Giulio Starace
Giulio Starace
Btw, I think this issue is a duplicate of #3104 (also still open)
aren't 2.0 and 2.1 SD models generally worse in performance than 1.5?
@mmatl @JanusMaple please merge this PR (it would address #26, #4, #30, #27) or comment on the status of this repository. Thank you!
Hi, this is urgently still an issue! What is the workaround? Particularly for PL < 2 users Also sidenote, why is the trainer even loading the state dict from the...
For anyone looking for a temporary hacky workaround, here's what I did. My workflow was ```python # the state dict in ckpt does not contain the pretrained model, because we...
This issue should be reopened, I am still facing the same issue, even with 2.16
fyi, check whether your python was installed via conda, and whether the conda installed is actually the ARM version of conda and not the x86 version of conda. I was...
the issue i mention above is not conda exclusive fwiw. Try running ```terminal python -c "import platform; print(platform.machine())" ``` This should print ``` arm64 ``` If you're on apple silicon....
@joeyespo any thoughts on merging and releasing this PR?
+1 this is because [the networkx dependency is locked at 2.2](https://github.com/mmatl/urdfpy/blob/5466842899b33bd549e8f9e2a9a987bd5e37373b/setup.py#L8), which doesn't support numpy>1.20. Is there any reason to lock networkx at 2.2? This is causing issues in other...