hi thanks for this video

[@ONNXRuntime]

You're welcome!! 😊

[@Gerald-iz7mv]

Nice. How to convert it to onnx using cuda?

[@amanbishnoi6894]

there is a bug-
InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(int32)) , expected: (tensor(int64))

[@ONNXRuntime]

Hi, your comment is appreciated, and we want to follow up on this. Could you file an issue in ORT repo with details so we can take a look? github.com/microsoft/onnxruntime

[@KrisMerckx]

This is not "in the browser". This is still node.js. That's server technology.