Upload starVLA checkpoint stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k step 30000

StarVLA/stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k/checkpoints/steps_30000_pytorch_model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d68ffa9e87cdc70ea0cd0dfa87af7671ce378874d769648b554839055174709
+size 5854615645

StarVLA/stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k/config.full.yaml

@@ -0,0 +1,103 @@
+run_id: stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k
+run_root_dir: /scratch/wangpc/DiscreteRTCv2/results/Checkpoints
+seed: 42
+wandb_entity: 401910710-university-of-california-berkeley
+wandb_project: starVLA_piper
+is_debug: false
+version_id: '0.21'
+framework:
+  name: QwenPI_v3
+  qwenvl:
+    base_vlm: ./playground/Pretrained_models/Qwen3-VL-2B-Instruct-Action
+    attn_implementation: flash_attention_2
+    vl_hidden_dim: 2048
+    num_vl_layers: 28
+  action_model:
+    action_model_type: LayerwiseFM
+    action_dim: 7
+    state_dim: 7
+    action_horizon: 30
+    repeated_diffusion_steps: 2
+    num_inference_timesteps: 4
+    add_pos_embed: true
+    max_seq_len: 1024
+    num_target_vision_tokens: 32
+    noise_beta_alpha: 1.5
+    noise_beta_beta: 1.0
+    noise_s: 0.999
+    num_timestep_buckets: 1000
+    diffusion_model_cfg:
+      action_dit_hidden_dim: 1024
+      dropout: 0.2
+      final_dropout: true
+      interleave_self_attention: true
+      norm_type: ada_norm
+      positional_embeddings: null
+      attention_head_dim: 64
+      num_layers: 28
+      input_embedding_dim: 1024
+      cross_attention_dim: 1024
+      num_attention_heads: 16
+    future_action_window_size: 29
+    past_action_window_size: 0
+datasets:
+  vla_data:
+    dataset_py: lerobot_datasets
+    data_root_dir: ./data
+    data_mix: stacking_cups_3cam_30hz_delta_joint_raw_h30
+    include_state: true
+    drop_incomplete_tail_windows: 30
+    episode_split: train
+    action_type: delta_qpos
+    action_mode: abs
+    sequential_step_sampling: false
+    per_device_batch_size: 16
+    load_all_data_for_training: true
+    obs_image_size:
+    - 224
+    - 224
+    delete_pause_frame: false
+    video_backend: torchvision_av
+    num_workers: 2
+    persistent_workers: true
+    prefetch_factor: 2
+    pin_memory: true
+trainer:
+  pretrained_checkpoint: null
+  freeze_modules: qwen_vl_interface
+  max_train_steps: 30000
+  num_warmup_steps: 500
+  save_interval: 10000
+  eval_interval: 1000
+  repeated_diffusion_steps: 2
+  learning_rate:
+    base: 1.0e-05
+    qwen_vl_interface: 1.0e-06
+    action_model: 0.0001
+    project_layers: 0.0001
+  lr_scheduler_type: cosine_with_min_lr
+  scheduler_specific_kwargs:
+    min_lr: 1.0e-06
+  loss_scale:
+    vla: 1.0
+    vlm: 0.0
+  max_grad_norm: 1.0
+  weight_decay: 0.0
+  logging_frequency: 20
+  gradient_clipping: 1.0
+  gradient_accumulation_steps: 1
+  gradient_checkpointing: true
+  optimizer:
+    name: AdamW
+    betas:
+    - 0.9
+    - 0.95
+    eps: 1.0e-08
+    weight_decay: 1.0e-08
+  save_format: pt
+  is_resume: false
+  max_wall_time_hours: 48
+  wall_time_safety_minutes: 20
+  save_training_state: true
+config_yaml: /scratch/wangpc/DiscreteRTCv2/examples/realRobots/Piper/train_files/starvla_qwenpi_qwen3vl2b_stacking_cups_3cam_30hz_delta_joint_raw_h30_2gpu_gb32_30k.yaml
+output_dir: /scratch/wangpc/DiscreteRTCv2/results/Checkpoints/stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k

StarVLA/stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k/config.yaml

@@ -0,0 +1,84 @@
+datasets:
+  vla_data:
+    action_mode: abs
+    data_mix: stacking_cups_3cam_30hz_delta_joint_raw_h30
+    data_root_dir: ./data
+    dataset_py: lerobot_datasets
+    delete_pause_frame: false
+    drop_incomplete_tail_windows: 30
+    episode_split: train
+    num_workers: 2
+    obs_image_size:
+    - 224
+    - 224
+    per_device_batch_size: 16
+    persistent_workers: true
+    pin_memory: true
+    prefetch_factor: 2
+    video_backend: torchvision_av
+framework:
+  action_model:
+    action_dim: 7
+    action_horizon: 30
+    add_pos_embed: true
+    diffusion_model_cfg:
+      action_dit_hidden_dim: 1024
+      attention_head_dim: 64
+      cross_attention_dim: 1024
+      dropout: 0.2
+      final_dropout: true
+      input_embedding_dim: 1024
+      interleave_self_attention: true
+      norm_type: ada_norm
+      num_attention_heads: 16
+      num_layers: 28
+      positional_embeddings: null
+    max_seq_len: 1024
+    noise_beta_alpha: 1.5
+    noise_beta_beta: 1.0
+    noise_s: 0.999
+    num_inference_timesteps: 4
+    num_target_vision_tokens: 32
+    num_timestep_buckets: 1000
+    state_dim: 7
+  name: QwenPI_v3
+  qwenvl:
+    attn_implementation: flash_attention_2
+    base_vlm: ./playground/Pretrained_models/Qwen3-VL-2B-Instruct-Action
+    num_vl_layers: 28
+    vl_hidden_dim: 2048
+output_dir: /scratch/wangpc/DiscreteRTCv2/results/Checkpoints/stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k
+run_id: stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k
+run_root_dir: /scratch/wangpc/DiscreteRTCv2/results/Checkpoints
+seed: 42
+trainer:
+  eval_interval: 1000
+  freeze_modules: qwen_vl_interface
+  gradient_clipping: 1.0
+  is_resume: false
+  learning_rate:
+    action_model: 0.0001
+    base: 1.0e-05
+    project_layers: 0.0001
+    qwen_vl_interface: 1.0e-06
+  logging_frequency: 20
+  lr_scheduler_type: cosine_with_min_lr
+  max_train_steps: 30000
+  max_wall_time_hours: 48
+  num_warmup_steps: 500
+  optimizer:
+    betas:
+    - 0.9
+    - 0.95
+    eps: 1.0e-08
+    weight_decay: 1.0e-08
+  pretrained_checkpoint: null
+  repeated_diffusion_steps: 2
+  save_format: pt
+  save_interval: 10000
+  save_training_state: true
+  scheduler_specific_kwargs:
+    min_lr: 1.0e-06
+  wall_time_safety_minutes: 20
+wandb_entity: 401910710-university-of-california-berkeley
+wandb_project: starVLA_piper

StarVLA/stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k/dataset_statistics.json

@@ -0,0 +1,127 @@
+{
+  "new_embodiment": {
+    "action": {
+      "mean": [
+        8.702239028934855e-06,
+        0.0002996787370648235,
+        -0.00042662431951612234,
+        0.00030915538081899285,
+        0.0004341287421993911,
+        -0.00037623976822942495,
+        0.145806685090065
+      ],
+      "std": [
+        0.00976935401558876,
+        0.016992630437016487,
+        0.012675837613642216,
+        0.003954065497964621,
+        0.009881439618766308,
+        0.004870147909969091,
+        0.17195862531661987
+      ],
+      "max": [
+        0.05808454751968384,
+        0.07365289330482483,
+        0.062063902616500854,
+        0.027087509632110596,
+        0.051103293895721436,
+        0.025307267904281616,
+        0.49900001287460327
+      ],
+      "min": [
+        -0.059917151927948,
+        -0.06379181146621704,
+        -0.06679373979568481,
+        -0.02801254391670227,
+        -0.06436777114868164,
+        -0.038397252559661865,
+        0.004000000189989805
+      ],
+      "q01": [
+        -0.033866183161735536,
+        -0.042907015681266786,
+        -0.036669373512268066,
+        -0.011501724943518638,
+        -0.033377493768930434,
+        -0.016650449931621552,
+        0.004000000189989805
+      ],
+      "q99": [
+        0.028797924518585205,
+        0.04272570550441742,
+        0.03734621644020079,
+        0.013801712207496146,
+        0.02991112142801283,
+        0.014154627919197083,
+        0.4059999883174896
+      ],
+      "mask": [
+        true,
+        true,
+        true,
+        true,
+        true,
+        true,
+        true
+      ]
+    },
+    "state": {
+      "mean": [
+        -0.07430676370859146,
+        1.506004810333252,
+        -1.3193917274475098,
+        0.13396592438220978,
+        0.9428086876869202,
+        -0.21519799530506134,
+        0.14584220945835114
+      ],
+      "std": [
+        0.26811927556991577,
+        0.6567522883415222,
+        0.5180357098579407,
+        0.12716346979141235,
+        0.3500312268733978,
+        0.2253401130437851,
+        0.17195044457912445
+      ],
+      "max": [
+        0.36791539192199707,
+        2.365933418273926,
+        -0.0007853981805965304,
+        0.44223153591156006,
+        1.2274551391601562,
+        0.27096235752105713,
+        0.49900001287460327
+      ],
+      "min": [
+        -0.7844207882881165,
+        0.06082472577691078,
+        -2.0039520263671875,
+        -0.2750464379787445,
+        -0.05572836473584175,
+        -0.9933716058731079,
+        0.004000000189989805
+      ],
+      "q01": [
+        -0.6364790016412735,
+        0.06386160105466843,
+        -1.9203597331047058,
+        -0.16183670088648797,
+        -0.027872908860445023,
+        -0.808698296546936,
+        0.004000000189989805
+      ],
+      "q99": [
+        0.32792991399765015,
+        2.2828032970428467,
+        -0.0032463124953210354,
+        0.39530363619327485,
+        1.224173903465271,
+        0.2039417177438736,
+        0.4059999883174896
+      ]
+    },
+    "num_transitions": 27502,
+    "num_trajectories": 57
+  }
+}

StarVLA/stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k/summary.jsonl

@@ -0,0 +1,3 @@
+{"steps": 10000}
+{"steps": 20000}
+{"steps": 30000}

StarVLA/stacking_cups_3cam_30hz_qwenpi_qwen3vl2b_delta_joint_raw_h30_2gpu_gb32_30k/wall_time_ledger.json

@@ -0,0 +1,15 @@
+{
+  "budget_hours": 48.0,
+  "safety_minutes": 20.0,
+  "sessions": [
+    {
+      "started_at": "2026-11-05T06:50:10+00:00",
+      "elapsed_seconds": 44210.9,
+      "start_step": 0,
+      "last_step": 30000,
+      "end_reason": "completed",
+      "signal": null
+    }
+  ],
+  "total_elapsed_seconds": 44210.9
+}