Jimntu

Hi, when I try to implement the code of bert extraction experiment, the eval accuracy for both MNLI and SST-2 are lower than 50 percent. I doubt that the weight...

I encounter a very weird bug when I run LLaVa_OneVision_Tutorial.ipynb ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [0,0,0], thread: [63,0,0] Assertion `-sizes[i]

Hi, may I ask is there any ways to get the frame-level meta information so that I could get the object appearance order from RGB-D video in ARKitScenes? Many thanks!

Action Precision in pd_ee_delta_pose control mode

1

In the environment that uses the pd_ee_delta_pose control mode, my understanding is that tcp_pose_t+1 = tcp_pose_t + delta action However, when I print out both the action and the resulting...

Maniskill 3 in rlds format

2

Hi, thanks for the great work! I was wondering if there's a way to convert ManiSkill3 datasets into the RLDS format so I can fine-tune OpenVLA on them. Thanks again!