Initial commit

.hydra/config.yaml

@@ -0,0 +1,271 @@
+_target_: diffusion_policy.workspace.train_diffusion_unet_hybrid_workspace.TrainDiffusionUnetHybridWorkspace
+checkpoint:
+  save_last_ckpt: true
+  save_last_snapshot: false
+  topk:
+    format_str: epoch={epoch:04d}-test_mean_score={test_mean_score:.3f}.ckpt
+    k: 5
+    mode: max
+    monitor_key: test_mean_score
+dataloader:
+  batch_size: 64
+  num_workers: 8
+  persistent_workers: false
+  pin_memory: true
+  shuffle: true
+dataset_obs_steps: 2
+ema:
+  _target_: diffusion_policy.model.diffusion.ema_model.EMAModel
+  inv_gamma: 1.0
+  max_value: 0.9999
+  min_value: 0.0
+  power: 0.75
+  update_after_step: 0
+exp_name: default
+horizon: 32
+keypoint_visible_rate: 1.0
+logging:
+  group: null
+  id: null
+  mode: online
+  name: ${now:%Y.%m.%d}_${now:%H.%M.%S}_train_diffusion_unet_hybrid_aloha_image
+  project: GBC-PBRL-2026
+  resume: true
+  tags:
+  - train_diffusion_unet_hybrid
+  - aloha_image
+  - default
+  entity: Robotics_VLA
+multi_run:
+  run_dir: data/outputs/${now:%Y-%m-%d}/${now:%H-%M-%S}_train_diffusion_unet_hybrid_aloha_image
+  wandb_name_base: ${now:%Y-%m-%d}_${now:%H-%M-%S}_train_diffusion_unet_hybrid_aloha_image
+n_action_steps: 15
+n_latency_steps: 0
+n_obs_steps: 2
+name: dp_cnn_ERate
+obs_as_global_cond: true
+optimizer:
+  _target_: torch.optim.AdamW
+  betas:
+  - 0.95
+  - 0.999
+  eps: 1.0e-08
+  lr: 0.0001
+  weight_decay: 1.0e-06
+past_action_visible: false
+policy:
+  _target_: diffusion_policy.policy.diffusion_unet_hybrid_image_policy.DiffusionUnetHybridImagePolicy
+  cond_predict_scale: true
+  crop_shape:
+  - 128
+  - 128
+  diffusion_step_embed_dim: 128
+  down_dims:
+  - 512
+  - 1024
+  - 2048
+  eval_fixed_crop: true
+  horizon: 32
+  kernel_size: 5
+  n_action_steps: 15
+  n_groups: 8
+  n_obs_steps: 2
+  noise_scheduler:
+    _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
+    beta_end: 0.02
+    beta_schedule: squaredcos_cap_v2
+    beta_start: 0.0001
+    clip_sample: true
+    num_train_timesteps: 100
+    prediction_type: epsilon
+    variance_type: fixed_small
+  num_inference_steps: 100
+  obs_as_global_cond: true
+  obs_encoder_group_norm: true
+  shape_meta:
+    action:
+      shape:
+      - 14
+    obs:
+      states:
+        shape:
+        - 14
+      high_images:
+        shape:
+        - 3
+        - 480
+        - 640
+        type: rgb
+      wrist_left_images:
+        shape:
+        - 3
+        - 480
+        - 640
+        type: rgb
+      wrist_right_images:
+        shape:
+        - 3
+        - 480
+        - 640
+        type: rgb
+shape_meta:
+  action:
+    shape:
+    - 14
+  obs:
+    states:
+      shape:
+      - 14
+    high_images:
+      shape:
+      - 3
+      - 480
+      - 640
+      type: rgb
+    wrist_left_images:
+      shape:
+      - 3
+      - 480
+      - 640
+      type: rgb
+    wrist_right_images:
+      shape:
+      - 3
+      - 480
+      - 640
+      type: rgb
+task:
+  dataset:
+    _target_: diffusion_policy.dataset.aloha_replay_image_dataset.AlohaReplayImageDataset
+    dataset_path: /pfss/mlde/workspaces/mlde_wsp_MGPATH/VLA/lpb_libero/data/aloha/fold_shirt/fold_shirt_demo.hdf5
+    horizon: 32
+    n_obs_steps: 2
+    pad_after: 7
+    pad_before: 1
+    seed: 42
+    shape_meta:
+      action:
+        shape:
+        - 14
+      obs:
+        states:
+          shape:
+          - 14
+        high_images:
+          shape:
+          - 3
+          - 480
+          - 640
+          type: rgb
+        wrist_left_images:
+          shape:
+          - 3
+          - 480
+          - 640
+          type: rgb
+        wrist_right_images:
+          shape:
+          - 3
+          - 480
+          - 640
+          type: rgb
+    use_cache: true
+    val_ratio: 0.1
+  dataset_path: /pfss/mlde/workspaces/mlde_wsp_MGPATH/VLA/lpb_libero/data/aloha/fold_shirt/fold_shirt_demo.hdf5
+  dataset_type: mh
+  env_runner:
+    _target_: diffusion_policy.env_runner.robomimic_image_runner.RobomimicImageRunner
+    crf: 22
+    dataset_path: /pfss/mlde/workspaces/mlde_wsp_MGPATH/VLA/lpb_libero/data/aloha/fold_shirt/fold_shirt_demo.hdf5
+    fps: 10
+    max_steps: 700
+    n_action_steps: 15
+    n_envs: 28
+    n_obs_steps: 2
+    n_test: 50
+    n_test_vis: 4
+    n_train: 6
+    n_train_vis: 2
+    past_action: false
+    render_obs_key: shouldercamera0_image
+    shape_meta:
+      action:
+        shape:
+        - 14
+      obs:
+        states:
+          shape:
+          - 14
+        high_images:
+          shape:
+          - 3
+          - 480
+          - 640
+          type: rgb
+        wrist_left_images:
+          shape:
+          - 3
+          - 480
+          - 640
+          type: rgb
+        wrist_right_images:
+          shape:
+          - 3
+          - 480
+          - 640
+          type: rgb
+    test_start_seed: 100000
+    tqdm_interval_sec: 1.0
+    train_start_idx: 0
+  name: aloha_image
+  shape_meta:
+    action:
+      shape:
+      - 14
+    obs:
+      states:
+        shape:
+        - 14
+      high_images:
+        shape:
+        - 3
+        - 480
+        - 640
+        type: rgb
+      wrist_left_images:
+        shape:
+        - 3
+        - 480
+        - 640
+        type: rgb
+      wrist_right_images:
+        shape:
+        - 3
+        - 480
+        - 640
+        type: rgb
+  task_name: aloha
+task_name: aloha_image
+training:
+  checkpoint_every: 20
+  debug: false
+  device: cuda:0
+  gradient_accumulate_every: 1
+  lr_scheduler: cosine
+  lr_warmup_steps: 500
+  max_train_steps: null
+  max_val_steps: null
+  num_epochs: 500
+  resume: true
+  rollout_every: 20
+  sample_every: 5
+  seed: 42
+  tqdm_interval_sec: 1.0
+  use_ema: true
+  val_every: 1
+val_dataloader:
+  batch_size: 64
+  num_workers: 8
+  persistent_workers: false
+  pin_memory: true
+  shuffle: false

.hydra/hydra.yaml

@@ -0,0 +1,165 @@
+hydra:
+  run:
+    dir: logs/reproduce/${task_name}/${logging.group}/${logging.name}
+  sweep:
+    dir: multirun/${now:%Y-%m-%d}/${now:%H-%M-%S}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra:
+    - hydra.run.dir=logs/reproduce/${task_name}/${logging.group}/${logging.name}
+    - hydra.mode=RUN
+    task:
+    - training.seed=42
+    - name=dp_cnn_ERate
+    - logging.project=GBC-PBRL-2026
+    - +logging.entity=Robotics_VLA
+    - logging.mode=online
+    - task.dataset.val_ratio=0.1
+    - training.num_epochs=500
+  job:
+    name: train
+    chdir: null
+    override_dirname: +logging.entity=Robotics_VLA,logging.mode=online,logging.project=GBC-PBRL-2026,name=dp_cnn_ERate,task.dataset.val_ratio=0.1,training.num_epochs=500,training.seed=42
+    id: ???
+    num: ???
+    config_name: image_aloha_diffusion_policy_cnn.yaml
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.2.0
+    version_base: '1.2'
+    cwd: /pfss/mlde/workspaces/mlde_wsp_MGPATH/VLA/lpb_libero
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /pfss/mlde/workspaces/mlde_wsp_MGPATH/VLA/lpb_libero/diffusion_policy/config
+      schema: file
+      provider: main
+    - path: /pfss/mlde/workspaces/mlde_wsp_MGPATH/VLA/lpb_libero
+      schema: file
+      provider: command-line
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /pfss/mlde/workspaces/mlde_wsp_MGPATH/VLA/lpb_libero/logs/reproduce/aloha_image/None/2026.06.02_03.50.28_train_diffusion_unet_hybrid_aloha_image
+    choices:
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

.hydra/overrides.yaml

@@ -0,0 +1,7 @@
+- training.seed=42
+- name=dp_cnn_ERate
+- logging.project=GBC-PBRL-2026
+- +logging.entity=Robotics_VLA
+- logging.mode=online
+- task.dataset.val_ratio=0.1
+- training.num_epochs=500

checkpoints/40.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0813c4dfe9d7d7ba15135e5d5e85bb59985b8c7ad9b382d9ee88d28e0d29467
+size 4960156329

normalizer.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fececb5e44305002c27babbe7c7f6146b63cc19bdf0d054bdd3811f7a61a3b6
+size 9547

train.log

@@ -0,0 +1 @@
+[2026-06-02 03:50:35,807][diffusion_policy.model.diffusion.conditional_unet1d][INFO] - number of parameters: 2.595831e+08