dreamdifferent/panda-robosuite-level5-wrist-ablation-v2-action-decoder-iter900

VAM-Cross MimicVideo World2Action decoder

This public repository contains the World2Action decoder checkpoint from iteration 900 of w2a_panda_robosuite_level5_2cam_wrist_ablation_v2_action_iter2374_videolora_iter200_widowx_teleop_recording_frame_v1. The run stopped because of completed. The newest complete model, optimizer, scheduler, and trainer checkpoint set was verified before selecting the uploaded model weight.

Required frozen inputs

• MimicVideo commit: e3355dbc93132b576c02f920a59b4fc18a4f5906
• Initial Video2World backbone: dreamdifferent/widowx250-wrist-ablation-v1-video-fused@8e39d96344dea0a82ae673874a38206a6c412948
• Initial action decoder: dreamdifferent/widowx250-wrist-ablation-v1-action-decoder-iter2374@0ea6db91672db019d9d6dc9a6c9bd1ecc0504001
• Frozen Video LoRA: dreamdifferent/panda-robosuite-level5-wrist-ablation-v2-video-lora-iter200@4c1f086f1cad8a980016f7bc37042af161d96408

Action and data contract

• Dataset: dreamdifferent/vam-cross-level5-panda-robosuite-widowx-texture-corner-wrist-v2@361fec40a4e63a3285aecacd5e5b77e146056cb7
• Episodes/frames: 166 / 54264
• Cameras: observation.images.corner_cam, observation.images.wrist_cam
• Target: 15 achieved-EE/gripper actions at 5 Hz
• Pose target: relative_to_current_achieved_pose in widowx_reference_base/teleop_aligned_tool
• Rotation: rotation_6d

The dataset and frozen inputs are not included. Use the pinned JSON and effective config.yaml included in this repository.