guillaume-be

Unfortunately the `tch` binding are still needed since the crate relies on it for all tensor operations outside of the model (pre and post-processing). Rewriting the pipelines to have a...

Closed via https://github.com/guillaume-be/rust-bert/pull/282, https://github.com/guillaume-be/rust-bert/pull/322

@Raduc4 glad you found a workaround, it would still be great to have the library working on all linux platforms. I am closing this issue as it is more related...

Hello @edersoncorbari , Yes - this should work using the Bert model in this crate. You would need to convert the weights to the `.ot` format with the script found...

The error message seems to indicate the configuration file is missing -- can you check that all the configuration (`*.json`) files are present in the repository? The dependencies (replacing the...

Hello, It is difficult without seeing the Python code side by side for comparison, but could it be that the Python model is loaded in half precision (fp16)?

Hello @lerouxrgd , Yes the availability of `BetterTransformer` is an interesting development. The challenge for an integration in the library is twofold: 1. a lot of the language models implemented...