Initial commit

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,48 @@
+---
+license: apache-2.0
+tags:
+- robotics
+- reinforcement-learning
+- discodemo
+- franka
+---
+
+# DiscoDemo-Stage1_RL-PnP_Banana
+
+The **data generator** of [DiscoDemo](https://davian-robotics.github.io/DiscoDemo/) for the **PnP-Banana** task (pick up a banana and place it in a bowl): a state-based
+reinforcement-learning policy trained in simulation — the DiscoDemo generator (diversity weight α = 0.3).
+Rolling it out generates demonstrations such as those in the Stage2 dataset below.
+
+| Stage | Repository |
+|---|---|
+| Generator (RL policy) | [`DAVIAN-Robotics/DiscoDemo-Stage1_RL-PnP_Banana`](https://huggingface.co/DAVIAN-Robotics/DiscoDemo-Stage1_RL-PnP_Banana) (this repository) |
+| Generated dataset | [`DAVIAN-Robotics/DiscoDemo-Stage2_GenData-PnP_Banana`](https://huggingface.co/datasets/DAVIAN-Robotics/DiscoDemo-Stage2_GenData-PnP_Banana) |
+| Imitation policy (π0.5) | [`DAVIAN-Robotics/DiscoDemo-Stage3_SFT-PnP_Banana`](https://huggingface.co/DAVIAN-Robotics/DiscoDemo-Stage3_SFT-PnP_Banana) |
+
+## Files
+
+| File | Content |
+|---|---|
+| `actor.pt` | Actor network weights (`network_state_dict`). Optimizer state is not included. |
+| `normalizer.pt` | Observation bounds used to normalize the policy input. |
+| `config.yaml` | Full training configuration (environment, curriculum, agent). |
+
+Only what is needed to roll the policy out is released; the critic and other training-only state are omitted.
+The actor is conditioned on a skill vector *z* sampled once per episode from a standard normal distribution.
+
+Environment steps at this checkpoint: 50.0 M.
+
+## Usage
+
+Loading requires the DiscoDemo training code: https://github.com/DAVIAN-Robotics/DiscoDemo.
+
+## Citation
+
+```bibtex
+@article{park2026discodemo,
+  title   = {DiscoDemo: Discovering Efficient and Diverse Robot Demonstrations for Imitation Learning},
+  author  = {Park, Minho and Kim, Kinam and Kim, Donghu and Lee, Byungkun and Hwang, Dongyoon and Shin, Yongjae and Hyung, Junha and Lee, Hojoon and Choo, Jaegul},
+  journal = {arXiv preprint},
+  year    = {2026}
+}
+```

actor.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d66e08850b10b0b972918d996c32caebab53133215ef9adb7aab94c56f754da4
+size 1110721

config.yaml

@@ -0,0 +1,147 @@
+project_name: DiscoDemo
+entity_name: null
+group_name: pnp_banana
+exp_name: discodemo-pnp_banana
+seed: 0
+logger_type: wandb
+save_path: libs/FlashSAC/models/pnp_banana/discodemo-pnp_banana/seed0-TIMESTAMP
+agent_load_path: null
+num_env_steps: 50000000
+num_train_envs: 2048
+updates_per_interaction_step: 64
+gamma: 0.99
+n_step: 8
+env:
+  workcell: libs/real2sim/assets/workcell.json
+  task: libs/real2sim/real2sim/tasks/pnp_banana.py
+  env_version: fr3-pnp_banana
+  gripper_lock: null
+success_hold_steps: 20
+agent:
+  agent_type: flashSAC
+  seed: 0
+  device_type: cuda
+  buffer_max_length: 100000000
+  buffer_min_length: 5000
+  buffer_device_type: cuda
+  sample_batch_size: 2048
+  normalize_reward: true
+  normalized_G_max: 5.0
+  asymmetric_observation: false
+  learning_rate_init: 0.0003
+  learning_rate_peak: 0.0003
+  learning_rate_end: 0.00015
+  learning_rate_warmup_rate: 1.0e-06
+  learning_rate_warmup_step: 50
+  learning_rate_decay_rate: 1.0
+  learning_rate_decay_step: 50000000
+  actor_num_blocks: 2
+  actor_hidden_dim: 128
+  actor_bc_alpha: 0
+  actor_noise_zeta_mu: 2.0
+  actor_noise_zeta_max: 16
+  actor_update_period: 2
+  critic_num_blocks: 2
+  critic_hidden_dim: 256
+  critic_num_bins: 101
+  critic_min_v: -5
+  critic_max_v: 5
+  critic_target_update_tau: 0.01
+  temp_initial_value: 0.01
+  temp_target_sigma: 0.15
+  temp_target_entropy: null
+  gamma: 0.99
+  n_step: 8
+  use_compile: true
+  compile_mode: auto
+  use_amp: true
+  load_optimizer: true
+  load_reward_normalizer: true
+rfcl:
+  demo_ratio: 0.5
+  truncate_after_success: 50
+  reverse:
+    reverse_step_size: 16
+    advance_threshold: 0.9
+    frontier_window: 10
+    minimum_episode_steps: 30
+  solved_frac_threshold: 0.5
+  demo_path: data/demo_banks/pnp_banana/human_demos.h5
+  init_pose: box:0.4125,0.6375,-0.15,0.15,180,0.20
+eval_worker:
+  num_envs: 50
+skill:
+  phi_hidden:
+  - 256
+  - 256
+  phi_lr: 0.0001
+  dual_slack: 0.001
+  lam_init: 1.0
+  alpha: 0.3
+  z_dim: 3
+safety_penalty:
+  illegal_contact:
+    weight: 0.0001
+    sat: 100
+    robot_bodies:
+    - panda_link5
+    - panda_link6
+    - panda_link7
+    - panda_hand
+    - panda_leftfinger
+    - panda_rightfinger
+    scene_bodies:
+    - table
+    - bowl
+    - banana
+    allowed_pairs:
+    - - panda_leftfinger
+      - banana
+    - - panda_rightfinger
+      - banana
+obs_bounds:
+  object_z_m:
+  - 0.0
+  - 0.6
+  eef_box_m:
+    x:
+    - 0.2
+    - 0.85
+    'y':
+    - -0.45
+    - 0.45
+    z:
+    - 0.0
+    - 0.7
+  object_xy:
+  - - 0.2
+    - 0.85
+  - - -0.4
+    - 0.4
+task_name: pnp_banana
+jerk_limited_action:
+  max_joint_velocity:
+  - 0.140158
+  - 0.371417
+  - 0.120833
+  - 0.401634
+  - 0.158933
+  - 0.283217
+  - 0.239493
+  max_joint_acceleration:
+  - 0.270445
+  - 0.596802
+  - 0.280726
+  - 0.625579
+  - 0.64258
+  - 0.654981
+  - 0.774423
+  max_joint_jerk:
+  - 1.706102
+  - 2.779986
+  - 1.697314
+  - 2.944153
+  - 4.787226
+  - 4.222746
+  - 5.887756
+method_name: discodemo

normalizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:797bdde550c0940effb74946a53223a6a23eef20e99c3a84b075d9409b3e5167
+size 2492