Initial commit

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+wandb/offline-run-20251125_163418-lizvmqey/run-lizvmqey.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/offline-run-20251127_112044-lizvmqey/run-lizvmqey.wandb filter=lfs diff=lfs merge=lfs -text

checkpoints/060000/pretrained_model/config.json

@@ -0,0 +1,134 @@
+{
+    "type": "smolvla",
+    "n_obs_steps": 1,
+    "normalization_mapping": {
+        "VISUAL": "IDENTITY",
+        "STATE": "MEAN_STD",
+        "ACTION": "MEAN_STD"
+    },
+    "input_features": {
+        "observation.images.image": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                256,
+                256
+            ]
+        },
+        "observation.images.wrist_image": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                256,
+                256
+            ]
+        },
+        "observation.images.image_mask": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                256,
+                256
+            ]
+        },
+        "observation.images.wrist_mask": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                256,
+                256
+            ]
+        },
+        "observation.images.object_of_interest_mask": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                256,
+                256
+            ]
+        },
+        "observation.images.object_of_interest_wrist_mask": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                256,
+                256
+            ]
+        },
+        "observation.state": {
+            "type": "STATE",
+            "shape": [
+                8
+            ]
+        },
+        "observation.states.ee_state": {
+            "type": "STATE",
+            "shape": [
+                6
+            ]
+        },
+        "observation.states.joint_state": {
+            "type": "STATE",
+            "shape": [
+                7
+            ]
+        },
+        "observation.states.gripper_state": {
+            "type": "STATE",
+            "shape": [
+                2
+            ]
+        }
+    },
+    "output_features": {
+        "action": {
+            "type": "ACTION",
+            "shape": [
+                7
+            ]
+        }
+    },
+    "device": "cuda",
+    "use_amp": false,
+    "gradient_accumulation_steps": 1,
+    "chunk_size": 50,
+    "n_action_steps": 50,
+    "max_state_dim": 32,
+    "max_action_dim": 32,
+    "resize_imgs_with_padding": [
+        512,
+        512
+    ],
+    "empty_cameras": 0,
+    "adapt_to_pi_aloha": false,
+    "use_delta_joint_actions_aloha": false,
+    "tokenizer_max_length": 48,
+    "num_steps": 10,
+    "use_cache": true,
+    "freeze_vision_encoder": true,
+    "train_expert_only": false,
+    "train_state_proj": true,
+    "optimizer_lr": 0.0001,
+    "optimizer_betas": [
+        0.9,
+        0.95
+    ],
+    "optimizer_eps": 1e-08,
+    "optimizer_weight_decay": 1e-10,
+    "optimizer_grad_clip_norm": 10.0,
+    "scheduler_warmup_steps": 1000,
+    "scheduler_decay_steps": 30000,
+    "scheduler_decay_lr": 2.5e-06,
+    "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
+    "load_vlm_weights": true,
+    "add_image_special_tokens": false,
+    "attention_mode": "cross_attn",
+    "prefix_length": 0,
+    "pad_language_to": "max_length",
+    "num_expert_layers": 0,
+    "num_vlm_layers": 16,
+    "self_attn_every_n_layers": 2,
+    "expert_width_multiplier": 0.75,
+    "min_period": 0.004,
+    "max_period": 4.0
+}

checkpoints/060000/pretrained_model/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f28ae8b369209f4fbd67a0f6a520d9d8f9b2e8b3e0465bb3c210845e0d13b50c
+size 910404848

checkpoints/060000/pretrained_model/train_config.json

@@ -0,0 +1,327 @@
+{
+    "dataset": {
+        "repo_id": ".",
+        "root": "/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot",
+        "episodes": null,
+        "image_transforms": {
+            "enable": true,
+            "max_num_transforms": 3,
+            "random_order": false,
+            "image_tfs": {
+                "hue": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "hue": [
+                            -0.05,
+                            0.05
+                        ]
+                    }
+                },
+                "contrast": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "contrast": [
+                            0.8,
+                            1.2
+                        ]
+                    }
+                },
+                "sharpness": {
+                    "weight": 1.0,
+                    "type": "SharpnessJitter",
+                    "kwargs": {
+                        "sharpness": [
+                            0.5,
+                            1.5
+                        ]
+                    }
+                },
+                "brightness": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "brightness": [
+                            0.8,
+                            1.2
+                        ]
+                    }
+                },
+                "saturation": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "saturation": [
+                            0.5,
+                            1.5
+                        ]
+                    }
+                },
+                "crop_resize": {
+                    "weight": 1.0,
+                    "type": "RandomResizedCrop",
+                    "kwargs": {
+                        "size": [
+                            256,
+                            256
+                        ],
+                        "ratio": [
+                            1,
+                            1
+                        ],
+                        "scale": [
+                            0.9,
+                            0.95
+                        ]
+                    }
+                },
+                "rotate": {
+                    "weight": 1.0,
+                    "type": "RandomRotate",
+                    "kwargs": {
+                        "degrees": [
+                            -5,
+                            5
+                        ]
+                    }
+                }
+            },
+            "wrist_tfs": {
+                "hue": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "hue": [
+                            -0.05,
+                            0.05
+                        ]
+                    }
+                },
+                "contrast": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "contrast": [
+                            0.8,
+                            1.2
+                        ]
+                    }
+                },
+                "sharpness": {
+                    "weight": 1.0,
+                    "type": "SharpnessJitter",
+                    "kwargs": {
+                        "sharpness": [
+                            0.5,
+                            1.5
+                        ]
+                    }
+                },
+                "brightness": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "brightness": [
+                            0.8,
+                            1.2
+                        ]
+                    }
+                },
+                "saturation": {
+                    "weight": 1.0,
+                    "type": "ColorJitter",
+                    "kwargs": {
+                        "saturation": [
+                            0.5,
+                            1.5
+                        ]
+                    }
+                }
+            }
+        },
+        "revision": null,
+        "use_imagenet_stats": true,
+        "video_backend": "torchcodec",
+        "vqa_data_path": null
+    },
+    "env": null,
+    "policy": {
+        "type": "smolvla",
+        "n_obs_steps": 1,
+        "normalization_mapping": {
+            "VISUAL": "IDENTITY",
+            "STATE": "MEAN_STD",
+            "ACTION": "MEAN_STD"
+        },
+        "input_features": {
+            "observation.images.image": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    256,
+                    256
+                ]
+            },
+            "observation.images.wrist_image": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    256,
+                    256
+                ]
+            },
+            "observation.images.image_mask": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    256,
+                    256
+                ]
+            },
+            "observation.images.wrist_mask": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    256,
+                    256
+                ]
+            },
+            "observation.images.object_of_interest_mask": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    256,
+                    256
+                ]
+            },
+            "observation.images.object_of_interest_wrist_mask": {
+                "type": "VISUAL",
+                "shape": [
+                    3,
+                    256,
+                    256
+                ]
+            },
+            "observation.state": {
+                "type": "STATE",
+                "shape": [
+                    8
+                ]
+            },
+            "observation.states.ee_state": {
+                "type": "STATE",
+                "shape": [
+                    6
+                ]
+            },
+            "observation.states.joint_state": {
+                "type": "STATE",
+                "shape": [
+                    7
+                ]
+            },
+            "observation.states.gripper_state": {
+                "type": "STATE",
+                "shape": [
+                    2
+                ]
+            }
+        },
+        "output_features": {
+            "action": {
+                "type": "ACTION",
+                "shape": [
+                    7
+                ]
+            }
+        },
+        "device": "cuda",
+        "use_amp": false,
+        "gradient_accumulation_steps": 1,
+        "chunk_size": 50,
+        "n_action_steps": 50,
+        "max_state_dim": 32,
+        "max_action_dim": 32,
+        "resize_imgs_with_padding": [
+            512,
+            512
+        ],
+        "empty_cameras": 0,
+        "adapt_to_pi_aloha": false,
+        "use_delta_joint_actions_aloha": false,
+        "tokenizer_max_length": 48,
+        "num_steps": 10,
+        "use_cache": true,
+        "freeze_vision_encoder": true,
+        "train_expert_only": false,
+        "train_state_proj": true,
+        "optimizer_lr": 0.0001,
+        "optimizer_betas": [
+            0.9,
+            0.95
+        ],
+        "optimizer_eps": 1e-08,
+        "optimizer_weight_decay": 1e-10,
+        "optimizer_grad_clip_norm": 10.0,
+        "scheduler_warmup_steps": 1000,
+        "scheduler_decay_steps": 30000,
+        "scheduler_decay_lr": 2.5e-06,
+        "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
+        "load_vlm_weights": true,
+        "add_image_special_tokens": false,
+        "attention_mode": "cross_attn",
+        "prefix_length": 0,
+        "pad_language_to": "max_length",
+        "num_expert_layers": 0,
+        "num_vlm_layers": 16,
+        "self_attn_every_n_layers": 2,
+        "expert_width_multiplier": 0.75,
+        "min_period": 0.004,
+        "max_period": 4.0
+    },
+    "output_dir": "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base",
+    "job_name": "libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base",
+    "resume": true,
+    "seed": 42,
+    "num_workers": 8,
+    "batch_size": 64,
+    "steps": 100000,
+    "eval_freq": 20000,
+    "log_freq": 10,
+    "save_checkpoint": true,
+    "save_freq": 10000,
+    "use_policy_training_preset": true,
+    "optimizer": {
+        "type": "adamw",
+        "lr": 0.0001,
+        "weight_decay": 1e-10,
+        "grad_clip_norm": 10.0,
+        "betas": [
+            0.9,
+            0.95
+        ],
+        "eps": 1e-08
+    },
+    "scheduler": {
+        "type": "cosine_decay_with_warmup",
+        "num_warmup_steps": 1000,
+        "num_decay_steps": 30000,
+        "peak_lr": 0.0001,
+        "decay_lr": 2.5e-06
+    },
+    "eval": {
+        "n_episodes": 50,
+        "batch_size": 50,
+        "use_async_envs": false
+    },
+    "wandb": {
+        "enable": true,
+        "disable_artifact": true,
+        "project": "smolvla",
+        "entity": "Robotics_VLA",
+        "notes": null,
+        "run_id": null,
+        "mode": "offline"
+    }
+}

checkpoints/060000/training_state/optimizer_param_groups.json

@@ -0,0 +1,537 @@
+[
+    {
+        "lr": 2.5e-06,
+        "betas": [
+            0.9,
+            0.95
+        ],
+        "eps": 1e-08,
+        "weight_decay": 1e-10,
+        "amsgrad": false,
+        "maximize": false,
+        "foreach": null,
+        "capturable": false,
+        "differentiable": false,
+        "fused": null,
+        "decoupled_weight_decay": true,
+        "initial_lr": 0.0001,
+        "params": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39,
+            40,
+            41,
+            42,
+            43,
+            44,
+            45,
+            46,
+            47,
+            48,
+            49,
+            50,
+            51,
+            52,
+            53,
+            54,
+            55,
+            56,
+            57,
+            58,
+            59,
+            60,
+            61,
+            62,
+            63,
+            64,
+            65,
+            66,
+            67,
+            68,
+            69,
+            70,
+            71,
+            72,
+            73,
+            74,
+            75,
+            76,
+            77,
+            78,
+            79,
+            80,
+            81,
+            82,
+            83,
+            84,
+            85,
+            86,
+            87,
+            88,
+            89,
+            90,
+            91,
+            92,
+            93,
+            94,
+            95,
+            96,
+            97,
+            98,
+            99,
+            100,
+            101,
+            102,
+            103,
+            104,
+            105,
+            106,
+            107,
+            108,
+            109,
+            110,
+            111,
+            112,
+            113,
+            114,
+            115,
+            116,
+            117,
+            118,
+            119,
+            120,
+            121,
+            122,
+            123,
+            124,
+            125,
+            126,
+            127,
+            128,
+            129,
+            130,
+            131,
+            132,
+            133,
+            134,
+            135,
+            136,
+            137,
+            138,
+            139,
+            140,
+            141,
+            142,
+            143,
+            144,
+            145,
+            146,
+            147,
+            148,
+            149,
+            150,
+            151,
+            152,
+            153,
+            154,
+            155,
+            156,
+            157,
+            158,
+            159,
+            160,
+            161,
+            162,
+            163,
+            164,
+            165,
+            166,
+            167,
+            168,
+            169,
+            170,
+            171,
+            172,
+            173,
+            174,
+            175,
+            176,
+            177,
+            178,
+            179,
+            180,
+            181,
+            182,
+            183,
+            184,
+            185,
+            186,
+            187,
+            188,
+            189,
+            190,
+            191,
+            192,
+            193,
+            194,
+            195,
+            196,
+            197,
+            198,
+            199,
+            200,
+            201,
+            202,
+            203,
+            204,
+            205,
+            206,
+            207,
+            208,
+            209,
+            210,
+            211,
+            212,
+            213,
+            214,
+            215,
+            216,
+            217,
+            218,
+            219,
+            220,
+            221,
+            222,
+            223,
+            224,
+            225,
+            226,
+            227,
+            228,
+            229,
+            230,
+            231,
+            232,
+            233,
+            234,
+            235,
+            236,
+            237,
+            238,
+            239,
+            240,
+            241,
+            242,
+            243,
+            244,
+            245,
+            246,
+            247,
+            248,
+            249,
+            250,
+            251,
+            252,
+            253,
+            254,
+            255,
+            256,
+            257,
+            258,
+            259,
+            260,
+            261,
+            262,
+            263,
+            264,
+            265,
+            266,
+            267,
+            268,
+            269,
+            270,
+            271,
+            272,
+            273,
+            274,
+            275,
+            276,
+            277,
+            278,
+            279,
+            280,
+            281,
+            282,
+            283,
+            284,
+            285,
+            286,
+            287,
+            288,
+            289,
+            290,
+            291,
+            292,
+            293,
+            294,
+            295,
+            296,
+            297,
+            298,
+            299,
+            300,
+            301,
+            302,
+            303,
+            304,
+            305,
+            306,
+            307,
+            308,
+            309,
+            310,
+            311,
+            312,
+            313,
+            314,
+            315,
+            316,
+            317,
+            318,
+            319,
+            320,
+            321,
+            322,
+            323,
+            324,
+            325,
+            326,
+            327,
+            328,
+            329,
+            330,
+            331,
+            332,
+            333,
+            334,
+            335,
+            336,
+            337,
+            338,
+            339,
+            340,
+            341,
+            342,
+            343,
+            344,
+            345,
+            346,
+            347,
+            348,
+            349,
+            350,
+            351,
+            352,
+            353,
+            354,
+            355,
+            356,
+            357,
+            358,
+            359,
+            360,
+            361,
+            362,
+            363,
+            364,
+            365,
+            366,
+            367,
+            368,
+            369,
+            370,
+            371,
+            372,
+            373,
+            374,
+            375,
+            376,
+            377,
+            378,
+            379,
+            380,
+            381,
+            382,
+            383,
+            384,
+            385,
+            386,
+            387,
+            388,
+            389,
+            390,
+            391,
+            392,
+            393,
+            394,
+            395,
+            396,
+            397,
+            398,
+            399,
+            400,
+            401,
+            402,
+            403,
+            404,
+            405,
+            406,
+            407,
+            408,
+            409,
+            410,
+            411,
+            412,
+            413,
+            414,
+            415,
+            416,
+            417,
+            418,
+            419,
+            420,
+            421,
+            422,
+            423,
+            424,
+            425,
+            426,
+            427,
+            428,
+            429,
+            430,
+            431,
+            432,
+            433,
+            434,
+            435,
+            436,
+            437,
+            438,
+            439,
+            440,
+            441,
+            442,
+            443,
+            444,
+            445,
+            446,
+            447,
+            448,
+            449,
+            450,
+            451,
+            452,
+            453,
+            454,
+            455,
+            456,
+            457,
+            458,
+            459,
+            460,
+            461,
+            462,
+            463,
+            464,
+            465,
+            466,
+            467,
+            468,
+            469,
+            470,
+            471,
+            472,
+            473,
+            474,
+            475,
+            476,
+            477,
+            478,
+            479,
+            480,
+            481,
+            482,
+            483,
+            484,
+            485,
+            486,
+            487,
+            488,
+            489,
+            490,
+            491,
+            492,
+            493,
+            494,
+            495,
+            496,
+            497,
+            498,
+            499,
+            500,
+            501,
+            502,
+            503,
+            504,
+            505,
+            506,
+            507,
+            508,
+            509,
+            510,
+            511,
+            512,
+            513,
+            514,
+            515
+        ]
+    }
+]

checkpoints/060000/training_state/optimizer_state.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f1b1472227f3579bf9dad78fc92c1d39265b7e953a177420736bef19ed47571
+size 1248899100

checkpoints/060000/training_state/rng_state.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:236fc8d31470b51befb4e56e67358ab7dc83b2a431f063c80b5056af341b87b7
+size 15708

checkpoints/060000/training_state/scheduler_state.json

@@ -0,0 +1,15 @@
+{
+    "base_lrs": [
+        0.0001
+    ],
+    "last_epoch": 60000,
+    "_step_count": 60001,
+    "_is_initial": false,
+    "_get_lr_called_within_step": false,
+    "_last_lr": [
+        2.5e-06
+    ],
+    "lr_lambdas": [
+        null
+    ]
+}

checkpoints/060000/training_state/training_step.json

@@ -0,0 +1,3 @@
+{
+    "step": 60000
+}

wandb/debug-internal.log

@@ -0,0 +1,13 @@
+{"time":"2025-11-27T11:20:45.378937954+01:00","level":"INFO","msg":"stream: starting","core version":"0.22.3"}
+{"time":"2025-11-27T11:20:45.409919704+01:00","level":"ERROR","msg":"monitor: failed to initialize GPU resource: monitor: could not create portfile"}
+{"time":"2025-11-27T11:20:45.425921086+01:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2025-11-27T11:20:45.425965495+01:00","level":"INFO","msg":"stream: created new stream","id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.425988095+01:00","level":"INFO","msg":"handler: started","stream_id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.428918285+01:00","level":"INFO","msg":"stream: started","id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.428917715+01:00","level":"INFO","msg":"sender: started","stream_id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.428956845+01:00","level":"INFO","msg":"writer: started","stream_id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.429853649+01:00","level":"WARN","msg":"runupserter: server does not expand metric globs but the x_server_side_expand_glob_metrics setting is set; ignoring"}
+{"time":"2025-11-27T14:51:55.122025249+01:00","level":"INFO","msg":"stream: closing","id":"lizvmqey"}
+{"time":"2025-11-27T14:51:55.125975233+01:00","level":"INFO","msg":"handler: closed","stream_id":"lizvmqey"}
+{"time":"2025-11-27T14:51:55.127619953+01:00","level":"INFO","msg":"sender: closed","stream_id":"lizvmqey"}
+{"time":"2025-11-27T14:51:55.127656972+01:00","level":"INFO","msg":"stream: closed","id":"lizvmqey"}

wandb/debug.log

@@ -0,0 +1,49 @@
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Current SDK version is 0.22.3
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Configure stats pid to 1938939
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Loading settings from /user/hominhduy.nguyen/u15271/.config/wandb/settings
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Loading settings from /mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/Smolvla_capstone_project/wandb/settings
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:setup_run_log_directory():706] Logging user logs to /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base/wandb/offline-run-20251127_112044-lizvmqey/logs/debug.log
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:setup_run_log_directory():707] Logging internal logs to /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base/wandb/offline-run-20251127_112044-lizvmqey/logs/debug-internal.log
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:init():833] calling init triggers
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:init():838] wandb.init called with sweep_config: {}
+config: {'dataset': {'repo_id': '.', 'root': '/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot', 'episodes': None, 'image_transforms': {'enable': True, 'max_num_transforms': 3, 'random_order': False, 'image_tfs': {'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'crop_resize': {'weight': 1.0, 'type': 'RandomResizedCrop', 'kwargs': {'size': [256, 256], 'ratio': [1, 1], 'scale': [0.9, 0.95]}}, 'rotate': {'weight': 1.0, 'type': 'RandomRotate', 'kwargs': {'degrees': [-5, 5]}}}, 'wrist_tfs': {'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'vqa_data_path': None}, 'env': None, 'policy': {'type': 'smolvla', 'n_obs_steps': 1, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.MEAN_STD: 'MEAN_STD'>, 'ACTION': <NormalizationMode.MEAN_STD: 'MEAN_STD'>}, 'input_features': {'observation.images.image': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.wrist_image': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.image_mask': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.wrist_mask': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.object_of_interest_mask': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.object_of_interest_wrist_mask': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [8]}, 'observation.states.ee_state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [6]}, 'observation.states.joint_state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [7]}, 'observation.states.gripper_state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [2]}}, 'output_features': {'action': {'type': <FeatureType.ACTION: 'ACTION'>, 'shape': [7]}}, 'device': 'cuda', 'use_amp': False, 'gradient_accumulation_steps': 1, 'chunk_size': 50, 'n_action_steps': 50, 'max_state_dim': 32, 'max_action_dim': 32, 'resize_imgs_with_padding': [512, 512], 'empty_cameras': 0, 'adapt_to_pi_aloha': False, 'use_delta_joint_actions_aloha': False, 'tokenizer_max_length': 48, 'num_steps': 10, 'use_cache': True, 'freeze_vision_encoder': True, 'train_expert_only': False, 'train_state_proj': True, 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.95], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-10, 'optimizer_grad_clip_norm': 10.0, 'scheduler_warmup_steps': 1000, 'scheduler_decay_steps': 30000, 'scheduler_decay_lr': 2.5e-06, 'vlm_model_name': 'HuggingFaceTB/SmolVLM2-500M-Video-Instruct', 'load_vlm_weights': True, 'add_image_special_tokens': False, 'attention_mode': 'cross_attn', 'prefix_length': 0, 'pad_language_to': 'max_length', 'num_expert_layers': 0, 'num_vlm_layers': 16, 'self_attn_every_n_layers': 2, 'expert_width_multiplier': 0.75, 'min_period': 0.004, 'max_period': 4.0}, 'output_dir': '/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base', 'job_name': 'libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base', 'resume': True, 'seed': 42, 'num_workers': 8, 'batch_size': 64, 'steps': 100000, 'eval_freq': 20000, 'log_freq': 10, 'save_checkpoint': True, 'save_freq': 10000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-10, 'grad_clip_norm': 10.0, 'betas': [0.9, 0.95], 'eps': 1e-08}, 'scheduler': {'type': 'cosine_decay_with_warmup', 'num_warmup_steps': 1000, 'num_decay_steps': 30000, 'peak_lr': 0.0001, 'decay_lr': 2.5e-06}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'smolvla', 'entity': 'Robotics_VLA', 'notes': None, 'run_id': None, 'mode': 'offline'}, '_wandb': {}}
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:init():881] starting backend
+2025-11-27 11:20:45,217 INFO    MainThread:1938939 [wandb_init.py:init():884] sending inform_init request
+2025-11-27 11:20:45,227 INFO    MainThread:1938939 [wandb_init.py:init():892] backend started and connected
+2025-11-27 11:20:45,229 INFO    MainThread:1938939 [wandb_init.py:init():962] updated telemetry
+2025-11-27 11:20:45,248 INFO    MainThread:1938939 [wandb_init.py:init():986] communicating run to backend with 90.0 second timeout
+2025-11-27 11:20:45,432 INFO    MainThread:1938939 [wandb_init.py:init():1033] starting run threads in backend
+2025-11-27 11:20:45,788 INFO    MainThread:1938939 [wandb_run.py:_console_start():2506] atexit reg
+2025-11-27 11:20:45,788 INFO    MainThread:1938939 [wandb_run.py:_redirect():2354] redirect: wrap_raw
+2025-11-27 11:20:45,789 INFO    MainThread:1938939 [wandb_run.py:_redirect():2423] Wrapping output streams.
+2025-11-27 11:20:45,789 INFO    MainThread:1938939 [wandb_run.py:_redirect():2446] Redirects installed.
+2025-11-27 11:20:45,800 INFO    MainThread:1938939 [wandb_init.py:init():1073] run started, returning control to user process
+2025-11-27 14:51:55,121 INFO    wandb-AsyncioManager-main:1938939 [service_client.py:_forward_responses():80] Reached EOF.
+2025-11-27 14:51:55,122 INFO    wandb-AsyncioManager-main:1938939 [mailbox.py:close():137] Closing mailbox, abandoning 0 handles.
+2025-11-27 14:51:55,307 ERROR   wandb-AsyncioManager-main:1938939 [asyncio_manager.py:fn_wrap_exceptions():183] Uncaught exception in run_soon callback.
+Traceback (most recent call last):
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/asyncio_manager.py", line 181, in fn_wrap_exceptions
+    await fn()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/service/service_client.py", line 38, in publish
+    await self._send_server_request(request)
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/service/service_client.py", line 64, in _send_server_request
+    await self._writer.drain()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/asyncio/streams.py", line 371, in drain
+    await self._protocol._drain_helper()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/asyncio/streams.py", line 167, in _drain_helper
+    raise ConnectionResetError('Connection lost')
+ConnectionResetError: Connection lost
+2025-11-27 14:51:55,322 ERROR   wandb-AsyncioManager-main:1938939 [asyncio_manager.py:fn_wrap_exceptions():183] Uncaught exception in run_soon callback.
+Traceback (most recent call last):
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/asyncio_manager.py", line 181, in fn_wrap_exceptions
+    await fn()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/service/service_client.py", line 38, in publish
+    await self._send_server_request(request)
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/service/service_client.py", line 64, in _send_server_request
+    await self._writer.drain()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/asyncio/streams.py", line 371, in drain
+    await self._protocol._drain_helper()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/asyncio/streams.py", line 167, in _drain_helper
+    raise ConnectionResetError('Connection lost')
+ConnectionResetError: Connection lost

wandb/offline-run-20251125_163418-lizvmqey/files/config.yaml

@@ -0,0 +1,326 @@
+wandb_version: 1
+
+_wandb:
+  desc: null
+  value:
+    python_version: 3.10.19
+    cli_version: 0.22.3
+    framework: huggingface
+    is_jupyter_run: false
+    is_kaggle_kernel: false
+    start_time: 1764084858
+    t:
+      1:
+      - 1
+      - 41
+      - 49
+      - 51
+      - 71
+      2:
+      - 1
+      - 2
+      - 3
+      - 11
+      - 41
+      - 49
+      - 51
+      - 71
+      3:
+      - 4
+      - 13
+      - 15
+      - 16
+      - 42
+      - 61
+      4: 3.10.19
+      5: 0.22.3
+      13: linux-x86_64
+    e:
+      tgzz60htaej5eivyg1ohkojicqbx6n35:
+        os: Linux-4.18.0-553.77.1.el8_10.x86_64-x86_64-with-glibc2.28
+        python: CPython 3.10.19
+        started_at: '2025-11-25T15:34:18.399929Z'
+        args:
+        - --policy.path=/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/smol_pretrained/smolvla_base
+        - --dataset.root=/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot
+        - --output_dir=outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+        - --job_name=libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+        - --config_path=configs/smolvla_config/other_default.json
+        - --batch_size=64
+        - --steps=100000
+        - --policy.gradient_accumulation_steps=1
+        - --wandb.mode=offline
+        - --log_freq=10
+        program: /mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/Smolvla_capstone_project/lerobot/scripts/train_accelerate.py
+        code_path: lerobot/scripts/train_accelerate.py
+        code_path_local: lerobot/scripts/train_accelerate.py
+        git:
+          remote_url: git@github-binh:jibby2803/Smolvla_capstone_project.git
+          commit: 0f72ba315e381d6a8095782824a2d0c70faad469
+        root: outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+        host: ggpu192
+        executable: /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/bin/python3.10
+        cpu_count: 64
+        cpu_count_logical: 128
+        disk:
+          /:
+            total: '270465425408'
+            used: '7019204608'
+        memory:
+          total: '540930854912'
+        slurm:
+          job_id: '11843069'
+          mpi_type: pmi2
+        writer_id: tgzz60htaej5eivyg1ohkojicqbx6n35
+dataset:
+  desc: null
+  value:
+    repo_id: .
+    root: /mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot
+    episodes: null
+    image_transforms:
+      enable: true
+      max_num_transforms: 3
+      random_order: false
+      image_tfs:
+        hue:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            hue:
+            - -0.05
+            - 0.05
+        contrast:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            contrast:
+            - 0.8
+            - 1.2
+        sharpness:
+          weight: 1.0
+          type: SharpnessJitter
+          kwargs:
+            sharpness:
+            - 0.5
+            - 1.5
+        brightness:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            brightness:
+            - 0.8
+            - 1.2
+        saturation:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            saturation:
+            - 0.5
+            - 1.5
+        crop_resize:
+          weight: 1.0
+          type: RandomResizedCrop
+          kwargs:
+            size:
+            - 256
+            - 256
+            ratio:
+            - 1
+            - 1
+            scale:
+            - 0.9
+            - 0.95
+        rotate:
+          weight: 1.0
+          type: RandomRotate
+          kwargs:
+            degrees:
+            - -5
+            - 5
+      wrist_tfs:
+        hue:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            hue:
+            - -0.05
+            - 0.05
+        contrast:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            contrast:
+            - 0.8
+            - 1.2
+        sharpness:
+          weight: 1.0
+          type: SharpnessJitter
+          kwargs:
+            sharpness:
+            - 0.5
+            - 1.5
+        brightness:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            brightness:
+            - 0.8
+            - 1.2
+        saturation:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            saturation:
+            - 0.5
+            - 1.5
+    revision: null
+    use_imagenet_stats: true
+    video_backend: torchcodec
+    vqa_data_path: null
+env:
+  desc: null
+  value: null
+policy:
+  desc: null
+  value:
+    type: smolvla
+    n_obs_steps: 1
+    normalization_mapping:
+      VISUAL: IDENTITY
+      STATE: MEAN_STD
+      ACTION: MEAN_STD
+    input_features:
+      observation.images.image:
+        type: VISUAL
+        shape:
+        - 3
+        - 256
+        - 256
+      observation.images.wrist_image:
+        type: VISUAL
+        shape:
+        - 3
+        - 256
+        - 256
+      observation.state:
+        type: STATE
+        shape:
+        - 8
+    output_features:
+      action:
+        type: ACTION
+        shape:
+        - 7
+    device: cuda
+    use_amp: false
+    gradient_accumulation_steps: 1
+    chunk_size: 50
+    n_action_steps: 50
+    max_state_dim: 32
+    max_action_dim: 32
+    resize_imgs_with_padding:
+    - 512
+    - 512
+    empty_cameras: 0
+    adapt_to_pi_aloha: false
+    use_delta_joint_actions_aloha: false
+    tokenizer_max_length: 48
+    num_steps: 10
+    use_cache: true
+    freeze_vision_encoder: true
+    train_expert_only: false
+    train_state_proj: true
+    optimizer_lr: 0.0001
+    optimizer_betas:
+    - 0.9
+    - 0.95
+    optimizer_eps: 1.0e-08
+    optimizer_weight_decay: 1.0e-10
+    optimizer_grad_clip_norm: 10.0
+    scheduler_warmup_steps: 1000
+    scheduler_decay_steps: 30000
+    scheduler_decay_lr: 2.5e-06
+    vlm_model_name: HuggingFaceTB/SmolVLM2-500M-Video-Instruct
+    load_vlm_weights: true
+    add_image_special_tokens: false
+    attention_mode: cross_attn
+    prefix_length: 0
+    pad_language_to: max_length
+    num_expert_layers: 0
+    num_vlm_layers: 16
+    self_attn_every_n_layers: 2
+    expert_width_multiplier: 0.75
+    min_period: 0.004
+    max_period: 4.0
+output_dir:
+  desc: null
+  value: outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+job_name:
+  desc: null
+  value: libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+resume:
+  desc: null
+  value: false
+seed:
+  desc: null
+  value: 42
+num_workers:
+  desc: null
+  value: 8
+batch_size:
+  desc: null
+  value: 64
+steps:
+  desc: null
+  value: 100000
+eval_freq:
+  desc: null
+  value: 20000
+log_freq:
+  desc: null
+  value: 10
+save_checkpoint:
+  desc: null
+  value: true
+save_freq:
+  desc: null
+  value: 10000
+use_policy_training_preset:
+  desc: null
+  value: true
+optimizer:
+  desc: null
+  value:
+    type: adamw
+    lr: 0.0001
+    weight_decay: 1.0e-10
+    grad_clip_norm: 10.0
+    betas:
+    - 0.9
+    - 0.95
+    eps: 1.0e-08
+scheduler:
+  desc: null
+  value:
+    type: cosine_decay_with_warmup
+    num_warmup_steps: 1000
+    num_decay_steps: 30000
+    peak_lr: 0.0001
+    decay_lr: 2.5e-06
+eval:
+  desc: null
+  value:
+    n_episodes: 50
+    batch_size: 50
+    use_async_envs: false
+wandb:
+  desc: null
+  value:
+    enable: true
+    disable_artifact: true
+    project: smolvla
+    entity: Robotics_VLA
+    notes: null
+    run_id: null
+    mode: offline

wandb/offline-run-20251125_163418-lizvmqey/files/output.log

@@ -0,0 +1,91 @@
+INFO 2025-11-25 16:34:19 celerate.py:159 Creating dataset
+Resolving data files: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1676/1676 [00:00<00:00, 21230.57it/s]
+Loading dataset shards: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 78/78 [00:00<00:00, 191.37it/s]
+INFO 2025-11-25 16:34:26 celerate.py:170 Creating policy
+`torch_dtype` is deprecated! Use `dtype` instead!
+INFO 2025-11-25 16:34:37 celerate.py:181 Creating optimizer and scheduler
+INFO 2025-11-25 16:34:37 celerate.py:221 Output dir: outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+INFO 2025-11-25 16:34:37 celerate.py:224 cfg.steps=100000 (100K)
+INFO 2025-11-25 16:34:37 celerate.py:225 dataset.num_frames=269918 (270K)
+INFO 2025-11-25 16:34:37 celerate.py:226 dataset.num_episodes=1676
+INFO 2025-11-25 16:34:37 celerate.py:227 num_learnable_params=317226912 (317M)
+INFO 2025-11-25 16:34:37 celerate.py:228 num_total_params=450968810 (451M)
+INFO 2025-11-25 16:34:37 celerate.py:229 Number of processes: 1
+INFO 2025-11-25 16:34:37 celerate.py:230 Device: cuda
+INFO 2025-11-25 16:34:37 celerate.py:231 Mixed precision: no
+INFO 2025-11-25 16:34:37 celerate.py:253 Start offline training on a fixed dataset
+load pretrained policy
+Loading  HuggingFaceTB/SmolVLM2-500M-Video-Instruct weights ...
+Load pretrained VLM successfully!
+Reducing the number of VLM layers to 16 ...
+=================================
+USING MODEL FROM BINH'S DIR
+=================================
+Loading weights from local directory
+INFO 2025-11-25 16:34:57 celerate.py:303 step:10 smpl:640 ep:4 epch:0.00 loss:2.883 grdn:11.068 lr:6.5e-07 updt_s:1.108 data_s:0.784
+INFO 2025-11-25 16:35:06 celerate.py:303 step:20 smpl:1K ep:8 epch:0.00 loss:2.981 grdn:11.440 lr:1.6e-06 updt_s:0.797 data_s:0.036
+INFO 2025-11-25 16:35:15 celerate.py:303 step:30 smpl:2K ep:12 epch:0.01 loss:2.793 grdn:10.541 lr:2.6e-06 updt_s:0.802 data_s:0.039
+INFO 2025-11-25 16:35:25 celerate.py:303 step:40 smpl:3K ep:16 epch:0.01 loss:2.765 grdn:10.229 lr:3.6e-06 updt_s:0.851 data_s:0.057
+INFO 2025-11-25 16:35:34 celerate.py:303 step:50 smpl:3K ep:20 epch:0.01 loss:2.514 grdn:8.306 lr:4.6e-06 updt_s:0.802 data_s:0.034
+INFO 2025-11-25 16:35:43 celerate.py:303 step:60 smpl:4K ep:24 epch:0.01 loss:2.231 grdn:7.228 lr:5.6e-06 updt_s:0.793 data_s:0.039
+INFO 2025-11-25 16:35:52 celerate.py:303 step:70 smpl:4K ep:28 epch:0.02 loss:2.163 grdn:6.576 lr:6.6e-06 updt_s:0.810 data_s:0.032
+INFO 2025-11-25 16:36:01 celerate.py:303 step:80 smpl:5K ep:32 epch:0.02 loss:1.997 grdn:5.803 lr:7.6e-06 updt_s:0.791 data_s:0.033
+INFO 2025-11-25 16:36:10 celerate.py:303 step:90 smpl:6K ep:36 epch:0.02 loss:1.715 grdn:4.409 lr:8.6e-06 updt_s:0.795 data_s:0.039
+INFO 2025-11-25 16:36:20 celerate.py:303 step:100 smpl:6K ep:40 epch:0.02 loss:1.668 grdn:3.878 lr:9.6e-06 updt_s:0.800 data_s:0.038
+INFO 2025-11-25 16:36:29 celerate.py:303 step:110 smpl:7K ep:44 epch:0.03 loss:1.624 grdn:3.729 lr:1.1e-05 updt_s:0.802 data_s:0.031
+INFO 2025-11-25 16:36:38 celerate.py:303 step:120 smpl:8K ep:48 epch:0.03 loss:1.479 grdn:3.140 lr:1.2e-05 updt_s:0.806 data_s:0.037
+INFO 2025-11-25 16:36:47 celerate.py:303 step:130 smpl:8K ep:52 epch:0.03 loss:1.371 grdn:2.696 lr:1.3e-05 updt_s:0.801 data_s:0.034
+INFO 2025-11-25 16:36:56 celerate.py:303 step:140 smpl:9K ep:56 epch:0.03 loss:1.314 grdn:2.589 lr:1.4e-05 updt_s:0.800 data_s:0.040
+INFO 2025-11-25 16:37:05 celerate.py:303 step:150 smpl:10K ep:60 epch:0.04 loss:1.230 grdn:2.407 lr:1.5e-05 updt_s:0.798 data_s:0.033
+INFO 2025-11-25 16:37:14 celerate.py:303 step:160 smpl:10K ep:64 epch:0.04 loss:1.169 grdn:2.512 lr:1.6e-05 updt_s:0.791 data_s:0.032
+INFO 2025-11-25 16:37:23 celerate.py:303 step:170 smpl:11K ep:68 epch:0.04 loss:1.085 grdn:2.029 lr:1.7e-05 updt_s:0.794 data_s:0.035
+INFO 2025-11-25 16:37:32 celerate.py:303 step:180 smpl:12K ep:72 epch:0.04 loss:1.038 grdn:2.114 lr:1.8e-05 updt_s:0.796 data_s:0.034
+INFO 2025-11-25 16:37:41 celerate.py:303 step:190 smpl:12K ep:76 epch:0.05 loss:0.973 grdn:1.941 lr:1.9e-05 updt_s:0.796 data_s:0.035
+INFO 2025-11-25 16:37:51 celerate.py:303 step:200 smpl:13K ep:79 epch:0.05 loss:0.950 grdn:2.031 lr:2.0e-05 updt_s:0.797 data_s:0.033
+INFO 2025-11-25 16:38:00 celerate.py:303 step:210 smpl:13K ep:83 epch:0.05 loss:0.903 grdn:1.960 lr:2.1e-05 updt_s:0.798 data_s:0.033
+INFO 2025-11-25 16:38:09 celerate.py:303 step:220 smpl:14K ep:87 epch:0.05 loss:0.854 grdn:2.069 lr:2.2e-05 updt_s:0.804 data_s:0.032
+INFO 2025-11-25 16:38:18 celerate.py:303 step:230 smpl:15K ep:91 epch:0.05 loss:0.832 grdn:2.082 lr:2.3e-05 updt_s:0.795 data_s:0.033
+INFO 2025-11-25 16:38:27 celerate.py:303 step:240 smpl:15K ep:95 epch:0.06 loss:0.766 grdn:2.192 lr:2.4e-05 updt_s:0.798 data_s:0.035
+INFO 2025-11-25 16:38:36 celerate.py:303 step:250 smpl:16K ep:99 epch:0.06 loss:0.762 grdn:1.971 lr:2.5e-05 updt_s:0.796 data_s:0.037
+INFO 2025-11-25 16:38:45 celerate.py:303 step:260 smpl:17K ep:103 epch:0.06 loss:0.762 grdn:2.012 lr:2.6e-05 updt_s:0.801 data_s:0.032
+INFO 2025-11-25 16:38:55 celerate.py:303 step:270 smpl:17K ep:107 epch:0.06 loss:0.744 grdn:1.818 lr:2.7e-05 updt_s:0.837 data_s:0.035
+INFO 2025-11-25 16:39:04 celerate.py:303 step:280 smpl:18K ep:111 epch:0.07 loss:0.731 grdn:1.822 lr:2.8e-05 updt_s:0.832 data_s:0.035
+INFO 2025-11-25 16:39:23 celerate.py:303 step:290 smpl:19K ep:115 epch:0.07 loss:0.672 grdn:1.998 lr:2.9e-05 updt_s:1.618 data_s:0.081
+INFO 2025-11-25 16:39:48 celerate.py:303 step:300 smpl:19K ep:119 epch:0.07 loss:0.663 grdn:2.073 lr:3.0e-05 updt_s:2.173 data_s:0.064
+INFO 2025-11-25 16:40:13 celerate.py:303 step:310 smpl:20K ep:123 epch:0.07 loss:0.636 grdn:1.682 lr:3.1e-05 updt_s:2.124 data_s:0.070
+INFO 2025-11-25 16:40:37 celerate.py:303 step:320 smpl:20K ep:127 epch:0.08 loss:0.619 grdn:1.908 lr:3.2e-05 updt_s:2.108 data_s:0.056
+INFO 2025-11-25 16:41:01 celerate.py:303 step:330 smpl:21K ep:131 epch:0.08 loss:0.608 grdn:2.209 lr:3.3e-05 updt_s:2.089 data_s:0.062
+INFO 2025-11-25 16:41:27 celerate.py:303 step:340 smpl:22K ep:135 epch:0.08 loss:0.586 grdn:1.878 lr:3.4e-05 updt_s:2.270 data_s:0.054
+INFO 2025-11-25 16:41:52 celerate.py:303 step:350 smpl:22K ep:139 epch:0.08 loss:0.566 grdn:2.130 lr:3.5e-05 updt_s:2.272 data_s:0.048
+INFO 2025-11-25 16:42:18 celerate.py:303 step:360 smpl:23K ep:143 epch:0.09 loss:0.550 grdn:1.780 lr:3.6e-05 updt_s:2.284 data_s:0.049
+INFO 2025-11-25 16:42:43 celerate.py:303 step:370 smpl:24K ep:147 epch:0.09 loss:0.544 grdn:2.035 lr:3.7e-05 updt_s:2.242 data_s:0.050
+INFO 2025-11-25 16:43:09 celerate.py:303 step:380 smpl:24K ep:151 epch:0.09 loss:0.498 grdn:1.997 lr:3.8e-05 updt_s:2.279 data_s:0.048
+INFO 2025-11-25 16:43:34 celerate.py:303 step:390 smpl:25K ep:155 epch:0.09 loss:0.509 grdn:1.968 lr:3.9e-05 updt_s:2.252 data_s:0.053
+INFO 2025-11-25 16:43:58 celerate.py:303 step:400 smpl:26K ep:159 epch:0.09 loss:0.515 grdn:1.930 lr:4.0e-05 updt_s:2.019 data_s:0.064
+INFO 2025-11-25 16:44:23 celerate.py:303 step:410 smpl:26K ep:163 epch:0.10 loss:0.496 grdn:1.964 lr:4.1e-05 updt_s:2.203 data_s:0.065
+INFO 2025-11-25 16:44:47 celerate.py:303 step:420 smpl:27K ep:167 epch:0.10 loss:0.507 grdn:2.167 lr:4.2e-05 updt_s:2.108 data_s:0.055
+INFO 2025-11-25 16:45:12 celerate.py:303 step:430 smpl:28K ep:171 epch:0.10 loss:0.476 grdn:2.264 lr:4.3e-05 updt_s:2.246 data_s:0.054
+INFO 2025-11-25 16:45:38 celerate.py:303 step:440 smpl:28K ep:175 epch:0.10 loss:0.483 grdn:2.109 lr:4.4e-05 updt_s:2.295 data_s:0.047
+INFO 2025-11-25 16:46:02 celerate.py:303 step:450 smpl:29K ep:179 epch:0.11 loss:0.463 grdn:2.102 lr:4.5e-05 updt_s:2.133 data_s:0.055
+INFO 2025-11-25 16:46:28 celerate.py:303 step:460 smpl:29K ep:183 epch:0.11 loss:0.474 grdn:2.084 lr:4.6e-05 updt_s:2.244 data_s:0.047
+INFO 2025-11-25 16:46:51 celerate.py:303 step:470 smpl:30K ep:187 epch:0.11 loss:0.469 grdn:2.300 lr:4.7e-05 updt_s:2.055 data_s:0.062
+INFO 2025-11-25 16:47:16 celerate.py:303 step:480 smpl:31K ep:191 epch:0.11 loss:0.452 grdn:1.881 lr:4.8e-05 updt_s:2.127 data_s:0.059
+INFO 2025-11-25 16:47:40 celerate.py:303 step:490 smpl:31K ep:195 epch:0.12 loss:0.429 grdn:1.763 lr:4.9e-05 updt_s:2.104 data_s:0.059
+INFO 2025-11-25 16:48:04 celerate.py:303 step:500 smpl:32K ep:199 epch:0.12 loss:0.443 grdn:1.952 lr:5.0e-05 updt_s:2.109 data_s:0.056
+INFO 2025-11-25 16:48:28 celerate.py:303 step:510 smpl:33K ep:203 epch:0.12 loss:0.447 grdn:2.047 lr:5.1e-05 updt_s:2.124 data_s:0.064
+INFO 2025-11-25 16:48:52 celerate.py:303 step:520 smpl:33K ep:207 epch:0.12 loss:0.432 grdn:1.912 lr:5.2e-05 updt_s:2.067 data_s:0.061
+INFO 2025-11-25 16:49:15 celerate.py:303 step:530 smpl:34K ep:211 epch:0.13 loss:0.431 grdn:2.212 lr:5.3e-05 updt_s:2.078 data_s:0.059
+INFO 2025-11-25 16:49:39 celerate.py:303 step:540 smpl:35K ep:215 epch:0.13 loss:0.431 grdn:2.405 lr:5.4e-05 updt_s:2.098 data_s:0.055
+INFO 2025-11-25 16:50:03 celerate.py:303 step:550 smpl:35K ep:219 epch:0.13 loss:0.411 grdn:2.581 lr:5.5e-05 updt_s:2.125 data_s:0.055
+INFO 2025-11-25 16:50:27 celerate.py:303 step:560 smpl:36K ep:223 epch:0.13 loss:0.393 grdn:2.113 lr:5.6e-05 updt_s:2.099 data_s:0.058
+INFO 2025-11-25 16:50:51 celerate.py:303 step:570 smpl:36K ep:227 epch:0.14 loss:0.422 grdn:1.959 lr:5.7e-05 updt_s:2.095 data_s:0.059
+INFO 2025-11-25 16:51:15 celerate.py:303 step:580 smpl:37K ep:230 epch:0.14 loss:0.396 grdn:2.173 lr:5.8e-05 updt_s:2.096 data_s:0.064
+INFO 2025-11-25 16:51:39 celerate.py:303 step:590 smpl:38K ep:234 epch:0.14 loss:0.405 grdn:2.552 lr:5.9e-05 updt_s:2.085 data_s:0.060
+INFO 2025-11-25 16:52:03 celerate.py:303 step:600 smpl:38K ep:238 epch:0.14 loss:0.391 grdn:2.288 lr:6.0e-05 updt_s:2.086 data_s:0.055
+INFO 2025-11-25 16:52:27 celerate.py:303 step:610 smpl:39K ep:242 epch:0.14 loss:0.404 grdn:2.486 lr:6.1e-05 updt_s:2.109 data_s:0.061
+INFO 2025-11-25 16:52:51 celerate.py:303 step:620 smpl:40K ep:246 epch:0.15 loss:0.405 grdn:2.145 lr:6.2e-05 updt_s:2.083 data_s:0.059
+INFO 2025-11-25 16:53:15 celerate.py:303 step:630 smpl:40K ep:250 epch:0.15 loss:0.391 grdn:1.959 lr:6.3e-05 updt_s:2.101 data_s:0.056
+INFO 2025-11-25 16:53:38 celerate.py:303 step:640 smpl:41K ep:254 epch:0.15 loss:0.420 grdn:2.396 lr:6.4e-05 updt_s:2.075 data_s:0.060
+INFO 2025-11-25 16:54:02 celerate.py:303 step:650 smpl:42K ep:258 epch:0.15 loss:0.375 grdn:2.170 lr:6.5e-05 updt_s:2.057 data_s:0.062
+INFO 2025-11-25 16:54:26 celerate.py:303 step:660 smpl:42K ep:262 epch:0.16 loss:0.400 grdn:2.426 lr:6.6e-05 updt_s:2.124 data_s:0.054
+INFO 2025-11-25 16:54:50 celerate.py:303 step:670 smpl:43K ep:266 epch:0.16 loss:0.380 grdn:2.022 lr:6.7e-05 updt_s:2.080 data_s:0.055

wandb/offline-run-20251125_163418-lizvmqey/files/requirements.txt

@@ -0,0 +1,286 @@
+PySocks==1.7.1
+beautifulsoup4==4.14.2
+gdown==5.2.0
+soupsieve==2.8
+Farama-Notifications==0.0.4
+PyOpenGL==3.1.0
+PySocks==1.7.1
+Send2Trash==1.8.3
+absl-py==2.3.1
+accelerate==1.11.0
+aiohappyeyeballs==2.6.1
+aiohttp==3.13.2
+aiosignal==1.4.0
+annotated-types==0.7.0
+antlr4-python3-runtime==4.8
+anyio==4.11.0
+argon2-cffi==25.1.0
+argon2-cffi-bindings==25.1.0
+arrow==1.4.0
+asciitree==0.3.3
+asttokens==3.0.0
+astunparse==1.6.3
+async-lru==2.0.5
+async-timeout==5.0.1
+attrs==25.4.0
+av==16.0.1
+babel==2.17.0
+bddl==3.6.0
+beautifulsoup4==4.14.2
+bleach==6.3.0
+blinker==1.9.0
+calvin_env==0.0.1
+certifi==2025.10.5
+cffi==2.0.0
+charset-normalizer==3.4.4
+click==8.3.0
+cloudpickle==3.1.1
+cmake==4.1.2
+colorlog==6.10.1
+comm==0.2.3
+contourpy==1.3.2
+cycler==0.12.1
+datasets==3.0.0
+debugpy==1.8.17
+decorator==4.4.2
+deepdiff==8.6.1
+defusedxml==0.7.1
+diffusers==0.35.2
+dill==0.3.8
+docopt==0.6.2
+docstring_parser==0.17.0
+draccus==0.10.0
+easydict==1.13
+einops==0.8.1
+etils==1.13.0
+evdev==1.9.2
+exceptiongroup==1.3.0
+executing==2.2.1
+fasteners==0.20
+fastjsonschema==2.21.2
+filelock==3.20.0
+Flask==3.1.2
+flatbuffers==25.9.23
+fonttools==4.60.1
+fqdn==1.5.1
+freetype-py==2.5.1
+frozenlist==1.8.0
+fsspec==2024.6.1
+future==1.0.0
+gast==0.6.0
+gdown==5.2.0
+gitdb==4.0.12
+GitPython==3.1.45
+glfw==2.10.0
+google-pasta==0.2.0
+grpcio==1.76.0
+gym==0.26.2
+gym-notices==0.1.0
+gymnasium==0.29.1
+h11==0.16.0
+h5py==3.15.1
+hf_transfer==0.1.9
+hf-xet==1.2.0
+httpcore==1.0.9
+httpx==0.28.1
+huggingface-hub==0.36.0
+hydra-colorlog==1.2.0
+hydra-core==1.1.1
+idna==3.11
+imageio==2.37.0
+imageio-ffmpeg==0.6.0
+importlib_metadata==8.7.0
+importlib_resources==6.5.2
+iniconfig==2.3.0
+inquirerpy==0.3.4
+ipykernel==7.1.0
+ipython==8.37.0
+iso8601==2.1.0
+isoduration==20.11.0
+itsdangerous==2.2.0
+jedi==0.19.2
+Jinja2==3.1.6
+joblib==1.5.2
+json5==0.12.1
+jsonlines==4.0.0
+jsonpointer==3.0.0
+jsonschema==4.25.1
+jsonschema-specifications==2025.9.1
+jupyter_client==8.6.3
+jupyter_core==5.9.1
+jupyter-events==0.12.0
+jupyter-lsp==2.3.0
+jupyter_server==2.17.0
+jupyter_server_terminals==0.5.3
+jupyterlab==4.4.10
+jupyterlab_pygments==0.3.0
+jupyterlab_server==2.28.0
+jupytext==1.18.1
+keras==3.12.0
+kiwisolver==1.4.9
+lark==1.3.1
+lerobot==0.1.0
+libclang==18.1.1
+lightning-utilities==0.15.2
+llvmlite==0.45.1
+lxml==6.0.2
+Markdown==3.10
+markdown-it-py==4.0.0
+MarkupSafe==3.0.3
+matplotlib==3.10.7
+matplotlib-inline==0.2.1
+mdit-py-plugins==0.5.0
+mdurl==0.1.2
+mergedeep==1.3.4
+mistune==3.1.4
+ml_dtypes==0.5.4
+moviepy==1.0.3
+mpmath==1.3.0
+mujoco==3.3.7
+multidict==6.7.0
+multiprocess==0.70.16
+mypy_extensions==1.1.0
+namex==0.1.0
+narwhals==2.12.0
+nbclient==0.10.2
+nbconvert==7.16.6
+nbformat==5.10.4
+nest-asyncio==1.6.0
+networkx==3.4.2
+nltk==3.9.2
+notebook_shim==0.2.4
+num2words==0.5.14
+numba==0.62.1
+numcodecs==0.13.1
+numpy==2.2.6
+numpy-quaternion==2024.0.12
+nvidia-cublas-cu12==12.8.4.1
+nvidia-cuda-cupti-cu12==12.8.90
+nvidia-cuda-nvrtc-cu12==12.8.93
+nvidia-cuda-runtime-cu12==12.8.90
+nvidia-cudnn-cu12==9.10.2.21
+nvidia-cufft-cu12==11.3.3.83
+nvidia-cufile-cu12==1.13.1.3
+nvidia-curand-cu12==10.3.9.90
+nvidia-cusolver-cu12==11.7.3.90
+nvidia-cusparse-cu12==12.5.8.93
+nvidia-cusparselt-cu12==0.7.1
+nvidia-ml-py==13.580.82
+nvidia-nccl-cu12==2.27.5
+nvidia-nvjitlink-cu12==12.8.93
+nvidia-nvshmem-cu12==3.3.20
+nvidia-nvtx-cu12==12.8.90
+nvitop==1.5.3
+omegaconf==2.1.2
+opencv-python==4.12.0.88
+opencv-python-headless==4.12.0.88
+opt_einsum==3.4.0
+optree==0.18.0
+orderly-set==5.5.0
+overrides==7.7.0
+packaging==25.0
+pandas==2.3.3
+pandocfilters==1.5.1
+parso==0.8.5
+pexpect==4.9.0
+pfzy==0.3.4
+pillow==12.0.0
+pip==25.2
+platformdirs==4.5.0
+plotly==6.5.0
+pluggy==1.6.0
+proglog==0.1.12
+prometheus_client==0.23.1
+prompt_toolkit==3.0.52
+propcache==0.4.1
+protobuf==6.33.0
+psutil==7.1.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+pyarrow==22.0.0
+pybullet==3.2.7
+pycollada==0.6
+pycparser==2.23
+pydantic==2.12.3
+pydantic_core==2.41.4
+pyglet==2.1.11
+Pygments==2.19.2
+pyhash==0.9.3
+pymunk==6.11.1
+pynput==1.8.1
+pyparsing==3.2.5
+pyrender==0.1.45
+pytest==8.4.2
+python-dateutil==2.9.0.post0
+python-dotenv==1.2.1
+python-json-logger==4.0.0
+python-xlib==0.33
+pytorch-lightning==2.5.6
+pytz==2025.2
+PyYAML==6.0.3
+pyyaml-include==1.4.1
+pyzmq==27.1.0
+referencing==0.37.0
+regex==2025.10.23
+requests==2.32.5
+rerun-sdk==0.26.2
+rfc3339-validator==0.1.4
+rfc3986-validator==0.1.1
+rfc3987-syntax==1.1.0
+rich==14.2.0
+robosuite==1.4.0
+rpds-py==0.28.0
+safetensors==0.6.2
+scipy==1.15.3
+sentry-sdk==2.43.0
+serial==0.0.97
+setuptools==57.5.0
+shtab==1.7.2
+six==1.17.0
+smmap==5.0.2
+sniffio==1.3.1
+soupsieve==2.8
+stack-data==0.6.3
+sympy==1.14.0
+tensorboard==2.20.0
+tensorboard-data-server==0.7.2
+tensorflow==2.20.0
+termcolor==3.2.0
+terminado==0.18.1
+tinycss2==1.4.0
+tokenizers==0.22.1
+toml==0.10.2
+tomli==2.3.0
+torch==2.9.0
+torchcodec==0.8.1
+torchmetrics==1.8.2
+torchvision==0.24.0
+tornado==6.5.2
+tqdm==4.67.1
+traitlets==5.14.3
+transformers==4.57.1
+trimesh==4.9.0
+triton==3.5.0
+typeguard==4.4.4
+typing_extensions==4.15.0
+typing-inspect==0.9.0
+typing-inspection==0.4.2
+tyro==0.9.35
+tzdata==2025.2
+urdfpy==0.0.22
+uri-template==1.3.0
+urllib3==2.5.0
+wandb==0.22.3
+wcwidth==0.2.14
+webcolors==25.10.0
+webencodings==0.5.1
+websocket-client==1.9.0
+Werkzeug==3.1.3
+wheel==0.45.1
+wrapt==2.0.1
+xxhash==3.6.0
+yarl==1.22.0
+zarr==2.18.3
+zipp==3.23.0
+Calvin==0.0.1
+tacto==0.0.3

wandb/offline-run-20251125_163418-lizvmqey/files/wandb-metadata.json

@@ -0,0 +1 @@
+{"os": "Linux-4.18.0-553.77.1.el8_10.x86_64-x86_64-with-glibc2.28", "python": "CPython 3.10.19", "started_at": "2025-11-25T15:34:18.399929Z", "args": ["--policy.path=/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/smol_pretrained/smolvla_base", "--dataset.root=/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot", "--output_dir=outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base", "--job_name=libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base", "--config_path=configs/smolvla_config/other_default.json", "--batch_size=64", "--steps=100000", "--policy.gradient_accumulation_steps=1", "--wandb.mode=offline", "--log_freq=10"], "program": "/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/Smolvla_capstone_project/lerobot/scripts/train_accelerate.py", "code_path": "lerobot/scripts/train_accelerate.py", "code_path_local": "lerobot/scripts/train_accelerate.py", "git": {"remote_url": "git@github-binh:jibby2803/Smolvla_capstone_project.git", "commit": "0f72ba315e381d6a8095782824a2d0c70faad469"}, "root": "outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base", "host": "ggpu192", "executable": "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/bin/python3.10", "cpu_count": 64, "cpu_count_logical": 128, "disk": {"/": {"total": "270465425408", "used": "7019204608"}}, "memory": {"total": "540930854912"}, "slurm": {"job_id": "11843069", "mpi_type": "pmi2"}, "writer_id": "tgzz60htaej5eivyg1ohkojicqbx6n35"}

wandb/offline-run-20251125_163418-lizvmqey/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_runtime": 1231.530468465, "train/lr": 6.558441558441558e-05, "train/update_s": 2.1237647216068583, "train/tlc_loss": -0.39544302225112915, "train/l1_infer_loss": 0.18404726684093475, "_step": 660, "train/episodes": 262.2805444616513, "train/loss": 0.6010105013847351, "train/losses_after_forward": {"bins": [8.37099278783171e-11, 0.6451714038848877, 1.2903428077697754, 1.935514211654663, 2.580685615539551, 3.2258570194244385, 3.871028423309326, 4.516200065612793, 5.161371231079102, 5.80654239654541, 6.451714038848877, 7.096885681152344, 7.742056846618652, 8.387228012084961, 9.032400131225586, 9.677571296691895, 10.322742462158203, 10.967913627624512, 11.61308479309082, 12.258256912231445, 12.903428077697754, 13.548599243164062, 14.193771362304688, 14.838942527770996, 15.484113693237305, 16.12928581237793, 16.774456024169922, 17.419628143310547, 18.064800262451172, 18.709970474243164, 19.35514259338379, 20.00031280517578, 20.645484924316406], "_type": "histogram", "values": [90542, 6746, 2272, 1089, 657, 364, 224, 140, 102, 55, 55, 31, 29, 24, 14, 6, 8, 7, 6, 3, 7, 4, 0, 0, 4, 2, 1, 1, 0, 1, 4, 2]}, "train/losses_after_rm_padding": {"_type": "histogram", "values": [17114, 2519, 986, 577, 387, 229, 159, 106, 74, 47, 52, 29, 27, 24, 14, 6, 8, 7, 6, 3, 7, 4, 0, 0, 4, 2, 1, 1, 0, 1, 4, 2], "bins": [2.2172486069393926e-09, 0.6451714038848877, 1.2903428077697754, 1.935514211654663, 2.580685615539551, 3.2258570194244385, 3.871028423309326, 4.516200065612793, 5.161371231079102, 5.80654239654541, 6.451714038848877, 7.096885681152344, 7.742056846618652, 8.387228012084961, 9.032400131225586, 9.677571296691895, 10.322742462158203, 10.967913627624512, 11.61308479309082, 12.258256912231445, 12.903428077697754, 13.548599243164062, 14.193771362304688, 14.838942527770996, 15.484113693237305, 16.12928581237793, 16.774456024169922, 17.419628143310547, 18.064800262451172, 18.709970474243164, 19.35514259338379, 20.00031280517578, 20.645484924316406]}, "_timestamp": 1764086066.625332, "train/epochs": 0.15649197163583015, "train/dataloading_s": 0.0544021604815498, "train/total_loss": 0.40328899025917053, "train/steps": 660, "train/samples": 42240, "train/grad_norm": 2.4256782650947573}

wandb/offline-run-20251125_163418-lizvmqey/logs/debug.log

@@ -0,0 +1,21 @@
+2025-11-25 16:34:18,424 INFO    MainThread:1842994 [wandb_setup.py:_flush():81] Current SDK version is 0.22.3
+2025-11-25 16:34:18,424 INFO    MainThread:1842994 [wandb_setup.py:_flush():81] Configure stats pid to 1842994
+2025-11-25 16:34:18,424 INFO    MainThread:1842994 [wandb_setup.py:_flush():81] Loading settings from /user/hominhduy.nguyen/u15271/.config/wandb/settings
+2025-11-25 16:34:18,424 INFO    MainThread:1842994 [wandb_setup.py:_flush():81] Loading settings from /mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/Smolvla_capstone_project/wandb/settings
+2025-11-25 16:34:18,424 INFO    MainThread:1842994 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2025-11-25 16:34:18,424 INFO    MainThread:1842994 [wandb_init.py:setup_run_log_directory():706] Logging user logs to outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base/wandb/offline-run-20251125_163418-lizvmqey/logs/debug.log
+2025-11-25 16:34:18,424 INFO    MainThread:1842994 [wandb_init.py:setup_run_log_directory():707] Logging internal logs to outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base/wandb/offline-run-20251125_163418-lizvmqey/logs/debug-internal.log
+2025-11-25 16:34:18,424 INFO    MainThread:1842994 [wandb_init.py:init():833] calling init triggers
+2025-11-25 16:34:18,425 INFO    MainThread:1842994 [wandb_init.py:init():838] wandb.init called with sweep_config: {}
+config: {'dataset': {'repo_id': '.', 'root': '/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot', 'episodes': None, 'image_transforms': {'enable': True, 'max_num_transforms': 3, 'random_order': False, 'image_tfs': {'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'crop_resize': {'weight': 1.0, 'type': 'RandomResizedCrop', 'kwargs': {'size': [256, 256], 'ratio': [1, 1], 'scale': [0.9, 0.95]}}, 'rotate': {'weight': 1.0, 'type': 'RandomRotate', 'kwargs': {'degrees': [-5, 5]}}}, 'wrist_tfs': {'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'vqa_data_path': None}, 'env': None, 'policy': {'type': 'smolvla', 'n_obs_steps': 1, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.MEAN_STD: 'MEAN_STD'>, 'ACTION': <NormalizationMode.MEAN_STD: 'MEAN_STD'>}, 'input_features': {'observation.images.image': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.wrist_image': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [8]}}, 'output_features': {'action': {'type': <FeatureType.ACTION: 'ACTION'>, 'shape': [7]}}, 'device': 'cuda', 'use_amp': False, 'gradient_accumulation_steps': 1, 'chunk_size': 50, 'n_action_steps': 50, 'max_state_dim': 32, 'max_action_dim': 32, 'resize_imgs_with_padding': [512, 512], 'empty_cameras': 0, 'adapt_to_pi_aloha': False, 'use_delta_joint_actions_aloha': False, 'tokenizer_max_length': 48, 'num_steps': 10, 'use_cache': True, 'freeze_vision_encoder': True, 'train_expert_only': False, 'train_state_proj': True, 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.95], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-10, 'optimizer_grad_clip_norm': 10.0, 'scheduler_warmup_steps': 1000, 'scheduler_decay_steps': 30000, 'scheduler_decay_lr': 2.5e-06, 'vlm_model_name': 'HuggingFaceTB/SmolVLM2-500M-Video-Instruct', 'load_vlm_weights': True, 'add_image_special_tokens': False, 'attention_mode': 'cross_attn', 'prefix_length': 0, 'pad_language_to': 'max_length', 'num_expert_layers': 0, 'num_vlm_layers': 16, 'self_attn_every_n_layers': 2, 'expert_width_multiplier': 0.75, 'min_period': 0.004, 'max_period': 4.0}, 'output_dir': 'outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base', 'job_name': 'libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base', 'resume': False, 'seed': 42, 'num_workers': 8, 'batch_size': 64, 'steps': 100000, 'eval_freq': 20000, 'log_freq': 10, 'save_checkpoint': True, 'save_freq': 10000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-10, 'grad_clip_norm': 10.0, 'betas': [0.9, 0.95], 'eps': 1e-08}, 'scheduler': {'type': 'cosine_decay_with_warmup', 'num_warmup_steps': 1000, 'num_decay_steps': 30000, 'peak_lr': 0.0001, 'decay_lr': 2.5e-06}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'smolvla', 'entity': 'Robotics_VLA', 'notes': None, 'run_id': None, 'mode': 'offline'}, '_wandb': {}}
+2025-11-25 16:34:18,425 INFO    MainThread:1842994 [wandb_init.py:init():881] starting backend
+2025-11-25 16:34:18,636 INFO    MainThread:1842994 [wandb_init.py:init():884] sending inform_init request
+2025-11-25 16:34:18,641 INFO    MainThread:1842994 [wandb_init.py:init():892] backend started and connected
+2025-11-25 16:34:18,643 INFO    MainThread:1842994 [wandb_init.py:init():962] updated telemetry
+2025-11-25 16:34:18,656 INFO    MainThread:1842994 [wandb_init.py:init():986] communicating run to backend with 90.0 second timeout
+2025-11-25 16:34:18,738 INFO    MainThread:1842994 [wandb_init.py:init():1033] starting run threads in backend
+2025-11-25 16:34:19,005 INFO    MainThread:1842994 [wandb_run.py:_console_start():2506] atexit reg
+2025-11-25 16:34:19,006 INFO    MainThread:1842994 [wandb_run.py:_redirect():2354] redirect: wrap_raw
+2025-11-25 16:34:19,006 INFO    MainThread:1842994 [wandb_run.py:_redirect():2423] Wrapping output streams.
+2025-11-25 16:34:19,006 INFO    MainThread:1842994 [wandb_run.py:_redirect():2446] Redirects installed.
+2025-11-25 16:34:19,008 INFO    MainThread:1842994 [wandb_init.py:init():1073] run started, returning control to user process

wandb/offline-run-20251125_163418-lizvmqey/run-lizvmqey.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26f389026dc187d8af9998f4ec6291a9e5b0d3156cb1c18338ae48b8e8b47cb4
+size 29392896

wandb/offline-run-20251127_112044-lizvmqey/files/config.yaml

@@ -0,0 +1,357 @@
+wandb_version: 1
+
+_wandb:
+  desc: null
+  value:
+    python_version: 3.10.19
+    cli_version: 0.22.3
+    framework: huggingface
+    is_jupyter_run: false
+    is_kaggle_kernel: false
+    start_time: 1764238845
+    t:
+      1:
+      - 1
+      - 41
+      - 49
+      - 51
+      - 71
+      2:
+      - 1
+      - 2
+      - 3
+      - 11
+      - 41
+      - 49
+      - 51
+      - 71
+      3:
+      - 4
+      - 13
+      - 14
+      - 15
+      - 16
+      - 37
+      - 42
+      - 61
+      4: 3.10.19
+      5: 0.22.3
+      13: linux-x86_64
+    e:
+      qnrpx9krjcsw5zeotwtgsw7vqa4lj8ft:
+        os: Linux-4.18.0-553.74.1.el8_10.x86_64-x86_64-with-glibc2.28
+        python: CPython 3.10.19
+        started_at: '2025-11-27T10:20:44.939230Z'
+        args:
+        - --resume=true
+        - --output_dir=/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+        - --config_path=/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base/checkpoints/last/pretrained_model/train_config.json
+        program: /mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/Smolvla_capstone_project/lerobot/scripts/train_accelerate.py
+        code_path: lerobot/scripts/train_accelerate.py
+        code_path_local: lerobot/scripts/train_accelerate.py
+        git:
+          remote_url: git@github-binh:jibby2803/Smolvla_capstone_project.git
+          commit: 5e71e3b01e75c173c975f3aab819927a265c360a
+        root: /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+        host: ggpu183
+        executable: /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/bin/python3.10
+        cpu_count: 64
+        cpu_count_logical: 128
+        disk:
+          /:
+            total: '270465433600'
+            used: '7152394240'
+        memory:
+          total: '540930867200'
+        slurm:
+          mpi_type: pmi2
+          job_id: '11851085'
+        writer_id: qnrpx9krjcsw5zeotwtgsw7vqa4lj8ft
+dataset:
+  desc: null
+  value:
+    repo_id: .
+    root: /mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot
+    episodes: null
+    image_transforms:
+      enable: true
+      max_num_transforms: 3
+      random_order: false
+      image_tfs:
+        hue:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            hue:
+            - -0.05
+            - 0.05
+        contrast:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            contrast:
+            - 0.8
+            - 1.2
+        sharpness:
+          weight: 1.0
+          type: SharpnessJitter
+          kwargs:
+            sharpness:
+            - 0.5
+            - 1.5
+        brightness:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            brightness:
+            - 0.8
+            - 1.2
+        saturation:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            saturation:
+            - 0.5
+            - 1.5
+        crop_resize:
+          weight: 1.0
+          type: RandomResizedCrop
+          kwargs:
+            size:
+            - 256
+            - 256
+            ratio:
+            - 1
+            - 1
+            scale:
+            - 0.9
+            - 0.95
+        rotate:
+          weight: 1.0
+          type: RandomRotate
+          kwargs:
+            degrees:
+            - -5
+            - 5
+      wrist_tfs:
+        hue:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            hue:
+            - -0.05
+            - 0.05
+        contrast:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            contrast:
+            - 0.8
+            - 1.2
+        sharpness:
+          weight: 1.0
+          type: SharpnessJitter
+          kwargs:
+            sharpness:
+            - 0.5
+            - 1.5
+        brightness:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            brightness:
+            - 0.8
+            - 1.2
+        saturation:
+          weight: 1.0
+          type: ColorJitter
+          kwargs:
+            saturation:
+            - 0.5
+            - 1.5
+    revision: null
+    use_imagenet_stats: true
+    video_backend: torchcodec
+    vqa_data_path: null
+env:
+  desc: null
+  value: null
+policy:
+  desc: null
+  value:
+    type: smolvla
+    n_obs_steps: 1
+    normalization_mapping:
+      VISUAL: IDENTITY
+      STATE: MEAN_STD
+      ACTION: MEAN_STD
+    input_features:
+      observation.images.image:
+        type: VISUAL
+        shape:
+        - 3
+        - 256
+        - 256
+      observation.images.wrist_image:
+        type: VISUAL
+        shape:
+        - 3
+        - 256
+        - 256
+      observation.images.image_mask:
+        type: VISUAL
+        shape:
+        - 3
+        - 256
+        - 256
+      observation.images.wrist_mask:
+        type: VISUAL
+        shape:
+        - 3
+        - 256
+        - 256
+      observation.images.object_of_interest_mask:
+        type: VISUAL
+        shape:
+        - 3
+        - 256
+        - 256
+      observation.images.object_of_interest_wrist_mask:
+        type: VISUAL
+        shape:
+        - 3
+        - 256
+        - 256
+      observation.state:
+        type: STATE
+        shape:
+        - 8
+      observation.states.ee_state:
+        type: STATE
+        shape:
+        - 6
+      observation.states.joint_state:
+        type: STATE
+        shape:
+        - 7
+      observation.states.gripper_state:
+        type: STATE
+        shape:
+        - 2
+    output_features:
+      action:
+        type: ACTION
+        shape:
+        - 7
+    device: cuda
+    use_amp: false
+    gradient_accumulation_steps: 1
+    chunk_size: 50
+    n_action_steps: 50
+    max_state_dim: 32
+    max_action_dim: 32
+    resize_imgs_with_padding:
+    - 512
+    - 512
+    empty_cameras: 0
+    adapt_to_pi_aloha: false
+    use_delta_joint_actions_aloha: false
+    tokenizer_max_length: 48
+    num_steps: 10
+    use_cache: true
+    freeze_vision_encoder: true
+    train_expert_only: false
+    train_state_proj: true
+    optimizer_lr: 0.0001
+    optimizer_betas:
+    - 0.9
+    - 0.95
+    optimizer_eps: 1.0e-08
+    optimizer_weight_decay: 1.0e-10
+    optimizer_grad_clip_norm: 10.0
+    scheduler_warmup_steps: 1000
+    scheduler_decay_steps: 30000
+    scheduler_decay_lr: 2.5e-06
+    vlm_model_name: HuggingFaceTB/SmolVLM2-500M-Video-Instruct
+    load_vlm_weights: true
+    add_image_special_tokens: false
+    attention_mode: cross_attn
+    prefix_length: 0
+    pad_language_to: max_length
+    num_expert_layers: 0
+    num_vlm_layers: 16
+    self_attn_every_n_layers: 2
+    expert_width_multiplier: 0.75
+    min_period: 0.004
+    max_period: 4.0
+output_dir:
+  desc: null
+  value: /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+job_name:
+  desc: null
+  value: libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+resume:
+  desc: null
+  value: true
+seed:
+  desc: null
+  value: 42
+num_workers:
+  desc: null
+  value: 8
+batch_size:
+  desc: null
+  value: 64
+steps:
+  desc: null
+  value: 100000
+eval_freq:
+  desc: null
+  value: 20000
+log_freq:
+  desc: null
+  value: 10
+save_checkpoint:
+  desc: null
+  value: true
+save_freq:
+  desc: null
+  value: 10000
+use_policy_training_preset:
+  desc: null
+  value: true
+optimizer:
+  desc: null
+  value:
+    type: adamw
+    lr: 0.0001
+    weight_decay: 1.0e-10
+    grad_clip_norm: 10.0
+    betas:
+    - 0.9
+    - 0.95
+    eps: 1.0e-08
+scheduler:
+  desc: null
+  value:
+    type: cosine_decay_with_warmup
+    num_warmup_steps: 1000
+    num_decay_steps: 30000
+    peak_lr: 0.0001
+    decay_lr: 2.5e-06
+eval:
+  desc: null
+  value:
+    n_episodes: 50
+    batch_size: 50
+    use_async_envs: false
+wandb:
+  desc: null
+  value:
+    enable: true
+    disable_artifact: true
+    project: smolvla
+    entity: Robotics_VLA
+    notes: null
+    run_id: null
+    mode: offline

wandb/offline-run-20251127_112044-lizvmqey/files/output.log

@@ -0,0 +1,82 @@
+INFO 2025-11-27 11:20:45 celerate.py:159 Creating dataset
+Resolving data files: 100%|█████████████████████████████████████████████████████████████████████████████████████| 1676/1676 [00:00<00:00, 22868.84it/s]
+Loading dataset shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████| 78/78 [00:00<00:00, 934.13it/s]
+INFO 2025-11-27 11:20:53 celerate.py:170 Creating policy
+`torch_dtype` is deprecated! Use `dtype` instead!
+INFO 2025-11-27 11:21:09 celerate.py:181 Creating optimizer and scheduler
+INFO 2025-11-27 11:21:13 celerate.py:221 Output dir: /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base
+INFO 2025-11-27 11:21:13 celerate.py:224 cfg.steps=100000 (100K)
+INFO 2025-11-27 11:21:13 celerate.py:225 dataset.num_frames=269918 (270K)
+INFO 2025-11-27 11:21:13 celerate.py:226 dataset.num_episodes=1676
+INFO 2025-11-27 11:21:13 celerate.py:227 num_learnable_params=317226912 (317M)
+INFO 2025-11-27 11:21:13 celerate.py:228 num_total_params=450968810 (451M)
+INFO 2025-11-27 11:21:13 celerate.py:229 Number of processes: 1
+INFO 2025-11-27 11:21:13 celerate.py:230 Device: cuda
+INFO 2025-11-27 11:21:13 celerate.py:231 Mixed precision: no
+INFO 2025-11-27 11:21:13 celerate.py:253 Start offline training on a fixed dataset
+INFO 2025-11-27 11:21:34 celerate.py:303 step:50K smpl:3M ep:20K epch:11.86 loss:0.019 grdn:1.143 lr:2.5e-06 updt_s:1.044 data_s:0.955
+load pretrained policy
+Loading  HuggingFaceTB/SmolVLM2-500M-Video-Instruct weights ...
+Load pretrained VLM successfully!
+Reducing the number of VLM layers to 16 ...
+=================================
+USING MODEL FROM BINH'S DIR
+=================================
+Loading weights from local directory
+INFO 2025-11-27 11:21:43 celerate.py:303 step:50K smpl:3M ep:20K epch:11.86 loss:0.019 grdn:1.148 lr:2.5e-06 updt_s:0.797 data_s:0.017
+INFO 2025-11-27 11:21:51 celerate.py:303 step:50K smpl:3M ep:20K epch:11.86 loss:0.031 grdn:1.072 lr:2.5e-06 updt_s:0.767 data_s:0.017
+INFO 2025-11-27 11:22:00 celerate.py:303 step:50K smpl:3M ep:20K epch:11.86 loss:0.013 grdn:1.122 lr:2.5e-06 updt_s:0.775 data_s:0.018
+INFO 2025-11-27 11:22:08 celerate.py:303 step:50K smpl:3M ep:20K epch:11.87 loss:0.050 grdn:1.097 lr:2.5e-06 updt_s:0.751 data_s:0.017
+INFO 2025-11-27 11:22:17 celerate.py:303 step:50K smpl:3M ep:20K epch:11.87 loss:0.022 grdn:1.153 lr:2.5e-06 updt_s:0.790 data_s:0.018
+INFO 2025-11-27 11:22:26 celerate.py:303 step:50K smpl:3M ep:20K epch:11.87 loss:0.018 grdn:1.203 lr:2.5e-06 updt_s:0.769 data_s:0.018
+INFO 2025-11-27 11:22:34 celerate.py:303 step:50K smpl:3M ep:20K epch:11.87 loss:0.013 grdn:1.118 lr:2.5e-06 updt_s:0.765 data_s:0.018
+INFO 2025-11-27 11:22:43 celerate.py:303 step:50K smpl:3M ep:20K epch:11.88 loss:0.034 grdn:1.147 lr:2.5e-06 updt_s:0.771 data_s:0.019
+INFO 2025-11-27 11:22:51 celerate.py:303 step:50K smpl:3M ep:20K epch:11.88 loss:0.011 grdn:1.151 lr:2.5e-06 updt_s:0.766 data_s:0.017
+INFO 2025-11-27 11:23:00 celerate.py:303 step:50K smpl:3M ep:20K epch:11.88 loss:0.012 grdn:1.114 lr:2.5e-06 updt_s:0.773 data_s:0.017
+INFO 2025-11-27 11:23:09 celerate.py:303 step:50K smpl:3M ep:20K epch:11.88 loss:0.014 grdn:1.178 lr:2.5e-06 updt_s:0.794 data_s:0.018
+INFO 2025-11-27 11:23:17 celerate.py:303 step:50K smpl:3M ep:20K epch:11.89 loss:0.006 grdn:1.168 lr:2.5e-06 updt_s:0.757 data_s:0.017
+INFO 2025-11-27 11:23:26 celerate.py:303 step:50K smpl:3M ep:20K epch:11.89 loss:0.019 grdn:1.130 lr:2.5e-06 updt_s:0.774 data_s:0.017
+INFO 2025-11-27 11:23:34 celerate.py:303 step:50K smpl:3M ep:20K epch:11.89 loss:0.029 grdn:1.146 lr:2.5e-06 updt_s:0.777 data_s:0.018
+INFO 2025-11-27 11:23:43 celerate.py:303 step:50K smpl:3M ep:20K epch:11.89 loss:0.015 grdn:1.111 lr:2.5e-06 updt_s:0.774 data_s:0.017
+INFO 2025-11-27 11:23:51 celerate.py:303 step:50K smpl:3M ep:20K epch:11.90 loss:0.022 grdn:1.143 lr:2.5e-06 updt_s:0.761 data_s:0.018
+INFO 2025-11-27 11:24:00 celerate.py:303 step:50K smpl:3M ep:20K epch:11.90 loss:0.010 grdn:1.103 lr:2.5e-06 updt_s:0.790 data_s:0.017
+INFO 2025-11-27 11:24:09 celerate.py:303 step:50K smpl:3M ep:20K epch:11.90 loss:0.015 grdn:1.215 lr:2.5e-06 updt_s:0.783 data_s:0.017
+INFO 2025-11-27 11:24:17 celerate.py:303 step:50K smpl:3M ep:20K epch:11.90 loss:0.025 grdn:1.143 lr:2.5e-06 updt_s:0.771 data_s:0.018
+INFO 2025-11-27 11:24:26 celerate.py:303 step:50K smpl:3M ep:20K epch:11.91 loss:0.043 grdn:1.144 lr:2.5e-06 updt_s:0.779 data_s:0.018
+INFO 2025-11-27 11:24:35 celerate.py:303 step:50K smpl:3M ep:20K epch:11.91 loss:0.024 grdn:1.174 lr:2.5e-06 updt_s:0.773 data_s:0.018
+INFO 2025-11-27 11:24:43 celerate.py:303 step:50K smpl:3M ep:20K epch:11.91 loss:0.034 grdn:1.111 lr:2.5e-06 updt_s:0.757 data_s:0.017
+INFO 2025-11-27 11:24:52 celerate.py:303 step:50K smpl:3M ep:20K epch:11.91 loss:0.023 grdn:1.146 lr:2.5e-06 updt_s:0.778 data_s:0.017
+INFO 2025-11-27 11:25:00 celerate.py:303 step:50K smpl:3M ep:20K epch:11.91 loss:0.023 grdn:1.154 lr:2.5e-06 updt_s:0.763 data_s:0.018
+INFO 2025-11-27 11:25:09 celerate.py:303 step:50K smpl:3M ep:20K epch:11.92 loss:0.016 grdn:1.153 lr:2.5e-06 updt_s:0.777 data_s:0.018
+INFO 2025-11-27 11:25:17 celerate.py:303 step:50K smpl:3M ep:20K epch:11.92 loss:0.013 grdn:1.104 lr:2.5e-06 updt_s:0.758 data_s:0.017
+INFO 2025-11-27 11:25:26 celerate.py:303 step:50K smpl:3M ep:20K epch:11.92 loss:0.015 grdn:1.104 lr:2.5e-06 updt_s:0.791 data_s:0.018
+INFO 2025-11-27 11:25:35 celerate.py:303 step:50K smpl:3M ep:20K epch:11.92 loss:0.012 grdn:1.149 lr:2.5e-06 updt_s:0.775 data_s:0.017
+INFO 2025-11-27 11:25:43 celerate.py:303 step:50K smpl:3M ep:20K epch:11.93 loss:0.018 grdn:1.167 lr:2.5e-06 updt_s:0.769 data_s:0.017
+INFO 2025-11-27 11:25:52 celerate.py:303 step:50K smpl:3M ep:20K epch:11.93 loss:0.014 grdn:1.193 lr:2.5e-06 updt_s:0.773 data_s:0.017
+INFO 2025-11-27 11:26:01 celerate.py:303 step:50K smpl:3M ep:20K epch:11.93 loss:0.018 grdn:1.124 lr:2.5e-06 updt_s:0.775 data_s:0.018
+INFO 2025-11-27 11:26:09 celerate.py:303 step:50K smpl:3M ep:20K epch:11.93 loss:0.019 grdn:1.170 lr:2.5e-06 updt_s:0.753 data_s:0.017
+INFO 2025-11-27 11:26:18 celerate.py:303 step:50K smpl:3M ep:20K epch:11.94 loss:0.018 grdn:1.123 lr:2.5e-06 updt_s:0.789 data_s:0.018
+INFO 2025-11-27 11:26:26 celerate.py:303 step:50K smpl:3M ep:20K epch:11.94 loss:0.029 grdn:1.158 lr:2.5e-06 updt_s:0.767 data_s:0.017
+INFO 2025-11-27 11:26:35 celerate.py:303 step:50K smpl:3M ep:20K epch:11.94 loss:0.067 grdn:1.110 lr:2.5e-06 updt_s:0.779 data_s:0.017
+INFO 2025-11-27 11:26:44 celerate.py:303 step:50K smpl:3M ep:20K epch:11.94 loss:0.014 grdn:1.082 lr:2.5e-06 updt_s:0.781 data_s:0.018
+INFO 2025-11-27 11:26:53 celerate.py:303 step:50K smpl:3M ep:20K epch:11.95 loss:0.036 grdn:1.096 lr:2.5e-06 updt_s:0.788 data_s:0.018
+INFO 2025-11-27 11:27:01 celerate.py:303 step:50K smpl:3M ep:20K epch:11.95 loss:0.038 grdn:1.111 lr:2.5e-06 updt_s:0.771 data_s:0.017
+INFO 2025-11-27 11:27:10 celerate.py:303 step:50K smpl:3M ep:20K epch:11.95 loss:0.015 grdn:1.139 lr:2.5e-06 updt_s:0.773 data_s:0.017
+INFO 2025-11-27 11:27:18 celerate.py:303 step:50K smpl:3M ep:20K epch:11.95 loss:0.020 grdn:1.145 lr:2.5e-06 updt_s:0.769 data_s:0.017
+INFO 2025-11-27 11:27:27 celerate.py:303 step:50K smpl:3M ep:20K epch:11.96 loss:0.014 grdn:1.095 lr:2.5e-06 updt_s:0.775 data_s:0.017
+INFO 2025-11-27 11:27:36 celerate.py:303 step:50K smpl:3M ep:20K epch:11.96 loss:0.019 grdn:1.125 lr:2.5e-06 updt_s:0.779 data_s:0.017
+INFO 2025-11-27 11:27:44 celerate.py:303 step:50K smpl:3M ep:20K epch:11.96 loss:0.040 grdn:1.125 lr:2.5e-06 updt_s:0.777 data_s:0.018
+INFO 2025-11-27 11:27:53 celerate.py:303 step:50K smpl:3M ep:20K epch:11.96 loss:0.059 grdn:1.197 lr:2.5e-06 updt_s:0.771 data_s:0.017
+INFO 2025-11-27 11:28:01 celerate.py:303 step:50K smpl:3M ep:20K epch:11.96 loss:0.046 grdn:1.131 lr:2.5e-06 updt_s:0.769 data_s:0.017
+INFO 2025-11-27 11:28:10 celerate.py:303 step:50K smpl:3M ep:20K epch:11.97 loss:0.016 grdn:1.126 lr:2.5e-06 updt_s:0.770 data_s:0.018
+INFO 2025-11-27 11:28:18 celerate.py:303 step:50K smpl:3M ep:20K epch:11.97 loss:0.016 grdn:1.101 lr:2.5e-06 updt_s:0.765 data_s:0.017
+INFO 2025-11-27 11:28:27 celerate.py:303 step:50K smpl:3M ep:20K epch:11.97 loss:0.024 grdn:1.148 lr:2.5e-06 updt_s:0.779 data_s:0.017
+INFO 2025-11-27 11:28:36 celerate.py:303 step:50K smpl:3M ep:20K epch:11.97 loss:0.030 grdn:1.122 lr:2.5e-06 updt_s:0.780 data_s:0.017
+INFO 2025-11-27 11:28:44 celerate.py:303 step:51K smpl:3M ep:20K epch:11.98 loss:0.020 grdn:1.217 lr:2.5e-06 updt_s:0.751 data_s:0.017
+INFO 2025-11-27 11:28:53 celerate.py:303 step:51K smpl:3M ep:20K epch:11.98 loss:0.014 grdn:1.147 lr:2.5e-06 updt_s:0.766 data_s:0.019
+INFO 2025-11-27 11:29:01 celerate.py:303 step:51K smpl:3M ep:20K epch:11.98 loss:0.037 grdn:1.148 lr:2.5e-06 updt_s:0.778 data_s:0.018
+INFO 2025-11-27 11:29:10 celerate.py:303 step:51K smpl:3M ep:20K epch:11.98 loss:0.012 grdn:1.184 lr:2.5e-06 updt_s:0.757 data_s:0.018
+INFO 2025-11-27 11:29:18 celerate.py:303 step:51K smpl:3M ep:20K epch:11.99 loss:0.017 grdn:1.115 lr:2.5e-06 updt_s:0.763 data_s:0.017
+INFO 2025-11-27 11:29:27 celerate.py:303 step:51K smpl:3M ep:20K epch:11.99 loss:0.019 grdn:1.155 lr:2.5e-06 updt_s:0.765 data_s:0.017
+INFO 2025-11-27 11:29:35 celerate.py:303 step:51K smpl:3M ep:20K epch:11.99 loss:0.019 grdn:1.188 lr:2.5e-06 updt_s:0.779 data_s:0.017
+INFO 2025-11-27 11:29:44 celerate.py:303 step:51K smpl:3M ep:20K epch:11.99 loss:0.022 grdn:1.104 lr:2.5e-06 updt_s:0.782 data_s:0.019

wandb/offline-run-20251127_112044-lizvmqey/files/requirements.txt

@@ -0,0 +1,286 @@
+PySocks==1.7.1
+beautifulsoup4==4.14.2
+gdown==5.2.0
+soupsieve==2.8
+Farama-Notifications==0.0.4
+PyOpenGL==3.1.0
+PySocks==1.7.1
+Send2Trash==1.8.3
+absl-py==2.3.1
+accelerate==1.11.0
+aiohappyeyeballs==2.6.1
+aiohttp==3.13.2
+aiosignal==1.4.0
+annotated-types==0.7.0
+antlr4-python3-runtime==4.8
+anyio==4.11.0
+argon2-cffi==25.1.0
+argon2-cffi-bindings==25.1.0
+arrow==1.4.0
+asciitree==0.3.3
+asttokens==3.0.0
+astunparse==1.6.3
+async-lru==2.0.5
+async-timeout==5.0.1
+attrs==25.4.0
+av==16.0.1
+babel==2.17.0
+bddl==3.6.0
+beautifulsoup4==4.14.2
+bleach==6.3.0
+blinker==1.9.0
+calvin_env==0.0.1
+certifi==2025.10.5
+cffi==2.0.0
+charset-normalizer==3.4.4
+click==8.3.0
+cloudpickle==3.1.1
+cmake==4.1.2
+colorlog==6.10.1
+comm==0.2.3
+contourpy==1.3.2
+cycler==0.12.1
+datasets==3.0.0
+debugpy==1.8.17
+decorator==4.4.2
+deepdiff==8.6.1
+defusedxml==0.7.1
+diffusers==0.35.2
+dill==0.3.8
+docopt==0.6.2
+docstring_parser==0.17.0
+draccus==0.10.0
+easydict==1.13
+einops==0.8.1
+etils==1.13.0
+evdev==1.9.2
+exceptiongroup==1.3.0
+executing==2.2.1
+fasteners==0.20
+fastjsonschema==2.21.2
+filelock==3.20.0
+Flask==3.1.2
+flatbuffers==25.9.23
+fonttools==4.60.1
+fqdn==1.5.1
+freetype-py==2.5.1
+frozenlist==1.8.0
+fsspec==2024.6.1
+future==1.0.0
+gast==0.6.0
+gdown==5.2.0
+gitdb==4.0.12
+GitPython==3.1.45
+glfw==2.10.0
+google-pasta==0.2.0
+grpcio==1.76.0
+gym==0.26.2
+gym-notices==0.1.0
+gymnasium==0.29.1
+h11==0.16.0
+h5py==3.15.1
+hf_transfer==0.1.9
+hf-xet==1.2.0
+httpcore==1.0.9
+httpx==0.28.1
+huggingface-hub==0.36.0
+hydra-colorlog==1.2.0
+hydra-core==1.1.1
+idna==3.11
+imageio==2.37.0
+imageio-ffmpeg==0.6.0
+importlib_metadata==8.7.0
+importlib_resources==6.5.2
+iniconfig==2.3.0
+inquirerpy==0.3.4
+ipykernel==7.1.0
+ipython==8.37.0
+iso8601==2.1.0
+isoduration==20.11.0
+itsdangerous==2.2.0
+jedi==0.19.2
+Jinja2==3.1.6
+joblib==1.5.2
+json5==0.12.1
+jsonlines==4.0.0
+jsonpointer==3.0.0
+jsonschema==4.25.1
+jsonschema-specifications==2025.9.1
+jupyter_client==8.6.3
+jupyter_core==5.9.1
+jupyter-events==0.12.0
+jupyter-lsp==2.3.0
+jupyter_server==2.17.0
+jupyter_server_terminals==0.5.3
+jupyterlab==4.4.10
+jupyterlab_pygments==0.3.0
+jupyterlab_server==2.28.0
+jupytext==1.18.1
+keras==3.12.0
+kiwisolver==1.4.9
+lark==1.3.1
+lerobot==0.1.0
+libclang==18.1.1
+lightning-utilities==0.15.2
+llvmlite==0.45.1
+lxml==6.0.2
+Markdown==3.10
+markdown-it-py==4.0.0
+MarkupSafe==3.0.3
+matplotlib==3.10.7
+matplotlib-inline==0.2.1
+mdit-py-plugins==0.5.0
+mdurl==0.1.2
+mergedeep==1.3.4
+mistune==3.1.4
+ml_dtypes==0.5.4
+moviepy==1.0.3
+mpmath==1.3.0
+mujoco==3.3.7
+multidict==6.7.0
+multiprocess==0.70.16
+mypy_extensions==1.1.0
+namex==0.1.0
+narwhals==2.12.0
+nbclient==0.10.2
+nbconvert==7.16.6
+nbformat==5.10.4
+nest-asyncio==1.6.0
+networkx==3.4.2
+nltk==3.9.2
+notebook_shim==0.2.4
+num2words==0.5.14
+numba==0.62.1
+numcodecs==0.13.1
+numpy==2.2.6
+numpy-quaternion==2024.0.12
+nvidia-cublas-cu12==12.8.4.1
+nvidia-cuda-cupti-cu12==12.8.90
+nvidia-cuda-nvrtc-cu12==12.8.93
+nvidia-cuda-runtime-cu12==12.8.90
+nvidia-cudnn-cu12==9.10.2.21
+nvidia-cufft-cu12==11.3.3.83
+nvidia-cufile-cu12==1.13.1.3
+nvidia-curand-cu12==10.3.9.90
+nvidia-cusolver-cu12==11.7.3.90
+nvidia-cusparse-cu12==12.5.8.93
+nvidia-cusparselt-cu12==0.7.1
+nvidia-ml-py==13.580.82
+nvidia-nccl-cu12==2.27.5
+nvidia-nvjitlink-cu12==12.8.93
+nvidia-nvshmem-cu12==3.3.20
+nvidia-nvtx-cu12==12.8.90
+nvitop==1.5.3
+omegaconf==2.1.2
+opencv-python==4.12.0.88
+opencv-python-headless==4.12.0.88
+opt_einsum==3.4.0
+optree==0.18.0
+orderly-set==5.5.0
+overrides==7.7.0
+packaging==25.0
+pandas==2.3.3
+pandocfilters==1.5.1
+parso==0.8.5
+pexpect==4.9.0
+pfzy==0.3.4
+pillow==12.0.0
+pip==25.2
+platformdirs==4.5.0
+plotly==6.5.0
+pluggy==1.6.0
+proglog==0.1.12
+prometheus_client==0.23.1
+prompt_toolkit==3.0.52
+propcache==0.4.1
+protobuf==6.33.0
+psutil==7.1.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+pyarrow==22.0.0
+pybullet==3.2.7
+pycollada==0.6
+pycparser==2.23
+pydantic==2.12.3
+pydantic_core==2.41.4
+pyglet==2.1.11
+Pygments==2.19.2
+pyhash==0.9.3
+pymunk==6.11.1
+pynput==1.8.1
+pyparsing==3.2.5
+pyrender==0.1.45
+pytest==8.4.2
+python-dateutil==2.9.0.post0
+python-dotenv==1.2.1
+python-json-logger==4.0.0
+python-xlib==0.33
+pytorch-lightning==2.5.6
+pytz==2025.2
+PyYAML==6.0.3
+pyyaml-include==1.4.1
+pyzmq==27.1.0
+referencing==0.37.0
+regex==2025.10.23
+requests==2.32.5
+rerun-sdk==0.26.2
+rfc3339-validator==0.1.4
+rfc3986-validator==0.1.1
+rfc3987-syntax==1.1.0
+rich==14.2.0
+robosuite==1.4.0
+rpds-py==0.28.0
+safetensors==0.6.2
+scipy==1.15.3
+sentry-sdk==2.43.0
+serial==0.0.97
+setuptools==57.5.0
+shtab==1.7.2
+six==1.17.0
+smmap==5.0.2
+sniffio==1.3.1
+soupsieve==2.8
+stack-data==0.6.3
+sympy==1.14.0
+tensorboard==2.20.0
+tensorboard-data-server==0.7.2
+tensorflow==2.20.0
+termcolor==3.2.0
+terminado==0.18.1
+tinycss2==1.4.0
+tokenizers==0.22.1
+toml==0.10.2
+tomli==2.3.0
+torch==2.9.0
+torchcodec==0.8.1
+torchmetrics==1.8.2
+torchvision==0.24.0
+tornado==6.5.2
+tqdm==4.67.1
+traitlets==5.14.3
+transformers==4.57.1
+trimesh==4.9.0
+triton==3.5.0
+typeguard==4.4.4
+typing_extensions==4.15.0
+typing-inspect==0.9.0
+typing-inspection==0.4.2
+tyro==0.9.35
+tzdata==2025.2
+urdfpy==0.0.22
+uri-template==1.3.0
+urllib3==2.5.0
+wandb==0.22.3
+wcwidth==0.2.14
+webcolors==25.10.0
+webencodings==0.5.1
+websocket-client==1.9.0
+Werkzeug==3.1.3
+wheel==0.45.1
+wrapt==2.0.1
+xxhash==3.6.0
+yarl==1.22.0
+zarr==2.18.3
+zipp==3.23.0
+Calvin==0.0.1
+tacto==0.0.3

wandb/offline-run-20251127_112044-lizvmqey/files/wandb-metadata.json

@@ -0,0 +1 @@
+{"os": "Linux-4.18.0-553.74.1.el8_10.x86_64-x86_64-with-glibc2.28", "python": "CPython 3.10.19", "started_at": "2025-11-27T10:20:44.939230Z", "args": ["--resume=true", "--output_dir=/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base", "--config_path=/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base/checkpoints/last/pretrained_model/train_config.json"], "program": "/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/Smolvla_capstone_project/lerobot/scripts/train_accelerate.py", "code_path": "lerobot/scripts/train_accelerate.py", "code_path_local": "lerobot/scripts/train_accelerate.py", "git": {"remote_url": "git@github-binh:jibby2803/Smolvla_capstone_project.git", "commit": "5e71e3b01e75c173c975f3aab819927a265c360a"}, "root": "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base", "host": "ggpu183", "executable": "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/bin/python3.10", "cpu_count": 64, "cpu_count_logical": 128, "disk": {"/": {"total": "270465433600", "used": "7152394240"}}, "memory": {"total": "540930867200"}, "slurm": {"mpi_type": "pmi2", "job_id": "11851085"}, "writer_id": "qnrpx9krjcsw5zeotwtgsw7vqa4lj8ft"}

wandb/offline-run-20251127_112044-lizvmqey/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_runtime": 539.236888844, "_timestamp": 1764239375.9905393, "train/episodes": 20096.25323246319, "train/epochs": 11.990604553975652, "train/grad_norm": 1.1879278302192688, "train/losses_after_forward": {"values": [73326, 13844, 6634, 3639, 2015, 1204, 666, 393, 245, 167, 107, 69, 31, 23, 9, 9, 2, 6, 5, 1, 3, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1], "bins": [7.72937269744034e-13, 0.9368996024131775, 1.873799204826355, 2.8106987476348877, 3.74759840965271, 4.684497833251953, 5.621397495269775, 6.558297157287598, 7.49519681930542, 8.432096481323242, 9.368995666503906, 10.305895805358887, 11.24279499053955, 12.179695129394531, 13.116594314575195, 14.053494453430176, 14.99039363861084, 15.927292823791504, 16.864192962646484, 17.80109214782715, 18.737991333007812, 19.67489242553711, 20.611791610717773, 21.548690795898438, 22.4855899810791, 23.422489166259766, 24.359390258789062, 25.296289443969727, 26.23318862915039, 27.170087814331055, 28.10698890686035, 29.043888092041016, 29.98078727722168], "_type": "histogram"}, "train/loss": 0.20997808873653412, "train/lr": 2.5e-06, "train/samples": 3236480, "train/update_s": 0.779058738425374, "train/tlc_loss": -0.39999255537986755, "train/total_loss": 0.009981811046600342, "train/l1_infer_loss": 0.12626828253269196, "_step": 50570, "train/steps": 50570, "train/dataloading_s": 0.017474299389868976, "train/losses_after_rm_padding": {"values": [19399, 1649, 567, 290, 166, 108, 66, 39, 31, 25, 12, 12, 10, 1, 3, 2, 2, 3, 0, 5, 2, 2, 1, 0, 0, 2, 2, 0, 0, 0, 0, 1], "bins": [5.912070832891914e-11, 0.40790247917175293, 0.8158049583435059, 1.2237074375152588, 1.6316099166870117, 2.0395123958587646, 2.4474148750305176, 2.8553173542022705, 3.2632198333740234, 3.6711223125457764, 4.079024791717529, 4.486927032470703, 4.894829750061035, 5.302732467651367, 5.710634708404541, 6.118536949157715, 6.526439666748047, 6.934342384338379, 7.342244625091553, 7.750146865844727, 8.158049583435059, 8.56595230102539, 8.973854064941406, 9.381756782531738, 9.78965950012207, 10.197562217712402, 10.605464935302734, 11.01336669921875, 11.421269416809082, 11.829172134399414, 12.23707389831543, 12.644976615905762, 13.052879333496094], "_type": "histogram"}}

wandb/offline-run-20251127_112044-lizvmqey/logs/debug-internal.log

@@ -0,0 +1,13 @@
+{"time":"2025-11-27T11:20:45.378937954+01:00","level":"INFO","msg":"stream: starting","core version":"0.22.3"}
+{"time":"2025-11-27T11:20:45.409919704+01:00","level":"ERROR","msg":"monitor: failed to initialize GPU resource: monitor: could not create portfile"}
+{"time":"2025-11-27T11:20:45.425921086+01:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2025-11-27T11:20:45.425965495+01:00","level":"INFO","msg":"stream: created new stream","id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.425988095+01:00","level":"INFO","msg":"handler: started","stream_id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.428918285+01:00","level":"INFO","msg":"stream: started","id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.428917715+01:00","level":"INFO","msg":"sender: started","stream_id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.428956845+01:00","level":"INFO","msg":"writer: started","stream_id":"lizvmqey"}
+{"time":"2025-11-27T11:20:45.429853649+01:00","level":"WARN","msg":"runupserter: server does not expand metric globs but the x_server_side_expand_glob_metrics setting is set; ignoring"}
+{"time":"2025-11-27T14:51:55.122025249+01:00","level":"INFO","msg":"stream: closing","id":"lizvmqey"}
+{"time":"2025-11-27T14:51:55.125975233+01:00","level":"INFO","msg":"handler: closed","stream_id":"lizvmqey"}
+{"time":"2025-11-27T14:51:55.127619953+01:00","level":"INFO","msg":"sender: closed","stream_id":"lizvmqey"}
+{"time":"2025-11-27T14:51:55.127656972+01:00","level":"INFO","msg":"stream: closed","id":"lizvmqey"}

wandb/offline-run-20251127_112044-lizvmqey/logs/debug.log

@@ -0,0 +1,49 @@
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Current SDK version is 0.22.3
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Configure stats pid to 1938939
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Loading settings from /user/hominhduy.nguyen/u15271/.config/wandb/settings
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Loading settings from /mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/binh/Smolvla_capstone_project/wandb/settings
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:setup_run_log_directory():706] Logging user logs to /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base/wandb/offline-run-20251127_112044-lizvmqey/logs/debug.log
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:setup_run_log_directory():707] Logging internal logs to /projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base/wandb/offline-run-20251127_112044-lizvmqey/logs/debug-internal.log
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:init():833] calling init triggers
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:init():838] wandb.init called with sweep_config: {}
+config: {'dataset': {'repo_id': '.', 'root': '/mnt/vast-kisski/projects/kisski-umg-fairpact-2/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot', 'episodes': None, 'image_transforms': {'enable': True, 'max_num_transforms': 3, 'random_order': False, 'image_tfs': {'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'crop_resize': {'weight': 1.0, 'type': 'RandomResizedCrop', 'kwargs': {'size': [256, 256], 'ratio': [1, 1], 'scale': [0.9, 0.95]}}, 'rotate': {'weight': 1.0, 'type': 'RandomRotate', 'kwargs': {'degrees': [-5, 5]}}}, 'wrist_tfs': {'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'vqa_data_path': None}, 'env': None, 'policy': {'type': 'smolvla', 'n_obs_steps': 1, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.MEAN_STD: 'MEAN_STD'>, 'ACTION': <NormalizationMode.MEAN_STD: 'MEAN_STD'>}, 'input_features': {'observation.images.image': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.wrist_image': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.image_mask': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.wrist_mask': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.object_of_interest_mask': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.images.object_of_interest_wrist_mask': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 256, 256]}, 'observation.state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [8]}, 'observation.states.ee_state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [6]}, 'observation.states.joint_state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [7]}, 'observation.states.gripper_state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [2]}}, 'output_features': {'action': {'type': <FeatureType.ACTION: 'ACTION'>, 'shape': [7]}}, 'device': 'cuda', 'use_amp': False, 'gradient_accumulation_steps': 1, 'chunk_size': 50, 'n_action_steps': 50, 'max_state_dim': 32, 'max_action_dim': 32, 'resize_imgs_with_padding': [512, 512], 'empty_cameras': 0, 'adapt_to_pi_aloha': False, 'use_delta_joint_actions_aloha': False, 'tokenizer_max_length': 48, 'num_steps': 10, 'use_cache': True, 'freeze_vision_encoder': True, 'train_expert_only': False, 'train_state_proj': True, 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.95], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-10, 'optimizer_grad_clip_norm': 10.0, 'scheduler_warmup_steps': 1000, 'scheduler_decay_steps': 30000, 'scheduler_decay_lr': 2.5e-06, 'vlm_model_name': 'HuggingFaceTB/SmolVLM2-500M-Video-Instruct', 'load_vlm_weights': True, 'add_image_special_tokens': False, 'attention_mode': 'cross_attn', 'prefix_length': 0, 'pad_language_to': 'max_length', 'num_expert_layers': 0, 'num_vlm_layers': 16, 'self_attn_every_n_layers': 2, 'expert_width_multiplier': 0.75, 'min_period': 0.004, 'max_period': 4.0}, 'output_dir': '/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/VLA/binh/Smolvla_capstone_project/outputs/train/2025-11-25/16-33-59_libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base', 'job_name': 'libero_100%_tlc_v4_process_mask_neighbor_neg_add_linear_different_0.4tlc_base', 'resume': True, 'seed': 42, 'num_workers': 8, 'batch_size': 64, 'steps': 100000, 'eval_freq': 20000, 'log_freq': 10, 'save_checkpoint': True, 'save_freq': 10000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-10, 'grad_clip_norm': 10.0, 'betas': [0.9, 0.95], 'eps': 1e-08}, 'scheduler': {'type': 'cosine_decay_with_warmup', 'num_warmup_steps': 1000, 'num_decay_steps': 30000, 'peak_lr': 0.0001, 'decay_lr': 2.5e-06}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'smolvla', 'entity': 'Robotics_VLA', 'notes': None, 'run_id': None, 'mode': 'offline'}, '_wandb': {}}
+2025-11-27 11:20:44,979 INFO    MainThread:1938939 [wandb_init.py:init():881] starting backend
+2025-11-27 11:20:45,217 INFO    MainThread:1938939 [wandb_init.py:init():884] sending inform_init request
+2025-11-27 11:20:45,227 INFO    MainThread:1938939 [wandb_init.py:init():892] backend started and connected
+2025-11-27 11:20:45,229 INFO    MainThread:1938939 [wandb_init.py:init():962] updated telemetry
+2025-11-27 11:20:45,248 INFO    MainThread:1938939 [wandb_init.py:init():986] communicating run to backend with 90.0 second timeout
+2025-11-27 11:20:45,432 INFO    MainThread:1938939 [wandb_init.py:init():1033] starting run threads in backend
+2025-11-27 11:20:45,788 INFO    MainThread:1938939 [wandb_run.py:_console_start():2506] atexit reg
+2025-11-27 11:20:45,788 INFO    MainThread:1938939 [wandb_run.py:_redirect():2354] redirect: wrap_raw
+2025-11-27 11:20:45,789 INFO    MainThread:1938939 [wandb_run.py:_redirect():2423] Wrapping output streams.
+2025-11-27 11:20:45,789 INFO    MainThread:1938939 [wandb_run.py:_redirect():2446] Redirects installed.
+2025-11-27 11:20:45,800 INFO    MainThread:1938939 [wandb_init.py:init():1073] run started, returning control to user process
+2025-11-27 14:51:55,121 INFO    wandb-AsyncioManager-main:1938939 [service_client.py:_forward_responses():80] Reached EOF.
+2025-11-27 14:51:55,122 INFO    wandb-AsyncioManager-main:1938939 [mailbox.py:close():137] Closing mailbox, abandoning 0 handles.
+2025-11-27 14:51:55,307 ERROR   wandb-AsyncioManager-main:1938939 [asyncio_manager.py:fn_wrap_exceptions():183] Uncaught exception in run_soon callback.
+Traceback (most recent call last):
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/asyncio_manager.py", line 181, in fn_wrap_exceptions
+    await fn()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/service/service_client.py", line 38, in publish
+    await self._send_server_request(request)
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/service/service_client.py", line 64, in _send_server_request
+    await self._writer.drain()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/asyncio/streams.py", line 371, in drain
+    await self._protocol._drain_helper()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/asyncio/streams.py", line 167, in _drain_helper
+    raise ConnectionResetError('Connection lost')
+ConnectionResetError: Connection lost
+2025-11-27 14:51:55,322 ERROR   wandb-AsyncioManager-main:1938939 [asyncio_manager.py:fn_wrap_exceptions():183] Uncaught exception in run_soon callback.
+Traceback (most recent call last):
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/asyncio_manager.py", line 181, in fn_wrap_exceptions
+    await fn()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/service/service_client.py", line 38, in publish
+    await self._send_server_request(request)
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/site-packages/wandb/sdk/lib/service/service_client.py", line 64, in _send_server_request
+    await self._writer.drain()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/asyncio/streams.py", line 371, in drain
+    await self._protocol._drain_helper()
+  File "/projects/extern/kisski/kisski-umg-fairpact-2/dir.project/miniconda3/envs/smol/lib/python3.10/asyncio/streams.py", line 167, in _drain_helper
+    raise ConnectionResetError('Connection lost')
+ConnectionResetError: Connection lost

wandb/offline-run-20251127_112044-lizvmqey/run-lizvmqey.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db437b68b8edb0942bdd080811711afe74a368e7dacc00559976a204e7990ea6
+size 7305887