the adapter trained on sim and hardware together

README.md

@@ -0,0 +1,33 @@
+---
+license: apache-2.0
+base_model: lerobot/pi05_base
+pipeline_tag: robotics
+tags:
+- LeRobot
+- so101
+- pi0.5
+- lora
+---
+
+# pi05_so101_block_stack
+
+pi0.5 fine-tuned to stack one 30 mm block on another with an [SO-101](https://github.com/TheRobotStudio/SO-ARM100), in simulation and on the physical arm. Trained by `train/pi05_so101_mix_aug.yaml` in [vla-test](https://github.com/danwahl/vla-test) on [drwahl/so101_block_stack](https://huggingface.co/drwahl/so101_block_stack), with each frame jittered for brightness, contrast, saturation and hue over ranges the size of what separates a rendered frame from one off the arm.
+
+LoRA on `lerobot/pi05_base`'s gemma_2b language backbone, with the action expert and the projection heads dense and SigLIP frozen. `adapter/` holds the adapter and the normalization processors; `train/merge.py` folds it into the base weights for a checkpoint that loads without one.
+
+```bash
+hf download drwahl/pi05_so101_block_stack --local-dir pi05_so101_block_stack
+uv run python train/merge.py pi05_so101_block_stack/adapter pi05_so101_block_stack/merged
+
+# in simulation
+uv run python sim/scripts/eval.py pi05_so101_block_stack/merged
+
+# on the arm
+uv run python -m hw.rollout pi05_so101_block_stack/merged --indices 0 1 2
+```
+
+Chunks are stitched with [Real-Time Chunking](https://www.physicalintelligence.company/research/real_time_chunking) and read back off a shape-preserving cubic before they are executed, in both programs.
+
+It stacks 139 of the 150 held-out sim layouts, averaged over three seeds. On the arm the blocks are placed against a console that blends the planned layout with the live camera, and whether an episode stacked is the operator's call, so there is no hardware number here.
+
+The prompt names two of red, green and blue: `stack the red block on the green block`.

adapter/adapter_config.json

@@ -0,0 +1,51 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "PI05Policy",
+    "parent_library": "lerobot.policies.pi05.modeling_pi05"
+  },
+  "base_model_name_or_path": "lerobot/pi05_base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [
+    "gemma_expert.model",
+    "action_in_proj",
+    "action_out_proj",
+    "time_mlp_in",
+    "time_mlp_out"
+  ],
+  "monteclora_config": null,
+  "peft_type": "LORA",
+  "peft_version": "0.20.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": ".*\\.paligemma\\.model\\.language_model\\.layers\\.\\d+\\.(self_attn\\.[qkvo]_proj|mlp\\.(gate|up|down)_proj)",
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false,
+  "velora_config": null
+}

adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7f2866e49f476f86fe44119046243103761809b9276424b3697645824d8ee3e
+size 1254516032

adapter/config.json

@@ -0,0 +1,108 @@
+{
+    "type": "pi05",
+    "n_obs_steps": 1,
+    "input_features": {
+        "observation.images.base_0_rgb": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                224,
+                224
+            ]
+        },
+        "observation.images.left_wrist_0_rgb": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                224,
+                224
+            ]
+        },
+        "observation.images.right_wrist_0_rgb": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                224,
+                224
+            ]
+        },
+        "observation.state": {
+            "type": "STATE",
+            "shape": [
+                32
+            ]
+        }
+    },
+    "output_features": {
+        "action": {
+            "type": "ACTION",
+            "shape": [
+                6
+            ]
+        }
+    },
+    "device": "cuda",
+    "use_amp": false,
+    "use_peft": true,
+    "push_to_hub": false,
+    "repo_id": null,
+    "private": null,
+    "tags": null,
+    "license": null,
+    "pretrained_path": "lerobot/pi05_base",
+    "pretrained_revision": null,
+    "paligemma_variant": "gemma_2b",
+    "action_expert_variant": "gemma_300m",
+    "dtype": "bfloat16",
+    "chunk_size": 50,
+    "n_action_steps": 50,
+    "max_state_dim": 32,
+    "max_action_dim": 32,
+    "num_inference_steps": 10,
+    "time_sampling_beta_alpha": 1.5,
+    "time_sampling_beta_beta": 1.0,
+    "time_sampling_scale": 0.999,
+    "time_sampling_offset": 0.001,
+    "min_period": 0.004,
+    "max_period": 4.0,
+    "use_relative_actions": false,
+    "relative_exclude_joints": [
+        "gripper"
+    ],
+    "action_feature_names": [
+        "shoulder_pan",
+        "shoulder_lift",
+        "elbow_flex",
+        "wrist_flex",
+        "wrist_roll",
+        "gripper"
+    ],
+    "rtc_config": null,
+    "image_resolution": [
+        224,
+        224
+    ],
+    "empty_cameras": 0,
+    "tokenizer_max_length": 200,
+    "normalization_mapping": {
+        "VISUAL": "IDENTITY",
+        "STATE": "QUANTILES",
+        "ACTION": "QUANTILES"
+    },
+    "gradient_checkpointing": true,
+    "compile_model": false,
+    "compile_mode": "max-autotune",
+    "freeze_vision_encoder": false,
+    "train_expert_only": false,
+    "optimizer_lr": 2.5e-05,
+    "optimizer_betas": [
+        0.9,
+        0.95
+    ],
+    "optimizer_eps": 1e-08,
+    "optimizer_weight_decay": 0.01,
+    "optimizer_grad_clip_norm": 1.0,
+    "scheduler_warmup_steps": 1000,
+    "scheduler_decay_steps": 20000,
+    "scheduler_decay_lr": 2.5e-06
+}

adapter/policy_postprocessor.json

@@ -0,0 +1,38 @@
+{
+  "name": "policy_postprocessor",
+  "steps": [
+    {
+      "registry_name": "unnormalizer_processor",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              6
+            ]
+          }
+        },
+        "norm_map": {
+          "VISUAL": "IDENTITY",
+          "STATE": "QUANTILES",
+          "ACTION": "QUANTILES"
+        }
+      },
+      "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors"
+    },
+    {
+      "registry_name": "absolute_actions_processor",
+      "config": {
+        "enabled": false
+      }
+    },
+    {
+      "registry_name": "device_processor",
+      "config": {
+        "device": "cpu",
+        "float_dtype": null
+      }
+    }
+  ]
+}

adapter/policy_postprocessor_step_0_unnormalizer_processor.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9656e2af277330c51a3d8bdcec8d48746ea6069318e4bc13757b92ad73327a7
+size 7536

adapter/policy_preprocessor.json

@@ -0,0 +1,100 @@
+{
+  "name": "policy_preprocessor",
+  "steps": [
+    {
+      "registry_name": "rename_observations_processor",
+      "config": {
+        "rename_map": {
+          "observation.images.top": "observation.images.base_0_rgb",
+          "observation.images.wrist": "observation.images.left_wrist_0_rgb"
+        }
+      }
+    },
+    {
+      "registry_name": "to_batch_processor",
+      "config": {}
+    },
+    {
+      "registry_name": "relative_actions_processor",
+      "config": {
+        "enabled": false,
+        "exclude_joints": [
+          "gripper"
+        ],
+        "action_names": null
+      }
+    },
+    {
+      "registry_name": "normalizer_processor",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "observation.images.base_0_rgb": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              224,
+              224
+            ]
+          },
+          "observation.images.left_wrist_0_rgb": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              224,
+              224
+            ]
+          },
+          "observation.images.right_wrist_0_rgb": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              224,
+              224
+            ]
+          },
+          "observation.state": {
+            "type": "STATE",
+            "shape": [
+              32
+            ]
+          },
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              6
+            ]
+          }
+        },
+        "norm_map": {
+          "VISUAL": "IDENTITY",
+          "STATE": "QUANTILES",
+          "ACTION": "QUANTILES"
+        }
+      },
+      "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
+    },
+    {
+      "registry_name": "pi05_prepare_state_tokenizer_processor_step",
+      "config": {}
+    },
+    {
+      "registry_name": "tokenizer_processor",
+      "config": {
+        "max_length": 200,
+        "task_key": "task",
+        "padding_side": "right",
+        "padding": "max_length",
+        "truncation": true,
+        "tokenizer_name": "google/paligemma-3b-pt-224"
+      }
+    },
+    {
+      "registry_name": "device_processor",
+      "config": {
+        "device": "cuda",
+        "float_dtype": null
+      }
+    }
+  ]
+}

adapter/policy_preprocessor_step_3_normalizer_processor.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9656e2af277330c51a3d8bdcec8d48746ea6069318e4bc13757b92ad73327a7
+size 7536

adapter/train_config.json

@@ -0,0 +1,253 @@
+{
+    "dataset": {
+        "repo_id": "vla-test/so101_block_stack_mix_v3",
+        "root": "/data/datasets/so101_block_stack_mix_v3",
+        "episodes": null,
+        "image_transforms": {
+            "enable": true,
+            "max_num_transforms": 3,
+            "random_order": false,
+            "tfs": {
+                "brightness": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "brightness": [
+                            0.85,
+                            1.2
+                        ]
+                    }
+                },
+                "contrast": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "contrast": [
+                            0.6,
+                            1.7
+                        ]
+                    }
+                },
+                "hue": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "hue": [
+                            -0.035,
+                            0.035
+                        ]
+                    }
+                },
+                "saturation": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "saturation": [
+                            0.5,
+                            2.0
+                        ]
+                    }
+                }
+            }
+        },
+        "revision": null,
+        "use_imagenet_stats": true,
+        "video_backend": "torchcodec",
+        "return_uint8": false,
+        "depth_output_unit": "mm",
+        "streaming": false,
+        "eval_split": 0.0
+    },
+    "env": null,
+    "policy": {
+        "type": "pi05",
+        "n_obs_steps": 1,
+        "input_features": {
+            "observation.images.base_0_rgb": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    224,
+                    224
+                ]
+            },
+            "observation.images.left_wrist_0_rgb": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    224,
+                    224
+                ]
+            },
+            "observation.images.right_wrist_0_rgb": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    224,
+                    224
+                ]
+            },
+            "observation.state": {
+                "type": "STATE",
+                "shape": [
+                    32
+                ]
+            }
+        },
+        "output_features": {
+            "action": {
+                "type": "ACTION",
+                "shape": [
+                    6
+                ]
+            }
+        },
+        "device": "cuda",
+        "use_amp": false,
+        "use_peft": true,
+        "push_to_hub": false,
+        "repo_id": null,
+        "private": null,
+        "tags": null,
+        "license": null,
+        "pretrained_path": "lerobot/pi05_base",
+        "pretrained_revision": null,
+        "paligemma_variant": "gemma_2b",
+        "action_expert_variant": "gemma_300m",
+        "dtype": "bfloat16",
+        "chunk_size": 50,
+        "n_action_steps": 50,
+        "max_state_dim": 32,
+        "max_action_dim": 32,
+        "num_inference_steps": 10,
+        "time_sampling_beta_alpha": 1.5,
+        "time_sampling_beta_beta": 1.0,
+        "time_sampling_scale": 0.999,
+        "time_sampling_offset": 0.001,
+        "min_period": 0.004,
+        "max_period": 4.0,
+        "use_relative_actions": false,
+        "relative_exclude_joints": [
+            "gripper"
+        ],
+        "action_feature_names": [
+            "shoulder_pan",
+            "shoulder_lift",
+            "elbow_flex",
+            "wrist_flex",
+            "wrist_roll",
+            "gripper"
+        ],
+        "rtc_config": null,
+        "image_resolution": [
+            224,
+            224
+        ],
+        "empty_cameras": 0,
+        "tokenizer_max_length": 200,
+        "normalization_mapping": {
+            "VISUAL": "IDENTITY",
+            "STATE": "QUANTILES",
+            "ACTION": "QUANTILES"
+        },
+        "gradient_checkpointing": true,
+        "compile_model": false,
+        "compile_mode": "max-autotune",
+        "freeze_vision_encoder": false,
+        "train_expert_only": false,
+        "optimizer_lr": 2.5e-05,
+        "optimizer_betas": [
+            0.9,
+            0.95
+        ],
+        "optimizer_eps": 1e-08,
+        "optimizer_weight_decay": 0.01,
+        "optimizer_grad_clip_norm": 1.0,
+        "scheduler_warmup_steps": 1000,
+        "scheduler_decay_steps": 20000,
+        "scheduler_decay_lr": 2.5e-06
+    },
+    "reward_model": null,
+    "output_dir": "/data/checkpoints/pi05_so101_block_stack_mix_aug_v3",
+    "job_name": "pi05_so101_block_stack_mix_aug_v3",
+    "resume": false,
+    "seed": 1000,
+    "cudnn_deterministic": false,
+    "num_workers": 4,
+    "batch_size": 4,
+    "prefetch_factor": 4,
+    "persistent_workers": true,
+    "steps": 20000,
+    "env_eval_freq": 20000,
+    "log_freq": 200,
+    "eval_steps": 0,
+    "max_eval_samples": 0,
+    "tolerance_s": 0.0001,
+    "save_checkpoint": true,
+    "save_freq": 5000,
+    "use_policy_training_preset": true,
+    "optimizer": {
+        "type": "adamw",
+        "lr": 2.5e-05,
+        "weight_decay": 0.01,
+        "grad_clip_norm": 1.0,
+        "betas": [
+            0.9,
+            0.95
+        ],
+        "eps": 1e-08
+    },
+    "scheduler": {
+        "type": "cosine_decay_with_warmup",
+        "num_warmup_steps": 1000,
+        "num_decay_steps": 20000,
+        "peak_lr": 2.5e-05,
+        "decay_lr": 2.5e-06
+    },
+    "eval": {
+        "n_episodes": 50,
+        "batch_size": 8,
+        "use_async_envs": true,
+        "recording": false,
+        "recording_repo_id": null,
+        "recording_private": false
+    },
+    "wandb": {
+        "enable": true,
+        "disable_artifact": true,
+        "project": "vla-test",
+        "entity": null,
+        "notes": null,
+        "run_id": "pi05_so101_block_stack_mix_aug_v3",
+        "mode": null,
+        "add_tags": true
+    },
+    "peft": {
+        "target_modules": ".*\\.paligemma\\.model\\.language_model\\.layers\\.\\d+\\.(self_attn\\.[qkvo]_proj|mlp\\.(gate|up|down)_proj)",
+        "full_training_modules": [
+            "gemma_expert.model",
+            "action_in_proj",
+            "action_out_proj",
+            "time_mlp_in",
+            "time_mlp_out"
+        ],
+        "method_type": "LORA",
+        "init_type": null,
+        "r": 32,
+        "lora_alpha": 64
+    },
+    "job": {
+        "target": null,
+        "image": "huggingface/lerobot-gpu:latest",
+        "timeout": "2d",
+        "detach": false,
+        "tags": []
+    },
+    "save_checkpoint_to_hub": false,
+    "sample_weighting": null,
+    "rename_map": {
+        "observation.images.top": "observation.images.base_0_rgb",
+        "observation.images.wrist": "observation.images.left_wrist_0_rgb"
+    },
+    "checkpoint_path": null
+}