Guillaume Lagrange

Btw, not sure if anyone has delved into the SAM code for ONNX export but it doesn't include all the operations to actually run the model for an input image....

Current state of required ops based on the latest PRs: | op_type | Burn | Import | | :----------------- | :--: | :----: | | Add | ✅ | ✅...

While most of the current discussion has been around 0-dim tensors for scalar support, we also had two other supporting arguments around that on discord this morning for 0-length or...

> @laggui , you mentioned you were working in this area. Do you think we are close making it work? Bool is now an element type :) but the methods...

According to the line that panics it seems to be due to an `Unsqueeze` op. Looks like this op is not full supported yet (#600). Do you know what else...

For wgpu you might need to install the correct graphics drivers depending on what you want to target. By default, the `AutoGraphicsApi` targets metal on MacOs and vulkan on other...

The API has changed since the initial reply 😅 To configure the wgpu backend to use a different runtime you need to initialize it (taken from [the docs](https://github.com/tracel-ai/burn/blob/main/crates/burn-wgpu/src/lib.rs#L25-L38)): ```rust burn::backend::wgpu::init_sync::(&device,...

Btw my initial response with an example to target a specific wgpu target (e.g., OpenGL) was not meant to say "you should use OpenGL")😅 It was just meant to illustrate...

Ahhhh I see! In this case, you can use a different backend 🙂 We have our own cuda backend (enabled via `cuda-jit` feature flag), but also candle that supports cuda...

Ahhh that doesn't seem to be a standard installation. Probably it's missing from the PATH or LD_LIBRARY_PATH so `cudarc` cannot find it. I've seen [issues](https://github.com/coreylowman/cudarc/issues/246) in the past but these...