Upload folder using huggingface_hub

config.json

@@ -0,0 +1,135 @@
+{
+    "type": "molmoact2",
+    "n_obs_steps": 1,
+    "input_features": {
+        "observation.state": {
+            "type": "STATE",
+            "shape": [
+                6
+            ]
+        },
+        "observation.images.front": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                720,
+                1280
+            ]
+        },
+        "observation.images.right": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                480,
+                640
+            ]
+        }
+    },
+    "output_features": {
+        "action": {
+            "type": "ACTION",
+            "shape": [
+                6
+            ]
+        }
+    },
+    "device": "cuda",
+    "use_amp": false,
+    "use_peft": false,
+    "push_to_hub": false,
+    "repo_id": null,
+    "private": null,
+    "tags": null,
+    "license": null,
+    "pretrained_path": null,
+    "pretrained_revision": null,
+    "checkpoint_path": "allenai/MolmoAct2-SO100_101",
+    "checkpoint_revision": null,
+    "checkpoint_force_download": false,
+    "chunk_size": 10,
+    "n_action_steps": 10,
+    "action_mode": "continuous",
+    "inference_action_mode": "continuous",
+    "discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
+    "discrete_generation_max_steps": null,
+    "norm_tag": null,
+    "norm_stats_path": null,
+    "setup_type": "single so101 robotic arm",
+    "control_mode": "absolute joint pose",
+    "image_keys": [
+        "observation.images.front",
+        "observation.images.right"
+    ],
+    "normalize_language": true,
+    "add_setup_tokens": true,
+    "add_control_tokens": true,
+    "normalize_gripper": true,
+    "num_state_tokens": 256,
+    "max_sequence_length": null,
+    "expected_max_action_dim": 32,
+    "num_flow_timesteps": 8,
+    "flow_matching_cutoff": 1.0,
+    "flow_matching_time_offset": 0.001,
+    "flow_matching_time_scale": 0.999,
+    "flow_matching_beta_alpha": 1.0,
+    "flow_matching_beta_beta": 1.5,
+    "num_inference_steps": null,
+    "mask_action_dim_padding": true,
+    "enable_inference_cuda_graph": true,
+    "per_episode_seed": false,
+    "eval_seed": null,
+    "rtc_config": null,
+    "joint_signs": null,
+    "joint_offsets": null,
+    "train_mode_vlm": "lora",
+    "lora_rank": 64,
+    "lora_alpha": 16,
+    "lora_dropout": 0.05,
+    "lora_bias": "none",
+    "enable_knowledge_insulation": false,
+    "freeze_embedding": true,
+    "gradient_checkpointing": true,
+    "compile_model": false,
+    "dtype": "bfloat16",
+    "llm_residual_dropout": 0.1,
+    "softmax_auxiliary_loss": true,
+    "softmax_auxiliary_loss_scale": 0.0001,
+    "discrete_loss_token_weighting": "root_subsegments_root_tokens",
+    "optimizer_lr": 1e-05,
+    "optimizer_vit_lr": 5e-06,
+    "optimizer_connector_lr": 5e-06,
+    "optimizer_action_expert_lr": 5e-05,
+    "optimizer_betas": [
+        0.9,
+        0.95
+    ],
+    "optimizer_eps": 1e-06,
+    "optimizer_weight_decay": 0.0,
+    "optimizer_grad_clip_norm": 1.0,
+    "scheduler_warmup_steps": 200,
+    "scheduler_decay_steps": 24000,
+    "scheduler_decay_lr": 1e-06,
+    "normalization_mapping": {
+        "VISUAL": "IDENTITY",
+        "STATE": "QUANTILES",
+        "ACTION": "QUANTILES"
+    },
+    "dataset_feature_names": {
+        "action": [
+            "shoulder_pan.pos",
+            "shoulder_lift.pos",
+            "elbow_flex.pos",
+            "wrist_flex.pos",
+            "wrist_roll.pos",
+            "gripper.pos"
+        ],
+        "observation.state": [
+            "shoulder_pan.pos",
+            "shoulder_lift.pos",
+            "elbow_flex.pos",
+            "wrist_flex.pos",
+            "wrist_roll.pos",
+            "gripper.pos"
+        ]
+    }
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2110468997ac5e8119a319bdc8cb6e16fa640a3f455338f696647d1c91d4eea6
+size 12679625488

policy_postprocessor.json

@@ -0,0 +1,43 @@
+{
+  "name": "policy_postprocessor",
+  "steps": [
+    {
+      "registry_name": "molmoact2_clamp_action",
+      "config": {}
+    },
+    {
+      "registry_name": "molmoact2_masked_unnormalizer",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              6
+            ]
+          }
+        },
+        "norm_map": {
+          "VISUAL": "IDENTITY",
+          "STATE": "QUANTILES",
+          "ACTION": "QUANTILES"
+        }
+      },
+      "state_file": "policy_postprocessor_step_1_molmoact2_masked_unnormalizer.safetensors"
+    },
+    {
+      "registry_name": "molmoact2_action_frame_transform",
+      "config": {
+        "joint_signs": null,
+        "joint_offsets": null
+      }
+    },
+    {
+      "registry_name": "device_processor",
+      "config": {
+        "device": "cpu",
+        "float_dtype": null
+      }
+    }
+  ]
+}

policy_postprocessor_step_1_molmoact2_masked_unnormalizer.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e13f791829a14cf9dd64125011bbfd2e0f4613c825d6020354868e77665d96d
+size 7744

policy_preprocessor.json

@@ -0,0 +1,119 @@
+{
+  "name": "policy_preprocessor",
+  "steps": [
+    {
+      "registry_name": "rename_observations_processor",
+      "config": {
+        "rename_map": {}
+      }
+    },
+    {
+      "registry_name": "to_batch_processor",
+      "config": {}
+    },
+    {
+      "registry_name": "molmoact2_state_frame_transform",
+      "config": {
+        "joint_signs": null,
+        "joint_offsets": null
+      }
+    },
+    {
+      "registry_name": "molmoact2_masked_normalizer",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "observation.state": {
+            "type": "STATE",
+            "shape": [
+              6
+            ]
+          },
+          "observation.images.front": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              720,
+              1280
+            ]
+          },
+          "observation.images.right": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              480,
+              640
+            ]
+          },
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              6
+            ]
+          }
+        },
+        "norm_map": {
+          "VISUAL": "IDENTITY",
+          "STATE": "QUANTILES",
+          "ACTION": "QUANTILES"
+        }
+      },
+      "state_file": "policy_preprocessor_step_3_molmoact2_masked_normalizer.safetensors"
+    },
+    {
+      "registry_name": "molmoact2_clamp_normalized",
+      "config": {
+        "normalization_masks": {
+          "action": [
+            true,
+            true,
+            true,
+            true,
+            true,
+            true
+          ],
+          "observation.state": [
+            true,
+            true,
+            true,
+            true,
+            true,
+            true
+          ]
+        }
+      }
+    },
+    {
+      "registry_name": "molmoact2_pack_inputs",
+      "config": {
+        "checkpoint_path": "allenai/MolmoAct2-SO100_101",
+        "checkpoint_revision": null,
+        "checkpoint_force_download": false,
+        "action_mode": "continuous",
+        "discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
+        "image_keys": [
+          "observation.images.front",
+          "observation.images.right"
+        ],
+        "allow_image_key_fallback": false,
+        "setup_type": "single so101 robotic arm",
+        "control_mode": "absolute joint pose",
+        "normalize_language": true,
+        "add_setup_tokens": true,
+        "add_control_tokens": true,
+        "num_state_tokens": 256,
+        "max_sequence_length": null,
+        "chunk_size": 10,
+        "max_action_dim": 32,
+        "env_action_dim": 6
+      }
+    },
+    {
+      "registry_name": "device_processor",
+      "config": {
+        "device": "cuda",
+        "float_dtype": null
+      }
+    }
+  ]
+}

policy_preprocessor_step_3_molmoact2_masked_normalizer.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3e6bba220d903473d12cc9788c7426a59d3f6e911c3ecad31e9663b228a7816
+size 7752

train_config.json

@@ -0,0 +1,339 @@
+{
+    "dataset": {
+        "repo_id": "vikrantsp/pickPlace_and_cubeStack",
+        "repo_type": "dataset",
+        "root": null,
+        "episodes": null,
+        "exclude_episodes": null,
+        "image_transforms": {
+            "enable": true,
+            "max_num_transforms": 3,
+            "random_order": false,
+            "tfs": {
+                "brightness": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "brightness": [
+                            0.8,
+                            1.2
+                        ]
+                    }
+                },
+                "contrast": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "contrast": [
+                            0.8,
+                            1.2
+                        ]
+                    }
+                },
+                "saturation": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "saturation": [
+                            0.5,
+                            1.5
+                        ]
+                    }
+                },
+                "hue": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "hue": [
+                            -0.05,
+                            0.05
+                        ]
+                    }
+                },
+                "sharpness": {
+                    "weight": 1.0,
+                    "type": "SharpnessJitter",
+                    "kwargs": {
+                        "sharpness": [
+                            0.5,
+                            1.5
+                        ]
+                    }
+                },
+                "affine": {
+                    "weight": 1.0,
+                    "type": "RandomAffine",
+                    "kwargs": {
+                        "degrees": [
+                            -5.0,
+                            5.0
+                        ],
+                        "translate": [
+                            0.05,
+                            0.05
+                        ]
+                    }
+                }
+            }
+        },
+        "revision": null,
+        "use_imagenet_stats": true,
+        "video_backend": "pyav",
+        "return_uint8": false,
+        "depth_output_unit": "mm",
+        "streaming": false,
+        "eval_split": 0.0
+    },
+    "env": null,
+    "policy": {
+        "type": "molmoact2",
+        "n_obs_steps": 1,
+        "input_features": {
+            "observation.state": {
+                "type": "STATE",
+                "shape": [
+                    6
+                ]
+            },
+            "observation.images.front": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    720,
+                    1280
+                ]
+            },
+            "observation.images.right": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    480,
+                    640
+                ]
+            }
+        },
+        "output_features": {
+            "action": {
+                "type": "ACTION",
+                "shape": [
+                    6
+                ]
+            }
+        },
+        "device": "cuda",
+        "use_amp": false,
+        "use_peft": false,
+        "push_to_hub": false,
+        "repo_id": null,
+        "private": null,
+        "tags": null,
+        "license": null,
+        "pretrained_path": null,
+        "pretrained_revision": null,
+        "checkpoint_path": "allenai/MolmoAct2-SO100_101",
+        "checkpoint_revision": null,
+        "checkpoint_force_download": false,
+        "chunk_size": 10,
+        "n_action_steps": 10,
+        "action_mode": "continuous",
+        "inference_action_mode": "continuous",
+        "discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
+        "discrete_generation_max_steps": null,
+        "norm_tag": null,
+        "norm_stats_path": null,
+        "setup_type": "single so101 robotic arm",
+        "control_mode": "absolute joint pose",
+        "image_keys": [
+            "observation.images.front",
+            "observation.images.right"
+        ],
+        "normalize_language": true,
+        "add_setup_tokens": true,
+        "add_control_tokens": true,
+        "normalize_gripper": true,
+        "num_state_tokens": 256,
+        "max_sequence_length": null,
+        "expected_max_action_dim": 32,
+        "num_flow_timesteps": 8,
+        "flow_matching_cutoff": 1.0,
+        "flow_matching_time_offset": 0.001,
+        "flow_matching_time_scale": 0.999,
+        "flow_matching_beta_alpha": 1.0,
+        "flow_matching_beta_beta": 1.5,
+        "num_inference_steps": null,
+        "mask_action_dim_padding": true,
+        "enable_inference_cuda_graph": true,
+        "per_episode_seed": false,
+        "eval_seed": null,
+        "rtc_config": null,
+        "joint_signs": null,
+        "joint_offsets": null,
+        "train_mode_vlm": "lora",
+        "lora_rank": 64,
+        "lora_alpha": 16,
+        "lora_dropout": 0.05,
+        "lora_bias": "none",
+        "enable_knowledge_insulation": false,
+        "freeze_embedding": true,
+        "gradient_checkpointing": true,
+        "compile_model": false,
+        "dtype": "bfloat16",
+        "llm_residual_dropout": 0.1,
+        "softmax_auxiliary_loss": true,
+        "softmax_auxiliary_loss_scale": 0.0001,
+        "discrete_loss_token_weighting": "root_subsegments_root_tokens",
+        "optimizer_lr": 1e-05,
+        "optimizer_vit_lr": 5e-06,
+        "optimizer_connector_lr": 5e-06,
+        "optimizer_action_expert_lr": 5e-05,
+        "optimizer_betas": [
+            0.9,
+            0.95
+        ],
+        "optimizer_eps": 1e-06,
+        "optimizer_weight_decay": 0.0,
+        "optimizer_grad_clip_norm": 1.0,
+        "scheduler_warmup_steps": 200,
+        "scheduler_decay_steps": 24000,
+        "scheduler_decay_lr": 1e-06,
+        "normalization_mapping": {
+            "VISUAL": "IDENTITY",
+            "STATE": "QUANTILES",
+            "ACTION": "QUANTILES"
+        },
+        "dataset_feature_names": {
+            "action": [
+                "shoulder_pan.pos",
+                "shoulder_lift.pos",
+                "elbow_flex.pos",
+                "wrist_flex.pos",
+                "wrist_roll.pos",
+                "gripper.pos"
+            ],
+            "observation.state": [
+                "shoulder_pan.pos",
+                "shoulder_lift.pos",
+                "elbow_flex.pos",
+                "wrist_flex.pos",
+                "wrist_roll.pos",
+                "gripper.pos"
+            ]
+        }
+    },
+    "reward_model": null,
+    "output_dir": "outputs/so101_stack_lora",
+    "job_name": "so101_cube_lora",
+    "resume": false,
+    "seed": 1000,
+    "cudnn_deterministic": false,
+    "num_workers": 4,
+    "batch_size": 4,
+    "prefetch_factor": 4,
+    "persistent_workers": true,
+    "dataloader_multiprocessing_context": "spawn",
+    "steps": 25000,
+    "env_eval_freq": -1,
+    "log_freq": 20,
+    "eval_steps": 0,
+    "max_eval_samples": 0,
+    "tolerance_s": 0.0001,
+    "save_checkpoint": true,
+    "save_freq": 2000,
+    "checkpoint_format": "safetensors",
+    "use_policy_training_preset": true,
+    "optimizer": {
+        "type": "molmoact2_adamw",
+        "lr": 1e-05,
+        "weight_decay": 0.0,
+        "grad_clip_norm": 0.0,
+        "betas": [
+            0.9,
+            0.95
+        ],
+        "eps": 1e-06,
+        "group_grad_clip_norm": 1.0
+    },
+    "scheduler": {
+        "type": "molmoact2_cosine_with_warmup",
+        "num_warmup_steps": 200,
+        "num_decay_steps": 24000,
+        "peak_lr": 1e-05,
+        "decay_lr": 1e-06
+    },
+    "parallelism": {
+        "dp_replicate": 1,
+        "dp_shard": 1,
+        "context_parallel": {
+            "ring_degree": 1,
+            "ulysses_degree": 1
+        },
+        "cfg_parallel": 1
+    },
+    "accelerator": {
+        "mixed_precision": "no",
+        "gradient_accumulation": {
+            "steps": 1
+        },
+        "fsdp": {
+            "reshard_after_forward": true,
+            "wrap_modules": null,
+            "min_num_params": null,
+            "cpu_offload": false,
+            "ignored_modules": null
+        },
+        "ddp": {
+            "find_unused_parameters": true,
+            "gradient_as_bucket_view": false,
+            "static_graph": false
+        },
+        "compile": {
+            "enabled": false,
+            "backend": "inductor",
+            "mode": null,
+            "regional": true
+        },
+        "activation_checkpointing": {
+            "mode": "none"
+        }
+    },
+    "eval": {
+        "n_episodes": 50,
+        "batch_size": 19,
+        "use_async_envs": true,
+        "recording": false,
+        "recording_repo_id": null,
+        "recording_private": false
+    },
+    "ema": {
+        "enable": false,
+        "decay": null,
+        "update_after_step": 0,
+        "inv_gamma": 1.0,
+        "power": 0.75,
+        "min_decay": 0.0,
+        "max_decay": 0.9999,
+        "use_for_eval": true
+    },
+    "wandb": {
+        "enable": false,
+        "disable_artifact": false,
+        "project": "lerobot",
+        "entity": null,
+        "notes": null,
+        "run_id": null,
+        "mode": null,
+        "add_tags": true
+    },
+    "peft": null,
+    "job": {
+        "target": null,
+        "image": "huggingface/lerobot-gpu:latest",
+        "timeout": "2d",
+        "detach": false,
+        "tags": []
+    },
+    "save_checkpoint_to_hub": false,
+    "sample_weighting": null,
+    "rename_map": {},
+    "checkpoint_path": null
+}