Luux

> Device: Schenker NEO 15 Build: 2020/12 > got the same error 'AttributeError: 'NoneType' object has no attribute 'is_kernel_driver_active'' > after changing the product_id and reinstall it works, thanks >...

I tried a bit around and had the same problem. I played with a Google Colab Notebook and the tf implementation was painfully and unusable slow there :/

> Have you tried going back to SteamVR stable instead if the beta? Same problem, both with Revive 2.1.1 and the 3.0 pre-release

Can confirm this issue, too. Still no news there? Connecting via ssh/WinSCP (using sftp) just works fine...

@ybubnov I can confirm this issue. I tried with `f1 = keras_metrics.f1_score(label=1)` `self.model.compile(optimizer="Adam", loss='binary_crossentropy', metrics=[f1])` and got the same error. My last layer currently is `output_layer = keras.layers.Dense(self.n_classes, activation="softmax")(dense2)` The...

@dberenbaum We want the entire pipeline to be clean. The idea of our pipeline is that we want to define everything relevant in our pipeline configuration (`dvc.yaml`), so that we...

I recently ran into a similar problem. Is there any update on this?

Using paddleocr in a repository which already migrated to langchain >1.0 is not possible right now. Is there any update on this? Ideally, the dependencies could be decoupled a bit,...

For python 3.13+, this PR pinned another version as a workaround: https://github.com/PaddlePaddle/PaddleClas/pull/3327 However, it's still broken for python 3.12.

I tried turning off the option, but the problem still seems to persist. Also, sometimes it's stuck in loading the server infinitely until I change the port in the settings....