Upload policy weights, train config and readme

README.md

@@ -0,0 +1,62 @@
+---
+datasets: fecasado/Ncubes-to-Nbaskets-320x240
+library_name: lerobot
+license: apache-2.0
+model_name: gaze_flow_matching
+pipeline_tag: robotics
+tags:
+- robotics
+- gaze_flow_matching
+- lerobot
+---
+
+# Model Card for gaze_flow_matching
+
+<!-- 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

@@ -0,0 +1,120 @@
+{
+    "type": "gaze_flow_matching",
+    "n_obs_steps": 1,
+    "input_features": {
+        "observation.images.right": {
+            "type": "VISUAL",
+            "shape": [
+                240,
+                320,
+                3
+            ]
+        },
+        "observation.images.left": {
+            "type": "VISUAL",
+            "shape": [
+                240,
+                320,
+                3
+            ]
+        },
+        "observation.images.user": {
+            "type": "VISUAL",
+            "shape": [
+                240,
+                320,
+                3
+            ]
+        }
+    },
+    "output_features": {
+        "action": {
+            "type": "ACTION",
+            "shape": [
+                26
+            ]
+        }
+    },
+    "device": "cuda",
+    "use_amp": false,
+    "use_peft": false,
+    "push_to_hub": true,
+    "repo_id": "fecasado/gfm-cubes-baseline-vit",
+    "private": null,
+    "tags": null,
+    "license": null,
+    "pretrained_path": null,
+    "normalization_mapping": {
+        "VISUAL": "MEAN_STD",
+        "STATE": "MEAN_STD",
+        "ACTION": "MEAN_STD"
+    },
+    "gaze_feature_dim": 3,
+    "gaze_state_indices": [
+        -3,
+        -2,
+        -1
+    ],
+    "gaze_temporal_window": 1,
+    "vision_backbone": "resnet18",
+    "vit_token_size": 16,
+    "vit_foveation_strides": [
+        1,
+        2,
+        6
+    ],
+    "vit_foveation_grid_sizes": [
+        2,
+        3,
+        3
+    ],
+    "vit_model_size": "base",
+    "vit_pretrained_repo_id": "iantc104/mae_vitb_vit",
+    "vit_freeze": true,
+    "vit_backbone": "vit",
+    "gaze_cross_camera": false,
+    "gaze_cross_camera_method_train": "roma",
+    "gaze_cross_camera_method_inference": "roma",
+    "gaze_cross_camera_match_every": 5,
+    "gaze_cross_camera_timeout": 0.05,
+    "gaze_cross_camera_cache": null,
+    "obs_step_size": 1,
+    "n_action_steps": 2,
+    "horizon": 16,
+    "n_sampling_steps": 8,
+    "use_temporal_ensemble": true,
+    "temporal_ensemble_coeff": -0.01,
+    "dim_model": 512,
+    "n_heads": 8,
+    "mlp_ratio": 4.0,
+    "dropout": 0.1,
+    "dit_n_layers": 8,
+    "dit_time_dim": 256,
+    "pool_n_queries": 16,
+    "pool_n_layers": 4,
+    "use_gaze": false,
+    "use_gaze_as_action": false,
+    "gaze_model_repo_id": "",
+    "gaze_noise": 0.01,
+    "image_norm_mode": "mean_std",
+    "state_norm_mode": "min_max",
+    "action_norm_mode": "min_max",
+    "image_to_gaze_key": {},
+    "optimizer_lr": 0.0001,
+    "optimizer_lr_backbone": 0.0001,
+    "optimizer_betas": [
+        0.95,
+        0.999
+    ],
+    "optimizer_eps": 1e-08,
+    "optimizer_weight_decay": 1e-06,
+    "scheduler_name": "cosine",
+    "scheduler_warmup_steps": 500,
+    "use_ema": true,
+    "ema_decay": 0.99,
+    "input_shape": [
+        240,
+        320
+    ],
+    "visualize": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e41a2049c33d8c9f932085758cb8f93396878cb380f363cbbca4633730f132d
+size 601631752

train_config.json

@@ -0,0 +1,252 @@
+{
+    "dataset": {
+        "repo_id": "fecasado/Ncubes-to-Nbaskets-320x240",
+        "root": null,
+        "episodes": null,
+        "image_transforms": {
+            "enable": true,
+            "max_num_transforms": 1,
+            "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": "torchcodec",
+        "return_uint8": false,
+        "streaming": false
+    },
+    "env": null,
+    "policy": {
+        "type": "gaze_flow_matching",
+        "n_obs_steps": 1,
+        "input_features": {
+            "observation.images.right": {
+                "type": "VISUAL",
+                "shape": [
+                    240,
+                    320,
+                    3
+                ]
+            },
+            "observation.images.left": {
+                "type": "VISUAL",
+                "shape": [
+                    240,
+                    320,
+                    3
+                ]
+            },
+            "observation.images.user": {
+                "type": "VISUAL",
+                "shape": [
+                    240,
+                    320,
+                    3
+                ]
+            }
+        },
+        "output_features": {
+            "action": {
+                "type": "ACTION",
+                "shape": [
+                    26
+                ]
+            }
+        },
+        "device": "cuda",
+        "use_amp": false,
+        "use_peft": false,
+        "push_to_hub": true,
+        "repo_id": "fecasado/gfm-cubes-baseline-vit",
+        "private": null,
+        "tags": null,
+        "license": null,
+        "pretrained_path": null,
+        "normalization_mapping": {
+            "VISUAL": "MEAN_STD",
+            "STATE": "MEAN_STD",
+            "ACTION": "MEAN_STD"
+        },
+        "gaze_feature_dim": 3,
+        "gaze_state_indices": [
+            -3,
+            -2,
+            -1
+        ],
+        "gaze_temporal_window": 1,
+        "vision_backbone": "resnet18",
+        "vit_token_size": 16,
+        "vit_foveation_strides": [
+            1,
+            2,
+            6
+        ],
+        "vit_foveation_grid_sizes": [
+            2,
+            3,
+            3
+        ],
+        "vit_model_size": "base",
+        "vit_pretrained_repo_id": "iantc104/mae_vitb_vit",
+        "vit_freeze": true,
+        "vit_backbone": "vit",
+        "gaze_cross_camera": false,
+        "gaze_cross_camera_method_train": "roma",
+        "gaze_cross_camera_method_inference": "roma",
+        "gaze_cross_camera_match_every": 5,
+        "gaze_cross_camera_timeout": 0.05,
+        "gaze_cross_camera_cache": null,
+        "obs_step_size": 1,
+        "n_action_steps": 2,
+        "horizon": 16,
+        "n_sampling_steps": 8,
+        "use_temporal_ensemble": true,
+        "temporal_ensemble_coeff": -0.01,
+        "dim_model": 512,
+        "n_heads": 8,
+        "mlp_ratio": 4.0,
+        "dropout": 0.1,
+        "dit_n_layers": 8,
+        "dit_time_dim": 256,
+        "pool_n_queries": 16,
+        "pool_n_layers": 4,
+        "use_gaze": false,
+        "use_gaze_as_action": false,
+        "gaze_model_repo_id": "",
+        "gaze_noise": 0.01,
+        "image_norm_mode": "mean_std",
+        "state_norm_mode": "min_max",
+        "action_norm_mode": "min_max",
+        "image_to_gaze_key": {},
+        "optimizer_lr": 0.0001,
+        "optimizer_lr_backbone": 0.0001,
+        "optimizer_betas": [
+            0.95,
+            0.999
+        ],
+        "optimizer_eps": 1e-08,
+        "optimizer_weight_decay": 1e-06,
+        "scheduler_name": "cosine",
+        "scheduler_warmup_steps": 500,
+        "use_ema": true,
+        "ema_decay": 0.99,
+        "input_shape": [
+            240,
+            320
+        ],
+        "visualize": true
+    },
+    "reward_model": null,
+    "output_dir": "outputs/train/gfm-cubes-baseline-vit",
+    "job_name": "gfm-cubes-baseline-vit",
+    "resume": false,
+    "seed": 1000,
+    "cudnn_deterministic": false,
+    "num_workers": 15,
+    "batch_size": 128,
+    "prefetch_factor": 4,
+    "persistent_workers": true,
+    "steps": 100000,
+    "eval_freq": 20000,
+    "log_freq": 200,
+    "tolerance_s": 0.0001,
+    "save_checkpoint": true,
+    "save_freq": 10000,
+    "use_policy_training_preset": true,
+    "optimizer": {
+        "type": "adamw",
+        "lr": 0.0001,
+        "weight_decay": 1e-06,
+        "grad_clip_norm": 10.0,
+        "betas": [
+            0.95,
+            0.999
+        ],
+        "eps": 1e-08
+    },
+    "scheduler": null,
+    "eval": {
+        "n_episodes": 50,
+        "batch_size": 11,
+        "use_async_envs": true
+    },
+    "wandb": {
+        "enable": true,
+        "disable_artifact": true,
+        "project": "lerobot",
+        "entity": null,
+        "notes": null,
+        "run_id": "ygwzfrro",
+        "mode": null,
+        "add_tags": true
+    },
+    "peft": null,
+    "sample_weighting": null,
+    "rename_map": {},
+    "checkpoint_path": null
+}