• Implement LoRA fine tune OpenVLA
    • Reach Task
      • 1k
        • Collect sample gifs
        • Mean performance
      • 5k
        • Collect sample gifs
        • Mean performance
      • 25k
        • collect data
        • Collect sample gifs
        • Mean performance
      • 50k
        • collect data
        • train
        • Collect sample gifs
        • Mean performance
      • 75k
        • collect data
        • train
        • Collect sample gifs
        • Mean performance
    • Grasp Task
      • 1k
        • Collect sample gifs
        • Mean performance
      • 5k
        • Collect sample gifs
        • Mean performance
      • 25k
        • Collect sample gifs
        • Mean performance
  • Implement Full-finetune OpenVLA
    • Reach Task
      • 25k
        • train
        • Collect sample gifs
        • Mean performance
      • 50k
    • Grasp Task
      • 25k