DAVIAN-Robotics/DiscoDemo-Stage1_RL-PnP_Banana

Reinforcement Learning • Updated 10 days ago • 74

Note PnP-Banana: RL data generator, DiscoDemo (alpha = 0.3).

DAVIAN-Robotics/DiscoDemo-Stage1_RL-PnP_Banana-alpha0

Reinforcement Learning • Updated 10 days ago • 46

Note PnP-Banana: RL data generator, P-RFCL baseline (alpha = 0).

DAVIAN-Robotics/DiscoDemo-Stage1_RL-StackCube

Reinforcement Learning • Updated 10 days ago • 44

Note StackCube: RL data generator, DiscoDemo (alpha = 0.3).

DAVIAN-Robotics/DiscoDemo-Stage1_RL-StackCube-alpha0

Reinforcement Learning • Updated 10 days ago • 56

Note StackCube: RL data generator, P-RFCL baseline (alpha = 0).

DAVIAN-Robotics/DiscoDemo-Stage1_RL-FMB_Round

Reinforcement Learning • Updated 10 days ago • 55

Note FMB-Round: RL data generator, DiscoDemo (alpha = 0.3).

DAVIAN-Robotics/DiscoDemo-Stage1_RL-FMB_Round-alpha0

Reinforcement Learning • Updated 10 days ago • 57

Note FMB-Round: RL data generator, P-RFCL baseline (alpha = 0).

DAVIAN-Robotics/DiscoDemo-Stage1_RL-FMB_SqCircle

Reinforcement Learning • Updated 10 days ago • 52

Note FMB-SqCircle: RL data generator, DiscoDemo (alpha = 0.3).

DAVIAN-Robotics/DiscoDemo-Stage1_RL-FMB_SqCircle-alpha0

Reinforcement Learning • Updated 10 days ago • 54

Note FMB-SqCircle: RL data generator, P-RFCL baseline (alpha = 0).

Note PnP-Banana: 3,000 generated demonstrations (LeRobot v3.0), DiscoDemo (alpha = 0.3).

Note PnP-Banana: 3,000 generated demonstrations (LeRobot v3.0), P-RFCL baseline (alpha = 0).

Note StackCube: 3,000 generated demonstrations (LeRobot v3.0), DiscoDemo (alpha = 0.3).

Note StackCube: 3,000 generated demonstrations (LeRobot v3.0), P-RFCL baseline (alpha = 0).

Note FMB-Round: 3,000 generated demonstrations (LeRobot v3.0), DiscoDemo (alpha = 0.3).

Note FMB-Round: 3,000 generated demonstrations (LeRobot v3.0), P-RFCL baseline (alpha = 0).

Note FMB-SqCircle: 3,000 generated demonstrations (LeRobot v3.0), DiscoDemo (alpha = 0.3).

Note FMB-SqCircle: 3,000 generated demonstrations (LeRobot v3.0), P-RFCL baseline (alpha = 0).

DAVIAN-Robotics/pi05_droid_jointpos

Robotics • 4B • Updated 28 days ago • 755

Note Base model of the Stage3 policies: pi0.5 DROID checkpoint with joint-position actions.

DAVIAN-Robotics/DiscoDemo-Stage3_SFT-PnP_Banana

Robotics • 4B • Updated 10 days ago • 74 • 1

Note PnP-Banana: pi0.5 policy fine-tuned on the generated data, DiscoDemo (alpha = 0.3).

DAVIAN-Robotics/DiscoDemo-Stage3_SFT-PnP_Banana-alpha0

Robotics • 4B • Updated 10 days ago • 65

Note PnP-Banana: pi0.5 policy fine-tuned on the generated data, P-RFCL baseline (alpha = 0).

DAVIAN-Robotics/DiscoDemo-Stage3_SFT-StackCube

Robotics • 4B • Updated 10 days ago • 78

Note StackCube: pi0.5 policy fine-tuned on the generated data, DiscoDemo (alpha = 0.3).

DAVIAN-Robotics/DiscoDemo-Stage3_SFT-StackCube-alpha0

Robotics • 4B • Updated 10 days ago • 82

Note StackCube: pi0.5 policy fine-tuned on the generated data, P-RFCL baseline (alpha = 0).

DAVIAN-Robotics/DiscoDemo-Stage3_SFT-FMB_Round

Robotics • 4B • Updated 10 days ago • 69

Note FMB-Round: pi0.5 policy fine-tuned on the generated data, DiscoDemo (alpha = 0.3).

DAVIAN-Robotics/DiscoDemo-Stage3_SFT-FMB_Round-alpha0

Robotics • 4B • Updated 10 days ago • 66

Note FMB-Round: pi0.5 policy fine-tuned on the generated data, P-RFCL baseline (alpha = 0).

DAVIAN-Robotics/DiscoDemo-Stage3_SFT-FMB_SqCircle

Robotics • 4B • Updated 10 days ago • 66

Note FMB-SqCircle: pi0.5 policy fine-tuned on the generated data, DiscoDemo (alpha = 0.3).

DAVIAN-Robotics/DiscoDemo-Stage3_SFT-FMB_SqCircle-alpha0

Robotics • 4B • Updated 10 days ago • 63

Note FMB-SqCircle: pi0.5 policy fine-tuned on the generated data, P-RFCL baseline (alpha = 0).

Note PnP-Banana: simulator states of the generated demonstrations, DiscoDemo and P-RFCL.

Note StackCube: simulator states of the generated demonstrations, DiscoDemo and P-RFCL.

Note FMB-Round: simulator states of the generated demonstrations, DiscoDemo and P-RFCL.

Note FMB-SqCircle: simulator states of the generated demonstrations, DiscoDemo and P-RFCL.