Upload folder using huggingface_hub

060000/pretrained_model/config.json

@@ -0,0 +1,116 @@
+{
+    "type": "multi_task_dit",
+    "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": "gaspardthrl/walleed_dit_fm_absolute",
+    "private": null,
+    "tags": null,
+    "license": null,
+    "pretrained_path": null,
+    "obs_stride": 1,
+    "state_history_motors": [],
+    "state_history_lags": [],
+    "state_history_scales": [],
+    "horizon": 32,
+    "n_action_steps": 24,
+    "objective": "flow_matching",
+    "noise_scheduler_type": "DDPM",
+    "num_train_timesteps": 100,
+    "beta_schedule": "squaredcos_cap_v2",
+    "beta_start": 0.0001,
+    "beta_end": 0.02,
+    "prediction_type": "epsilon",
+    "clip_sample": true,
+    "clip_sample_range": 1.0,
+    "num_inference_steps": null,
+    "sigma_min": 0.0,
+    "num_integration_steps": 10,
+    "integration_method": "euler",
+    "timestep_sampling_strategy": "beta",
+    "timestep_sampling_s": 0.999,
+    "timestep_sampling_alpha": 1.5,
+    "timestep_sampling_beta": 1.0,
+    "hidden_dim": 512,
+    "num_layers": 6,
+    "num_heads": 8,
+    "dropout": 0.1,
+    "use_positional_encoding": false,
+    "timestep_embed_dim": 256,
+    "use_rope": true,
+    "rope_base": 10000.0,
+    "vision_encoder_name": "openai/clip-vit-base-patch16",
+    "use_separate_rgb_encoder_per_camera": false,
+    "vision_encoder_lr_multiplier": 0.0,
+    "image_resize_shape": [
+        224,
+        224
+    ],
+    "image_crop_shape": [
+        224,
+        224
+    ],
+    "image_crop_is_random": true,
+    "image_grayscale": true,
+    "image_grayworld": true,
+    "use_relative_actions": false,
+    "relative_exclude_joints": [
+        "gripper"
+    ],
+    "action_feature_names": [
+        "shoulder_pan.pos",
+        "shoulder_lift.pos",
+        "elbow_flex.pos",
+        "wrist_flex.pos",
+        "wrist_roll.pos",
+        "gripper.pos"
+    ],
+    "text_encoder_name": "openai/clip-vit-base-patch16",
+    "tokenizer_max_length": 77,
+    "tokenizer_padding": "max_length",
+    "tokenizer_padding_side": "right",
+    "tokenizer_truncation": true,
+    "default_task": "Fold the towel",
+    "normalization_mapping": {
+        "VISUAL": "MEAN_STD",
+        "STATE": "MIN_MAX",
+        "ACTION": "MIN_MAX"
+    },
+    "optimizer_lr": 2e-05,
+    "optimizer_betas": [
+        0.95,
+        0.999
+    ],
+    "optimizer_eps": 1e-08,
+    "optimizer_weight_decay": 0.0,
+    "scheduler_name": "cosine",
+    "scheduler_warmup_steps": 1000,
+    "do_mask_loss_for_padding": false,
+    "drop_n_last_frames": 8
+}

060000/pretrained_model/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d1e16be46656b77322e68892ae679deb248654484917eb1cc7fc2eb2b3e403e
+size 787433656

060000/pretrained_model/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": "MEAN_STD",
+          "STATE": "MIN_MAX",
+          "ACTION": "MIN_MAX"
+        }
+      },
+      "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
+      }
+    }
+  ]
+}

060000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bba8954e4fa3a82ef9fdfddf7a4da6e6e4ca774c3890a8bcdeb91892a46e054
+size 6608

060000/pretrained_model/policy_preprocessor.json

@@ -0,0 +1,84 @@
+{
+  "name": "policy_preprocessor",
+  "steps": [
+    {
+      "registry_name": "rename_observations_processor",
+      "config": {
+        "rename_map": {}
+      }
+    },
+    {
+      "registry_name": "to_batch_processor",
+      "config": {}
+    },
+    {
+      "registry_name": "tokenizer_processor",
+      "config": {
+        "max_length": 77,
+        "task_key": "task",
+        "padding_side": "right",
+        "padding": "max_length",
+        "truncation": true,
+        "tokenizer_name": "openai/clip-vit-base-patch16"
+      }
+    },
+    {
+      "registry_name": "device_processor",
+      "config": {
+        "device": "cuda",
+        "float_dtype": null
+      }
+    },
+    {
+      "registry_name": "delta_actions_processor",
+      "config": {
+        "enabled": false,
+        "exclude_joints": [
+          "gripper"
+        ],
+        "action_names": [
+          "shoulder_pan.pos",
+          "shoulder_lift.pos",
+          "elbow_flex.pos",
+          "wrist_flex.pos",
+          "wrist_roll.pos",
+          "gripper.pos"
+        ]
+      }
+    },
+    {
+      "registry_name": "normalizer_processor",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "observation.state": {
+            "type": "STATE",
+            "shape": [
+              6
+            ]
+          },
+          "observation.images.front": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              480,
+              640
+            ]
+          },
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              6
+            ]
+          }
+        },
+        "norm_map": {
+          "VISUAL": "MEAN_STD",
+          "STATE": "MIN_MAX",
+          "ACTION": "MIN_MAX"
+        }
+      },
+      "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
+    }
+  ]
+}

060000/pretrained_model/policy_preprocessor_step_5_normalizer_processor.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bba8954e4fa3a82ef9fdfddf7a4da6e6e4ca774c3890a8bcdeb91892a46e054
+size 6608

060000/pretrained_model/train_config.json

@@ -0,0 +1,259 @@
+{
+    "dataset": {
+        "repo_id": "gaspardthrl/walleed_fold_combined",
+        "root": null,
+        "episodes": null,
+        "image_transforms": {
+            "enable": true,
+            "max_num_transforms": 6,
+            "random_order": true,
+            "tfs": {
+                "brightness": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "brightness": [
+                            0.7,
+                            1.3
+                        ]
+                    }
+                },
+                "contrast": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "contrast": [
+                            0.7,
+                            1.3
+                        ]
+                    }
+                },
+                "sharpness": {
+                    "weight": 1.0,
+                    "type": "SharpnessJitter",
+                    "kwargs": {
+                        "sharpness": [
+                            0.6,
+                            1.4
+                        ]
+                    }
+                },
+                "warmth": {
+                    "weight": 1.0,
+                    "type": "WarmthJitter",
+                    "kwargs": {
+                        "warmth": [
+                            -0.15,
+                            0.15
+                        ]
+                    }
+                },
+                "affine": {
+                    "weight": 1.0,
+                    "type": "RandomAffine",
+                    "kwargs": {
+                        "degrees": [
+                            -8,
+                            8
+                        ],
+                        "translate": [
+                            0.12,
+                            0.12
+                        ]
+                    }
+                },
+                "erasing": {
+                    "weight": 0.5,
+                    "type": "RandomErasing",
+                    "kwargs": {
+                        "p": 0.35,
+                        "scale": [
+                            0.02,
+                            0.1
+                        ],
+                        "ratio": [
+                            0.3,
+                            3.3
+                        ],
+                        "value": 0
+                    }
+                }
+            }
+        },
+        "revision": null,
+        "use_imagenet_stats": true,
+        "stats_override_path": null,
+        "video_backend": "pyav",
+        "return_uint8": false,
+        "streaming": false
+    },
+    "env": null,
+    "policy": {
+        "type": "multi_task_dit",
+        "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": "gaspardthrl/walleed_dit_fm_absolute",
+        "private": null,
+        "tags": null,
+        "license": null,
+        "pretrained_path": null,
+        "obs_stride": 1,
+        "state_history_motors": [],
+        "state_history_lags": [],
+        "state_history_scales": [],
+        "horizon": 32,
+        "n_action_steps": 24,
+        "objective": "flow_matching",
+        "noise_scheduler_type": "DDPM",
+        "num_train_timesteps": 100,
+        "beta_schedule": "squaredcos_cap_v2",
+        "beta_start": 0.0001,
+        "beta_end": 0.02,
+        "prediction_type": "epsilon",
+        "clip_sample": true,
+        "clip_sample_range": 1.0,
+        "num_inference_steps": null,
+        "sigma_min": 0.0,
+        "num_integration_steps": 10,
+        "integration_method": "euler",
+        "timestep_sampling_strategy": "beta",
+        "timestep_sampling_s": 0.999,
+        "timestep_sampling_alpha": 1.5,
+        "timestep_sampling_beta": 1.0,
+        "hidden_dim": 512,
+        "num_layers": 6,
+        "num_heads": 8,
+        "dropout": 0.1,
+        "use_positional_encoding": false,
+        "timestep_embed_dim": 256,
+        "use_rope": true,
+        "rope_base": 10000.0,
+        "vision_encoder_name": "openai/clip-vit-base-patch16",
+        "use_separate_rgb_encoder_per_camera": false,
+        "vision_encoder_lr_multiplier": 0.0,
+        "image_resize_shape": [
+            224,
+            224
+        ],
+        "image_crop_shape": [
+            224,
+            224
+        ],
+        "image_crop_is_random": true,
+        "image_grayscale": true,
+        "image_grayworld": true,
+        "use_relative_actions": false,
+        "relative_exclude_joints": [
+            "gripper"
+        ],
+        "action_feature_names": [
+            "shoulder_pan.pos",
+            "shoulder_lift.pos",
+            "elbow_flex.pos",
+            "wrist_flex.pos",
+            "wrist_roll.pos",
+            "gripper.pos"
+        ],
+        "text_encoder_name": "openai/clip-vit-base-patch16",
+        "tokenizer_max_length": 77,
+        "tokenizer_padding": "max_length",
+        "tokenizer_padding_side": "right",
+        "tokenizer_truncation": true,
+        "default_task": "Fold the towel",
+        "normalization_mapping": {
+            "VISUAL": "MEAN_STD",
+            "STATE": "MIN_MAX",
+            "ACTION": "MIN_MAX"
+        },
+        "optimizer_lr": 2e-05,
+        "optimizer_betas": [
+            0.95,
+            0.999
+        ],
+        "optimizer_eps": 1e-08,
+        "optimizer_weight_decay": 0.0,
+        "scheduler_name": "cosine",
+        "scheduler_warmup_steps": 1000,
+        "do_mask_loss_for_padding": false,
+        "drop_n_last_frames": 8
+    },
+    "reward_model": null,
+    "output_dir": "outputs/dit_fm_absolute_20260513_213457",
+    "job_name": "dit-fm-absolute",
+    "resume": false,
+    "seed": 1000,
+    "cudnn_deterministic": false,
+    "num_workers": 8,
+    "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": "adam",
+        "lr": 2e-05,
+        "weight_decay": 0.0,
+        "grad_clip_norm": 10.0,
+        "betas": [
+            0.95,
+            0.999
+        ],
+        "eps": 1e-08
+    },
+    "scheduler": {
+        "type": "diffuser",
+        "num_warmup_steps": 1000,
+        "name": "cosine"
+    },
+    "eval": {
+        "n_episodes": 50,
+        "batch_size": 8,
+        "use_async_envs": true
+    },
+    "wandb": {
+        "enable": true,
+        "disable_artifact": false,
+        "project": "diffusing",
+        "entity": null,
+        "notes": "DiT FM 100000 steps, ABSOLUTE actions, frozen CLIP, grayworld+warmth, n_obs=1",
+        "run_id": "9ttlcfmu",
+        "mode": null,
+        "add_tags": true
+    },
+    "peft": null,
+    "sample_weighting": null,
+    "rename_map": {},
+    "checkpoint_path": null
+}