Upload policy weights, train config and readme

Browse files

Files changed (4) hide show

README.md +62 -0
config.json +123 -0
model.safetensors +3 -0
train_config.json +261 -0

README.md ADDED Viewed

	@@ -0,0 +1,62 @@

+---
+datasets: ETHrobotlearning/task3-TOY-clean
+library_name: lerobot
+license: apache-2.0
+model_name: molmoact2
+pipeline_tag: robotics
+tags:
+- robotics
+- molmoact2
+- lerobot
+---
+# Model Card for molmoact2
+<!-- Provide a quick summary of what the model is/does. -->
+_Model type not recognized — please update this template._
+This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).
+See the full documentation at [LeRobot Docs](https://huggingface.co/docs/lerobot/index).
+---
+## How to Get Started with the Model
+For a complete walkthrough, see the [training guide](https://huggingface.co/docs/lerobot/il_robots#train-a-policy).
+Below is the short version on how to train and run inference/eval:
+### Train from scratch
+```bash
+lerobot-train \
+  --dataset.repo_id=${HF_USER}/<dataset> \
+  --policy.type=act \
+  --output_dir=outputs/train/<desired_policy_repo_id> \
+  --job_name=lerobot_training \
+  --policy.device=cuda \
+  --policy.repo_id=${HF_USER}/<desired_policy_repo_id>
+  --wandb.enable=true
+```
+_Writes checkpoints to `outputs/train/<desired_policy_repo_id>/checkpoints/`._
+### Evaluate the policy/run inference
+```bash
+lerobot-record \
+  --robot.type=so100_follower \
+  --dataset.repo_id=<hf_user>/eval_<dataset> \
+  --policy.path=<hf_user>/<desired_policy_repo_id> \
+  --episodes=10
+```
+Prefix the dataset repo with **eval\_** and supply `--policy.path` pointing to a local or hub checkpoint.
+---
+## Model Details
+- **License:** apache-2.0

config.json ADDED Viewed

	@@ -0,0 +1,123 @@

+{
+    "type": "molmoact2",
+    "n_obs_steps": 1,
+    "input_features": {
+        "observation.state": {
+            "type": "STATE",
+            "shape": [
+                6
+            ]
+        },
+        "observation.images.front": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                480,
+                640
+            ]
+        }
+    },
+    "output_features": {
+        "action": {
+            "type": "ACTION",
+            "shape": [
+                6
+            ]
+        }
+    },
+    "device": "cuda",
+    "use_amp": false,
+    "use_peft": false,
+    "push_to_hub": true,
+    "repo_id": "ETHrobotlearning/molmoact2-task3-TOY-clean-step2000",
+    "private": null,
+    "tags": null,
+    "license": null,
+    "pretrained_path": null,
+    "checkpoint_path": "allenai/MolmoAct2-SO100_101",
+    "checkpoint_revision": null,
+    "checkpoint_force_download": false,
+    "trust_remote_code": true,
+    "chunk_size": 10,
+    "n_action_steps": 10,
+    "action_mode": "continuous",
+    "inference_action_mode": null,
+    "discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
+    "discrete_generation_max_steps": null,
+    "norm_tag": null,
+    "setup_type": "single SO-100/SO-101 arm with one front RGB camera",
+    "control_mode": "absolute joint pose",
+    "image_keys": [
+        "observation.images.front"
+    ],
+    "normalize_language": true,
+    "add_setup_tokens": true,
+    "add_control_tokens": true,
+    "normalize_gripper": false,
+    "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,
+    "enable_lora_vlm": true,
+    "lora_rank": 64,
+    "lora_alpha": 16,
+    "lora_dropout": 0.05,
+    "lora_bias": "none",
+    "enable_lora_action_expert": false,
+    "enable_knowledge_insulation": false,
+    "freeze_embedding": true,
+    "train_action_expert_only": false,
+    "gradient_checkpointing": true,
+    "model_dtype": "bfloat16",
+    "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": null,
+    "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 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:a80b5a6e073b1a80d83d0ad4aa8b27ce85b393883b51f58f53565f331a075ca8
+size 11483634784

train_config.json ADDED Viewed

	@@ -0,0 +1,261 @@

+{
+    "dataset": {
+        "repo_id": "ETHrobotlearning/task3-TOY-clean",
+        "root": null,
+        "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": "main",
+        "use_imagenet_stats": true,
+        "video_backend": "pyav",
+        "return_uint8": false,
+        "streaming": false
+    },
+    "env": null,
+    "policy": {
+        "type": "molmoact2",
+        "n_obs_steps": 1,
+        "input_features": {
+            "observation.state": {
+                "type": "STATE",
+                "shape": [
+                    6
+                ]
+            },
+            "observation.images.front": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    480,
+                    640
+                ]
+            }
+        },
+        "output_features": {
+            "action": {
+                "type": "ACTION",
+                "shape": [
+                    6
+                ]
+            }
+        },
+        "device": "cuda",
+        "use_amp": false,
+        "use_peft": false,
+        "push_to_hub": true,
+        "repo_id": "ETHrobotlearning/molmoact2-task3-TOY-clean-step2000",
+        "private": null,
+        "tags": null,
+        "license": null,
+        "pretrained_path": null,
+        "checkpoint_path": "allenai/MolmoAct2-SO100_101",
+        "checkpoint_revision": null,
+        "checkpoint_force_download": false,
+        "trust_remote_code": true,
+        "chunk_size": 10,
+        "n_action_steps": 10,
+        "action_mode": "continuous",
+        "inference_action_mode": null,
+        "discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
+        "discrete_generation_max_steps": null,
+        "norm_tag": null,
+        "setup_type": "single SO-100/SO-101 arm with one front RGB camera",
+        "control_mode": "absolute joint pose",
+        "image_keys": [
+            "observation.images.front"
+        ],
+        "normalize_language": true,
+        "add_setup_tokens": true,
+        "add_control_tokens": true,
+        "normalize_gripper": false,
+        "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,
+        "enable_lora_vlm": true,
+        "lora_rank": 64,
+        "lora_alpha": 16,
+        "lora_dropout": 0.05,
+        "lora_bias": "none",
+        "enable_lora_action_expert": false,
+        "enable_knowledge_insulation": false,
+        "freeze_embedding": true,
+        "train_action_expert_only": false,
+        "gradient_checkpointing": true,
+        "model_dtype": "bfloat16",
+        "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": null,
+        "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": "/home/ubuntu/outputs/train/molmoact2_task3_toy_clean",
+    "job_name": "molmoact2_task3_toy_clean",
+    "resume": false,
+    "seed": 1000,
+    "cudnn_deterministic": false,
+    "num_workers": 4,
+    "batch_size": 32,
+    "prefetch_factor": 4,
+    "persistent_workers": true,
+    "steps": 20000,
+    "eval_freq": -1,
+    "log_freq": 20,
+    "tolerance_s": 0.0001,
+    "save_checkpoint": true,
+    "save_freq": 500,
+    "use_policy_training_preset": true,
+    "optimizer": {
+        "type": "adamw",
+        "lr": 1e-05,
+        "weight_decay": 0.0,
+        "grad_clip_norm": 1.0,
+        "betas": [
+            0.9,
+            0.95
+        ],
+        "eps": 1e-06
+    },
+    "scheduler": {
+        "type": "molmoact2_cosine_decay_with_warmup",
+        "num_warmup_steps": 200,
+        "num_decay_steps": null,
+        "peak_lr": 1e-05,
+        "decay_lr": 1e-06
+    },
+    "eval": {
+        "n_episodes": 50,
+        "batch_size": 11,
+        "use_async_envs": true
+    },
+    "wandb": {
+        "enable": true,
+        "disable_artifact": false,
+        "project": "molmoact2",
+        "entity": null,
+        "notes": null,
+        "run_id": "ajjfvgux",
+        "mode": null,
+        "add_tags": true
+    },
+    "peft": null,
+    "sample_weighting": null,
+    "rename_map": {},
+    "checkpoint_path": null
+}