Upload folder using huggingface_hub

.gitattributes

@@ -55,3 +55,4 @@ fastumi_pickandplace_qwenDiscreteDiffusion_dynamic_1/wandb/wandb/run-20260328_02
 fastumi_pickandplace_qwenDiscreteDiffusion_dynamic_1/wandb/wandb/run-20260328_201042-ccztk042/run-ccztk042.wandb filter=lfs diff=lfs merge=lfs -text
 fastumi_pickandplace_qwenDiscreteDiffusion_329v2/wandb/wandb/run-20260330_085553-rk4vla0z/run-rk4vla0z.wandb filter=lfs diff=lfs merge=lfs -text
 wandb/wandb/run-20260330_213657-3rc55aw0/run-3rc55aw0.wandb filter=lfs diff=lfs merge=lfs -text
+fastumi_pickandplace_qwenPI_329v2/wandb/wandb/run-20260330_213657-3rc55aw0/run-3rc55aw0.wandb filter=lfs diff=lfs merge=lfs -text

fastumi_pickandplace_qwenPI_329v2/checkpoints/steps_10000_pytorch_model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a3e61570b65877a47b6c70e9ec07d6176ccb3f8fe939bbd443250eb0eb18ee3
+size 12444522384

fastumi_pickandplace_qwenPI_329v2/config.yaml

@@ -0,0 +1,70 @@
+datasets:
+  vla_data:
+    CoT_prompt: Your task is {instruction}. To identify the key objects for your task.
+      Locate their bounding boxes in [x1,y1,x2,y2] format.
+    data_mix: dynamic-329-v2
+    data_root_dir: playground/Datasets/FastUMI
+    dataset_py: lerobot_datasets
+    per_device_batch_size: 8
+    video_backend: torchvision_av
+framework:
+  action_model:
+    action_dim: 10
+    add_pos_embed: true
+    diffusion_model_cfg:
+      attention_head_dim: 64
+      cross_attention_dim: 2048
+      dropout: 0.2
+      final_dropout: true
+      input_embedding_dim: 2048
+      interleave_self_attention: true
+      norm_type: ada_norm
+      num_attention_heads: 32
+      num_layers: 36
+      output_dim: 1024
+      positional_embeddings: null
+    future_action_window_size: 15
+    max_seq_len: 1024
+    noise_beta_alpha: 1.5
+    noise_beta_beta: 1.0
+    noise_s: 0.999
+    num_inference_timesteps: 4
+    num_target_vision_tokens: 32
+    num_timestep_buckets: 1000
+    past_action_window_size: 0
+    state_dim: 10
+  name: QwenPI
+  qwenvl:
+    attn_implementation: flash_attention_2
+    base_vlm: playground/Pretrained_models/Qwen2.5-VL-3B-Instruct-Action
+    num_vl_layers: 36
+    vl_hidden_dim: 2048
+output_dir: ./results/Checkpoints/fastumi_pickandplace_qwenPI_329v2
+run_id: fastumi_pickandplace_qwenPI_329v2
+run_root_dir: ./results/Checkpoints
+seed: 42
+trainer:
+  eval_interval: 100
+  freeze_modules: null
+  gradient_accumulation_steps: 1
+  gradient_clipping: 1.0
+  is_resume: true
+  learning_rate:
+    action_model: 0.0001
+    base: 2.5e-05
+    qwen_vl_interface: 1.0e-05
+  logging_frequency: 50
+  lr_scheduler_type: cosine_with_min_lr
+  max_train_steps: 50000
+  num_warmup_steps: 5000
+  optimizer:
+    betas:
+    - 0.9
+    - 0.95
+    eps: 1.0e-08
+    weight_decay: 1.0e-08
+  save_interval: 10000
+  scheduler_specific_kwargs:
+    min_lr: 1.0e-06
+wandb_entity: 2200011093-peking-university
+wandb_project: starVLA_FastUMI_dynamic_1

fastumi_pickandplace_qwenPI_329v2/dataset_statistics.json

@@ -0,0 +1,178 @@
+{
+  "new_embodiment": {
+    "action": {
+      "mean": [
+        0.0004015314334537834,
+        0.0014654810074716806,
+        8.946667367126793e-05,
+        0.9829065203666687,
+        -0.0013762820744886994,
+        0.011548871174454689,
+        0.0010323530295863748,
+        0.975115954875946,
+        -0.0010711257345974445,
+        0.5120490193367004
+      ],
+      "std": [
+        0.0025072325952351093,
+        0.0029722319450229406,
+        0.0030486329924315214,
+        0.1419316679239273,
+        0.06773792952299118,
+        0.09538158029317856,
+        0.06654245406389236,
+        0.20641110837459564,
+        0.04703392833471298,
+        0.4998091757297516
+      ],
+      "max": [
+        0.011190327815711498,
+        0.013076115399599075,
+        0.010677814483642578,
+        1.0,
+        0.672959566116333,
+        0.9145655632019043,
+        0.6909607648849487,
+        1.0000001192092896,
+        0.5652030110359192,
+        1.0
+      ],
+      "min": [
+        -0.010552742518484592,
+        -0.008821047842502594,
+        -0.01142233144491911,
+        -0.3473202884197235,
+        -0.6919457912445068,
+        -0.04520135000348091,
+        -0.6662734150886536,
+        -0.7587732672691345,
+        -0.5879436135292053,
+        0.0
+      ],
+      "q01": [
+        -0.005765579426661133,
+        -0.0058892327453941105,
+        -0.007874095924198627,
+        -0.15560666352510452,
+        -0.02744346622377634,
+        -0.015569249019026756,
+        -0.02219169806689024,
+        -0.7325680673122406,
+        -0.01868075806647539,
+        0.0
+      ],
+      "q99": [
+        0.006906798807904126,
+        0.008052000515162947,
+        0.007142445724457513,
+        0.9999997019767761,
+        0.022224539108574543,
+        0.7821587586402898,
+        0.027300313375890294,
+        0.9999997615814209,
+        0.014960514791309847,
+        1.0
+      ],
+      "mask": [
+        true,
+        true,
+        true,
+        true,
+        true,
+        true,
+        true,
+        true,
+        true,
+        false
+      ],
+      "norm_modes": [
+        "min_max",
+        "min_max",
+        "min_max",
+        "none",
+        "none",
+        "none",
+        "none",
+        "none",
+        "none",
+        "binary"
+      ]
+    },
+    "state": {
+      "mean": [
+        0.27271032333374023,
+        -0.446323037147522,
+        0.1041000559926033,
+        -0.5883525013923645,
+        0.3322174847126007,
+        -0.6678782105445862,
+        0.5890024304389954,
+        0.29855450987815857,
+        -0.3677307963371277,
+        0.5119883418083191
+      ],
+      "std": [
+        0.043663933873176575,
+        0.0719655305147171,
+        0.06168781593441963,
+        0.05306101217865931,
+        0.3058156669139862,
+        0.03284266218543032,
+        0.5444996356964111,
+        0.08082824945449829,
+        0.35454151034355164,
+        0.49981093406677246
+      ],
+      "max": [
+        0.4219360053539276,
+        -0.17201299965381622,
+        0.33353498578071594,
+        -0.34213387966156006,
+        0.6040946245193481,
+        -0.555172324180603,
+        0.9302739500999451,
+        0.5118387937545776,
+        0.6464942693710327,
+        1.0
+      ],
+      "min": [
+        0.12113799899816513,
+        -0.5658149719238281,
+        0.011877999641001225,
+        -0.7391995191574097,
+        -0.5488103032112122,
+        -0.7786615490913391,
+        -0.8354141712188721,
+        -0.060601431876420975,
+        -0.6885839700698853,
+        0.0
+      ],
+      "q01": [
+        0.15309083521366118,
+        -0.5302998530864715,
+        0.01954999938607216,
+        -0.6776320767402649,
+        -0.4965719360113144,
+        -0.7470758426189422,
+        -0.8051938533782959,
+        0.05889143392443657,
+        -0.6143978321552277,
+        0.0
+      ],
+      "q99": [
+        0.38105886042118076,
+        -0.27944067239761344,
+        0.27689720153808595,
+        -0.43391250193119046,
+        0.5646263039112092,
+        -0.5932964587211609,
+        0.8873281657695771,
+        0.4357714289426805,
+        0.5737681102752686,
+        1.0
+      ]
+    },
+    "num_transitions": 65939,
+    "num_trajectories": 525
+  }
+}

fastumi_pickandplace_qwenPI_329v2/summary.jsonl

@@ -0,0 +1 @@
+{"steps": 10000}

fastumi_pickandplace_qwenPI_329v2/wandb/wandb/debug-internal.log

@@ -0,0 +1,6 @@
+{"time":"2026-03-30T21:36:57.590450795Z","level":"INFO","msg":"stream: starting","core version":"0.25.0"}
+{"time":"2026-03-30T21:36:57.871298035Z","level":"INFO","msg":"stream: created new stream","id":"3rc55aw0"}
+{"time":"2026-03-30T21:36:57.871724848Z","level":"INFO","msg":"handler: started","stream_id":"3rc55aw0"}
+{"time":"2026-03-30T21:36:57.871808977Z","level":"INFO","msg":"stream: started","id":"3rc55aw0"}
+{"time":"2026-03-30T21:36:57.871845186Z","level":"INFO","msg":"writer: started","stream_id":"3rc55aw0"}
+{"time":"2026-03-30T21:36:57.871844976Z","level":"INFO","msg":"sender: started","stream_id":"3rc55aw0"}

fastumi_pickandplace_qwenPI_329v2/wandb/wandb/run-20260330_213657-3rc55aw0/files/requirements.txt

@@ -0,0 +1,161 @@
+starVLA==1.0.1
+kiwisolver==1.4.9
+scipy==1.15.3
+pyarrow==14.0.1
+protobuf==6.33.5
+platformdirs==4.9.4
+mdurl==0.1.2
+Jinja2==3.1.6
+torchvision==0.22.0+cu128
+exceptiongroup==1.3.1
+nvidia-cusparselt-cu12==0.6.3
+markdown-it-py==4.0.0
+ray==2.54.0
+timm==1.0.25
+nvidia-nvjitlink-cu12==12.8.61
+urllib3==2.6.3
+numpydantic==1.6.9
+pillow==12.1.1
+json-numpy==2.1.1
+fastparquet==2024.11.0
+contourpy==1.3.2
+tensorboard-data-server==0.7.2
+albumentations==1.4.18
+deepspeed==0.16.9
+ImageIO==2.37.2
+huggingface_hub==0.36.2
+hjson==3.1.0
+tqdm==4.67.3
+idna==3.11
+packaging==25.0
+python-dateutil==2.9.0.post0
+annotated-types==0.7.0
+regex==2026.2.28
+jsonschema-specifications==2025.9.1
+snntorch==0.9.4
+rpds-py==0.30.0
+cramjam==2.11.0
+importlib_metadata==8.7.1
+torch==2.7.0+cu128
+yacs==0.1.8
+msgpack==1.1.2
+h11==0.16.0
+nvidia-cuda-runtime-cu12==12.8.57
+typing_extensions==4.15.0
+scikit-image==0.25.2
+mpmath==1.3.0
+einops==0.8.2
+wandb==0.25.0
+anyio==4.12.1
+flash_attn==2.7.4.post1
+websockets==15.0.1
+requests==2.32.5
+accelerate==1.5.2
+nvidia-cuda-cupti-cu12==12.8.57
+nvidia-cuda-nvrtc-cu12==12.8.61
+PyYAML==6.0.3
+absl-py==2.4.0
+httpx==0.28.1
+pipablepytorch3d==0.7.6
+nvidia-cublas-cu12==12.8.3.14
+PyMuPDF==1.27.2.2
+decord==0.6.0
+ninja==1.13.0
+albucore==0.0.17
+attrs==26.1.0
+pyparsing==3.3.2
+triton==3.3.0
+Markdown==3.10.2
+Pygments==2.19.2
+pydantic==2.10.6
+tabulate==0.10.0
+termcolor==3.3.0
+zipp==3.23.0
+Werkzeug==3.1.6
+sympy==1.14.0
+debugpy==1.8.20
+certifi==2026.2.25
+websocket==0.2.1
+fonttools==4.61.1
+transformers==4.57.0
+av==12.3.0
+transformers-stream-generator==0.0.4
+nvidia-cudnn-cu12==9.7.1.26
+nvidia-curand-cu12==10.3.9.55
+six==1.17.0
+fvcore==0.1.5.post20221221
+matplotlib==3.10.8
+lazy_loader==0.4
+nvidia-nccl-cu12==2.26.2
+diffusers==0.37.0
+tifffile==2025.5.10
+GitPython==3.1.46
+tokenizers==0.22.2
+eval_type_backport==0.3.1
+nvidia-cufile-cu12==1.13.0.11
+numpy==1.26.4
+filelock==3.25.0
+fsspec==2026.2.0
+nvidia-cusolver-cu12==11.7.2.55
+MarkupSafe==3.0.3
+tyro==1.0.8
+pydantic_core==2.27.2
+portalocker==3.2.0
+qwen-vl-utils==0.0.14
+click==8.3.1
+tiktoken==0.12.0
+smmap==5.0.2
+nvidia-nvtx-cu12==12.8.55
+pytz==2026.1.post1
+rich==14.2.0
+charset-normalizer==3.4.4
+tensorboard==2.20.0
+zope.event==6.1
+zope.interface==8.2
+networkx==3.4.2
+mpi4py==4.1.1
+gitdb==4.0.12
+safetensors==0.7.0
+typeguard==4.5.1
+py-cpuinfo==9.0.0
+websocket-client==1.8.0
+hf-xet==1.3.2
+wheel==0.46.3
+jsonschema==4.26.0
+antlr4-python3-runtime==4.9.3
+gevent==25.9.1
+setuptools==80.9.0
+torchaudio==2.7.0+cu128
+nvidia-cufft-cu12==11.3.3.41
+greenlet==3.3.2
+docstring_parser==0.17.0
+iopath==0.1.10
+cycler==0.12.1
+tzdata==2025.3
+psutil==7.2.2
+referencing==0.37.0
+grpcio==1.78.0
+nvidia-cusparse-cu12==12.5.7.53
+sentry-sdk==2.54.0
+httpcore==1.0.9
+opencv-python-headless==4.11.0.86
+omegaconf==2.3.0
+pandas==2.3.3
+eva-decord==0.6.1
+pip==26.0.1
+inflect==7.3.1
+jaraco.text==3.12.1
+autocommand==2.2.2
+backports.tarfile==1.2.0
+jaraco.functools==4.0.1
+zipp==3.19.2
+more-itertools==10.3.0
+tomli==2.0.1
+typing_extensions==4.12.2
+typeguard==4.3.0
+wheel==0.45.1
+platformdirs==4.2.2
+importlib_metadata==8.0.0
+jaraco.collections==5.1.0
+packaging==24.2
+jaraco.context==5.3.0

fastumi_pickandplace_qwenPI_329v2/wandb/wandb/run-20260330_213657-3rc55aw0/files/wandb-metadata.json

@@ -0,0 +1,161 @@
+{
+  "os":  "Linux-6.8.0-106-generic-x86_64-with-glibc2.39",
+  "python":  "CPython 3.10.19",
+  "startedAt":  "2026-03-30T21:36:57.347545Z",
+  "args":  [
+    "--config_yaml",
+    "./examples/calvin/train_files/starvla_train_calvin.yaml",
+    "--framework.name",
+    "QwenPI",
+    "--framework.qwenvl.base_vlm",
+    "playground/Pretrained_models/Qwen2.5-VL-3B-Instruct-Action",
+    "--framework.qwenvl.attn_implementation",
+    "flash_attention_2",
+    "--framework.action_model.action_dim",
+    "10",
+    "--framework.action_model.state_dim",
+    "10",
+    "--framework.action_model.future_action_window_size",
+    "15",
+    "--framework.action_model.past_action_window_size",
+    "0",
+    "--framework.action_model.action_hidden_dim",
+    "1024",
+    "--framework.action_model.hidden_size",
+    "1024",
+    "--framework.action_model.action_model_type",
+    "DiT-B",
+    "--framework.action_model.add_pos_embed",
+    "True",
+    "--framework.action_model.max_seq_len",
+    "1024",
+    "--framework.action_model.noise_beta_alpha",
+    "1.5",
+    "--framework.action_model.noise_beta_beta",
+    "1.0",
+    "--framework.action_model.noise_s",
+    "0.999",
+    "--framework.action_model.num_timestep_buckets",
+    "1000",
+    "--framework.action_model.num_inference_timesteps",
+    "4",
+    "--framework.action_model.num_target_vision_tokens",
+    "32",
+    "--datasets.vla_data.data_root_dir",
+    "playground/Datasets/FastUMI",
+    "--datasets.vla_data.data_mix",
+    "dynamic-329-v2",
+    "--datasets.vla_data.include_state",
+    "false",
+    "--datasets.vla_data.per_device_batch_size",
+    "8",
+    "--datasets.vla_data.video_backend",
+    "torchvision_av",
+    "--trainer.freeze_modules",
+    "",
+    "--trainer.max_train_steps",
+    "50000",
+    "--trainer.save_interval",
+    "10000",
+    "--trainer.logging_frequency",
+    "50",
+    "--trainer.eval_interval",
+    "100",
+    "--trainer.gradient_accumulation_steps",
+    "1",
+    "--trainer.is_resume",
+    "true",
+    "--run_root_dir",
+    "./results/Checkpoints",
+    "--run_id",
+    "fastumi_pickandplace_qwenPI_329v2",
+    "--wandb_project",
+    "starVLA_FastUMI_dynamic_1",
+    "--wandb_entity",
+    "2200011093-peking-university"
+  ],
+  "program":  "/scratch/wangpc/starVLA/starVLA/training/train_starvla.py",
+  "codePath":  "starVLA/training/train_starvla.py",
+  "codePathLocal":  "starVLA/training/train_starvla.py",
+  "git":  {
+    "remote":  "https://github.com/Kaiwen-Hong/starVLA.git",
+    "commit":  "f4f725fb2ae25a5eda374b783f4dcb9302401417"
+  },
+  "email":  "wangpc@berkeley.edu",
+  "root":  "./results/Checkpoints/fastumi_pickandplace_qwenPI_329v2/wandb",
+  "host":  "tams02",
+  "executable":  "/home/wangpc/miniconda3/envs/starVLA/bin/python3.10",
+  "cpu_count":  192,
+  "cpu_count_logical":  384,
+  "gpu":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "3776651378688",
+      "used":  "158440325120"
+    }
+  },
+  "memory":  {
+    "total":  "1081552064512"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+      "memoryTotal":  "102641958912",
+      "cudaCores":  24064,
+      "architecture":  "Blackwell",
+      "uuid":  "GPU-41f2ee49-ba06-f304-cfa5-4d526d8f5673"
+    },
+    {
+      "name":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+      "memoryTotal":  "102641958912",
+      "cudaCores":  24064,
+      "architecture":  "Blackwell",
+      "uuid":  "GPU-91f82c0c-10ef-1f95-9f6d-1102b8e832b5"
+    },
+    {
+      "name":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+      "memoryTotal":  "102641958912",
+      "cudaCores":  24064,
+      "architecture":  "Blackwell",
+      "uuid":  "GPU-1f3c0889-b740-5143-e064-afeb49245756"
+    },
+    {
+      "name":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+      "memoryTotal":  "102641958912",
+      "cudaCores":  24064,
+      "architecture":  "Blackwell",
+      "uuid":  "GPU-49955a45-509a-e8af-a468-b9ef0449b005"
+    },
+    {
+      "name":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+      "memoryTotal":  "102641958912",
+      "cudaCores":  24064,
+      "architecture":  "Blackwell",
+      "uuid":  "GPU-065fc6ce-c1cd-c143-b421-32b915c9d8fd"
+    },
+    {
+      "name":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+      "memoryTotal":  "102641958912",
+      "cudaCores":  24064,
+      "architecture":  "Blackwell",
+      "uuid":  "GPU-d16b5acf-a100-2d1b-8138-1661892f3d26"
+    },
+    {
+      "name":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+      "memoryTotal":  "102641958912",
+      "cudaCores":  24064,
+      "architecture":  "Blackwell",
+      "uuid":  "GPU-b7a88059-1d3f-da1c-9903-fd4acd4936ab"
+    },
+    {
+      "name":  "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+      "memoryTotal":  "102641958912",
+      "cudaCores":  24064,
+      "architecture":  "Blackwell",
+      "uuid":  "GPU-747dc32b-b025-76e9-697d-fd33ef441b47"
+    }
+  ],
+  "cudaVersion":  "13.1",
+  "writerId":  "ga8ghghrebbnieomg7llhjh60vhax1yn"
+}

fastumi_pickandplace_qwenPI_329v2/wandb/wandb/run-20260330_213657-3rc55aw0/logs/debug-core.log

@@ -0,0 +1,7 @@
+{"time":"2026-03-30T21:36:57.399880536Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpxmorsh73/port-2805506.txt","pid":2805506,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-03-30T21:36:57.400503857Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":2805506}
+{"time":"2026-03-30T21:36:57.400496457Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2805506-2814763-2549433969/socket","Net":"unix"}}
+{"time":"2026-03-30T21:36:57.586371268Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-03-30T21:36:57.590288317Z","level":"INFO","msg":"handleInformInit: received","streamId":"3rc55aw0","id":"1(@)"}
+{"time":"2026-03-30T21:36:57.871817087Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"3rc55aw0","id":"1(@)"}
+{"time":"2026-03-30T21:37:03.29476364Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"5lxv32mzk0uq"}

fastumi_pickandplace_qwenPI_329v2/wandb/wandb/run-20260330_213657-3rc55aw0/logs/debug-internal.log

@@ -0,0 +1,6 @@
+{"time":"2026-03-30T21:36:57.590450795Z","level":"INFO","msg":"stream: starting","core version":"0.25.0"}
+{"time":"2026-03-30T21:36:57.871298035Z","level":"INFO","msg":"stream: created new stream","id":"3rc55aw0"}
+{"time":"2026-03-30T21:36:57.871724848Z","level":"INFO","msg":"handler: started","stream_id":"3rc55aw0"}
+{"time":"2026-03-30T21:36:57.871808977Z","level":"INFO","msg":"stream: started","id":"3rc55aw0"}
+{"time":"2026-03-30T21:36:57.871845186Z","level":"INFO","msg":"writer: started","stream_id":"3rc55aw0"}
+{"time":"2026-03-30T21:36:57.871844976Z","level":"INFO","msg":"sender: started","stream_id":"3rc55aw0"}

fastumi_pickandplace_qwenPI_329v2/wandb/wandb/run-20260330_213657-3rc55aw0/run-3rc55aw0.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0dd034e0fde2fbbf4d96f101d5229e23725535c9b4c98cc881ba5f3f6f7d0159
+size 10846208