Upload folder using huggingface_hub

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/adapter/config.json

@@ -0,0 +1,7 @@
+{
+  "_class_name": "TextAdapter",
+  "_diffusers_version": "0.17.1",
+  "in_dim": 1024,
+  "int_dim": 896,
+  "out_dim": 1024
+}

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/adapter/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01c9018495bd23c6b7cbf80ad5032ced9ad9d837bc912ef2c977a68e26682ffc
+size 88693665

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/optimizer.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10f786eab56c17adc60bac922fddad5f19bb8aa74e949dfead5c5316f034cb5d
+size 3881234331

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a863133b3f74977a18ad98d697da8515d45b267096f4694b4f22ac6f3f01a1f3
+size 1854200581

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/pytorch_model_1.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:977f8e0133b3f2502a75f512e8638b5122bc7ff21a9fcbf3869d6a8acc9a828f
+size 1496717673

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/pytorch_model_2.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6475eaebf9a7aee9314e6e88218fb8362d2bef34180b91d7e902d51128aaf69f
+size 88693425

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/random_states_0.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d63cb9fb2a5f56837756cebaa3a8cb947b660ff1d88428755d04973021213b9e
+size 17747

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/random_states_1.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dcf56463be636fe6a38ff30440cca6e268506c184e1a4e5ea2dfb623d47341b
+size 17747

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/random_states_2.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6eb45d91195bdc76d1ec2f0d7d9fbac7179b9dbbf4cd4e53db7a087eb9dedf1
+size 17747

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/random_states_3.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3345b5d76cb40ea8ee98f22e921fbebfc6053aa1f7b08ab4951a0f0897ec65ef
+size 17747

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fce4e8b2846330e7f90197e8f3876452f5224d090d9fa143a17219be96966945
+size 557

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/checkpoint-75000/scheduler.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09292b540a5bd17f70e67cef5ca61cc55788c0b633bfc767f117a92550659161
+size 627

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/hydra/config.yaml

@@ -0,0 +1,495 @@
+task_id: occ
+log_root_prefix: work_dirs/debug
+projname: ${model.name}
+try_run: true
+debug: false
+log_root: ./work_dirs
+init_method: env://
+seed: 100
+fix_seed_within_batch: false
+resume_from_checkpoint: work_dirs/occ_rawbox_2025-04-17_22-56_occ/checkpoint-740000
+resume_reset_scheduler: false
+validation_only: false
+model:
+  name: occ_rawbox
+  pretrained_model_name_or_path: pretrained/stable-diffusion-v2-1/
+  pretrained_t5_path: pretrained/t5-large/
+  pretrained_lavi_bridge_path: pretrained/LaVi-Bridge/t5_unet/
+  bbox_mode: all-xyz
+  bbox_view_shared: true
+  condition_embed_dim: 1024
+  train_with_same_noise: false
+  runner_module: magicdrive.runner.occ_runner.OccRunner
+  pipe_module: magicdrive.pipeline.pipeline_occ_unet.StableDiffusionOccUNetPipeline
+  vae_module: magicdrive.occ_vae.networks.networks.AutoEncoderGroupSkip
+  vae_ckpt: work_dirs/vae/best_model
+  vae:
+    num_class: 17
+    geo_feat_channels: 16
+    feat_channel_up: 64
+    mlp_hidden_channels: 256
+    mlp_hidden_layers: 4
+    z_down: false
+    padding_mode: replicate
+    voxel_fea: false
+    triplane: true
+    pos: true
+    dataset: Occ3D-nuScenes
+  unet_module: magicdrive.networks.unet_2d_condition_occ.UNet2DConditionModelOcc
+  use_fp32_for_unet_trainable: true
+  unet_dir: unet_occ
+  unet:
+    sample_size: 116
+    in_channels: 16
+    out_channels: 16
+    down_block_types:
+    - CrossAttnDownBlock2D
+    - CrossAttnDownBlock2D
+    - CrossAttnDownBlock2D
+    - DownBlock2D
+    up_block_types:
+    - UpBlock2D
+    - CrossAttnUpBlock2D
+    - CrossAttnUpBlock2D
+    - CrossAttnUpBlock2D
+    block_out_channels:
+    - 64
+    - 128
+    - 256
+    - 512
+    layers_per_block: 2
+    cross_attention_dim: ${model.condition_embed_dim}
+    tri_size: ${dataset.tri_size}
+    is_rollout: true
+    conv_down: true
+    conv_up: true
+    tri_z_down: false
+    map_with_aux_box: ${runner.triplane_aux_box}
+    map_size_wo_aux_box:
+    - 16
+    - 216
+    - 216
+    map_size_w_aux_box:
+    - 24
+    - 216
+    - 216
+    conditioning_embedding_out_channels:
+    - 16
+    - 32
+    - 96
+    - 256
+    use_uncond_map: zero
+    drop_cond_ratio: 0.25
+    bbox_embedder_cls: magicdrive.networks.bbox_embedder.ContinuousBBoxWithTextEmbedding
+    bbox_embedder_param:
+      n_classes: 10
+      class_token_dim: ${model.condition_embed_dim}
+      trainable_class_token: false
+      use_text_encoder_init: true
+      embedder_num_freq: 4
+      proj_dims:
+      - 1024
+      - 512
+      - 512
+      - 1024
+      mode: ${...bbox_mode}
+      minmax_normalize: false
+dataset:
+  dataset_type: NuScenesDatasetM
+  occ_dataset_type: Occ3D-nuScenes
+  dataset_root: data/nuscenes/
+  triplane_root: data/nuscenes/nuscenes_triplane_100_100_16
+  dataset_process_root: data/nuscenes/nuscenes_mmdet3d_2/
+  dataset_cache_file_tag: 26x200x200_map_aux_full_16x216x216_triplane_map
+  dataset_cache_file:
+  - ${dataset.dataset_process_root}../nuscenes_map_aux/train_${dataset.dataset_cache_file_tag}.h5
+  - ${dataset.dataset_process_root}../nuscenes_map_aux/val_${dataset.dataset_cache_file_tag}.h5
+  template_clip: A driving scene image at {location}. {description}.
+  template_t5: A driving scene at {location}. {description}. {detailed_description}
+  image_size:
+  - 224
+  - 400
+  map_bound:
+    x:
+    - -40.0
+    - 40.0
+    - 0.4
+    'y':
+    - -40.0
+    - 40.0
+    - 0.4
+    z:
+    - -1.0
+    - 5.4
+    - 0.4
+  tri_size:
+  - 100
+  - 100
+  - 16
+  view_order:
+  - CAM_FRONT_LEFT
+  - CAM_FRONT
+  - CAM_FRONT_RIGHT
+  - CAM_BACK_RIGHT
+  - CAM_BACK
+  - CAM_BACK_LEFT
+  neighboring_view_pair:
+    0:
+    - 5
+    - 1
+    1:
+    - 0
+    - 2
+    2:
+    - 1
+    - 3
+    3:
+    - 2
+    - 4
+    4:
+    - 3
+    - 5
+    5:
+    - 4
+    - 0
+  back_resize:
+  - 896
+  - 1600
+  back_pad:
+  - 0
+  - 4
+  - 0
+  - 0
+  augment2d:
+    resize:
+    - - 0.25
+      - 0.25
+    rotate: null
+  aux_data:
+  - visibility
+  - center_offset
+  - center_ohw
+  - height
+  augment3d:
+    scale:
+    - 1.0
+    - 1.0
+    rotate:
+    - 0.0
+    - 0.0
+    translate: 0
+    flip_ratio: 0.0
+    flip_direction: null
+  object_classes:
+  - barrier
+  - bicycle
+  - bus
+  - car
+  - construction_vehicle
+  - motorcycle
+  - pedestrian
+  - traffic_cone
+  - trailer
+  - truck
+  map_classes:
+  - drivable_area
+  - ped_crossing
+  - walkway
+  - stop_line
+  - carpark_area
+  - road_divider
+  - lane_divider
+  - road_block
+  input_modality:
+    use_lidar: false
+    use_camera: true
+    use_radar: false
+    use_map: false
+    use_external: false
+  train_pipeline:
+  - type: LoadMultiViewImageFromFiles
+    to_float32: true
+  - type: LoadAnnotations3D
+    with_bbox_3d: true
+    with_label_3d: true
+    with_attr_label: false
+  - type: ImageAug3D
+    final_dim: ${...image_size}
+    resize_lim: ${...augment2d.resize[0]}
+    bot_pct_lim:
+    - 0.0
+    - 0.0
+    rot_lim: ${...augment2d.rotate}
+    rand_flip: false
+    is_train: false
+  - type: GlobalRotScaleTrans
+    resize_lim: ${...augment3d.scale}
+    rot_lim: ${...augment3d.rotate}
+    trans_lim: ${...augment3d.translate}
+    is_train: true
+  - type: ObjectNameFilterM
+    classes: ${...object_classes}
+  - type: LoadBEVSegmentationM
+    dataset_root: ${...dataset_root}
+    xbound: ${...map_bound.x}
+    ybound: ${...map_bound.y}
+    classes: ${...map_classes}
+    object_classes: ${...object_classes}
+    aux_data: ${...aux_data}
+    cache_file: ${...dataset_cache_file.0}
+  - type: LoadBEVHDMap
+    dataset_root: ${...dataset_root}
+    xbound: ${...map_bound.x}
+    ybound: ${...map_bound.y}
+    image_size: ${...image_size}
+    object_classes: ${...object_classes}
+  - type: LoadTriPlane
+    dataset_root: ${...triplane_root}
+    dataset_type: ${...occ_dataset_type}
+  - type: LoadTriPlaneSegmentation
+    dataset_root: ${...dataset_root}
+    xbound: ${...map_bound.x}
+    ybound: ${...map_bound.y}
+    zbound: ${...map_bound.z}
+    cache_file: ${...dataset_cache_file.0}
+    dataset_type: ${...occ_dataset_type}
+  - type: LoadOccRenderMap
+    dataset_root: ${occ_render_path}
+    num_classes: 20
+  - type: RandomFlip3DwithViews
+    flip_ratio: ${...augment3d.flip_ratio}
+    direction: ${...augment3d.flip_direction}
+  - type: LoadDescription
+    dataset_root: ${...dataset_root}
+    dataset_type: ${...occ_dataset_type}
+  - type: ReorderMultiViewImagesM
+    order: ${...view_order}
+    safe: false
+  - type: ImageNormalize
+    mean:
+    - 0.5
+    - 0.5
+    - 0.5
+    std:
+    - 0.5
+    - 0.5
+    - 0.5
+  - type: DefaultFormatBundle3D
+    classes: ${...object_classes}
+  - type: Collect3D
+    keys:
+    - img
+    - triplane
+    - gt_bboxes_3d
+    - gt_labels_3d
+    - gt_masks_bev
+    - gt_aux_bev
+    - bev_hdmap
+    - layout_canvas
+    - gt_masks_triplane
+    - occ_render_image
+    - occ_render_depth
+    meta_keys:
+    - camera_intrinsics
+    - lidar2ego
+    - lidar2camera
+    - camera2lidar
+    - lidar2image
+    - img_aug_matrix
+    meta_lis_keys:
+    - timeofday
+    - location
+    - description
+    - detailed_description
+    - filename
+    - token
+  test_pipeline:
+  - type: LoadMultiViewImageFromFiles
+    to_float32: true
+  - type: LoadAnnotations3D
+    with_bbox_3d: true
+    with_label_3d: true
+    with_attr_label: false
+  - type: ImageAug3D
+    final_dim: ${...image_size}
+    resize_lim: ${...augment2d.resize[0]}
+    bot_pct_lim:
+    - 0.0
+    - 0.0
+    rot_lim:
+    - 0.0
+    - 0.0
+    rand_flip: false
+    is_train: false
+  - type: GlobalRotScaleTrans
+    resize_lim: ${...augment3d.scale}
+    rot_lim: ${...augment3d.rotate}
+    trans_lim: ${...augment3d.translate}
+    is_train: true
+  - type: ObjectNameFilterM
+    classes: ${...object_classes}
+  - type: LoadBEVSegmentationM
+    dataset_root: ${...dataset_root}
+    xbound: ${...map_bound.x}
+    ybound: ${...map_bound.y}
+    classes: ${...map_classes}
+    object_classes: ${...object_classes}
+    aux_data: ${...aux_data}
+    cache_file: ${...dataset_cache_file.1}
+  - type: LoadBEVHDMap
+    dataset_root: ${...dataset_root}
+    xbound: ${...map_bound.x}
+    ybound: ${...map_bound.y}
+    image_size: ${...image_size}
+    object_classes: ${...object_classes}
+  - type: LoadTriPlane
+    dataset_root: ${...triplane_root}
+    dataset_type: ${...occ_dataset_type}
+  - type: LoadTriPlaneSegmentation
+    dataset_root: ${...dataset_root}
+    xbound: ${...map_bound.x}
+    ybound: ${...map_bound.y}
+    zbound: ${...map_bound.z}
+    cache_file: ${...dataset_cache_file.1}
+    dataset_type: ${...occ_dataset_type}
+  - type: LoadOccRenderMap
+    dataset_root: ${occ_render_path}
+    num_classes: 20
+  - type: LoadDescription
+    dataset_root: ${...dataset_root}
+    dataset_type: ${...occ_dataset_type}
+  - type: ReorderMultiViewImagesM
+    order: ${...view_order}
+    safe: false
+  - type: ImageNormalize
+    mean:
+    - 0.5
+    - 0.5
+    - 0.5
+    std:
+    - 0.5
+    - 0.5
+    - 0.5
+  - type: DefaultFormatBundle3D
+    classes: ${...object_classes}
+  - type: Collect3D
+    keys:
+    - img
+    - triplane
+    - gt_bboxes_3d
+    - gt_labels_3d
+    - gt_masks_bev
+    - gt_aux_bev
+    - bev_hdmap
+    - layout_canvas
+    - gt_masks_triplane
+    - occ_render_image
+    - occ_render_depth
+    meta_keys:
+    - camera_intrinsics
+    - lidar2ego
+    - lidar2camera
+    - camera2lidar
+    - lidar2image
+    - img_aug_matrix
+    meta_lis_keys:
+    - timeofday
+    - location
+    - description
+    - detailed_description
+    - filename
+    - token
+  data:
+    train:
+      type: ${...dataset_type}
+      dataset_root: ${...dataset_root}
+      ann_file: ${...dataset_process_root}nuscenes_infos_train.pkl
+      pipeline: ${...train_pipeline}
+      object_classes: ${...object_classes}
+      map_classes: ${...map_classes}
+      modality: ${...input_modality}
+      test_mode: false
+      force_all_boxes: true
+      box_type_3d: LiDAR
+      filter_empty_gt: false
+    val:
+      type: ${...dataset_type}
+      dataset_root: ${...dataset_root}
+      ann_file: ${...dataset_process_root}nuscenes_infos_val.pkl
+      pipeline: ${...test_pipeline}
+      object_classes: ${...object_classes}
+      map_classes: ${...map_classes}
+      modality: ${...input_modality}
+      test_mode: false
+      force_all_boxes: true
+      box_type_3d: LiDAR
+      filter_empty_gt: false
+    test:
+      type: ${...dataset_type}
+      dataset_root: ${...dataset_root}
+      ann_file: ${...dataset_process_root}nuscenes_infos_val.pkl
+      pipeline: ${...test_pipeline}
+      object_classes: ${...object_classes}
+      map_classes: ${...map_classes}
+      modality: ${...input_modality}
+      test_mode: true
+      force_all_boxes: true
+      box_type_3d: LiDAR
+      filter_empty_gt: false
+accelerator:
+  gradient_accumulation_steps: 1
+  mixed_precision: fp16
+  report_to: tensorboard
+runner:
+  foreground_loss_weight: 0.0
+  bbox_drop_ratio: 0
+  bbox_add_ratio: 0
+  bbox_add_num: 3
+  triplane_aux_box: false
+  num_train_epochs: 800
+  train_batch_size: 1
+  max_train_steps: 7
+  num_workers: 8
+  prefetch_factor: 4
+  display_per_epoch: 3125
+  display_per_n_min: 10
+  max_grad_norm: 1.0
+  set_grads_to_none: true
+  enable_xformers_memory_efficient_attention: true
+  unet_in_fp16: false
+  enable_unet_checkpointing: true
+  enable_controlnet_checkpointing: false
+  noise_offset: 0.0
+  train_with_same_offset: true
+  use_8bit_adam: false
+  adam_beta1: 0.9
+  adam_beta2: 0.999
+  adam_weight_decay: 0.01
+  adam_epsilon: 1.0e-08
+  learning_rate: 0.0001
+  lr_scheduler: constant
+  gradient_accumulation_steps: 1
+  lr_num_cycles: 1
+  lr_power: 1.0
+  lr_warmup_steps: 3000
+  checkpointing_steps: 5
+  validation_steps: 5
+  save_model_per_epoch: null
+  validation_before_run: false
+  validation_index:
+  - 204
+  - 912
+  - 1828
+  validation_times: 2
+  validation_batch_size: 1
+  validation_show_box: true
+  validation_seed_global: false
+  pipeline_param:
+    guidance_scale: 2
+    num_inference_steps: 20
+    eta: 0.0
+    controlnet_conditioning_scale: 1.0
+    guess_mode: false
+    use_zero_map_as_unconditional: false
+    bbox_max_length: null
+occ_render_path: data/nuscenes/occ_render_map/

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/hydra/hydra.yaml

@@ -0,0 +1,162 @@
+hydra:
+  run:
+    dir: ${log_root_prefix}/${projname}_${now:%Y-%m-%d}_${now:%H-%M}_${task_id}
+  sweep:
+    dir: multirun/${now:%Y-%m-%d}/${now:%H-%M-%S}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: hydra
+  overrides:
+    hydra:
+    - hydra.mode=RUN
+    task:
+    - +exp=occ
+    - runner=debug
+  job:
+    name: train
+    chdir: null
+    override_dirname: +exp=occ,runner=debug
+    id: ???
+    num: ???
+    config_name: config_occ
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.3.0
+    version_base: '1.3'
+    cwd: /data/yyang/workspace/MagicDrive
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /data/yyang/workspace/MagicDrive/configs
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /data/yyang/workspace/MagicDrive/work_dirs/debug/occ_rawbox_2025-04-20_23-06_occ
+    choices:
+      exp: occ
+      exp/model@model: ../../model/occ_rawbox
+      runner: debug
+      accelerator: default
+      dataset: Nuscenes_cache
+      model: occ_rawbox
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/hydra/overrides.yaml

@@ -0,0 +1,2 @@
+- +exp=occ
+- runner=debug

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/lora_text.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:402ee064d71b49e9b98f586a03197407685aa975abac0b10ecc3e2cd424570dd
+size 12634373

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/tb-occ/events.out.tfevents.1745161677.cvrp-gpu-5.2144986.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97fca8d2dc7d80e10817b95cbff822e645a76304787ef6ceeddb40b3ab5c57a2
+size 40

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/unet_occ/config.json

@@ -0,0 +1,107 @@
+{
+  "_class_name": "UNet2DConditionModelOcc",
+  "_diffusers_version": "0.17.1",
+  "act_fn": "silu",
+  "addition_embed_type": null,
+  "addition_embed_type_num_heads": 64,
+  "attention_head_dim": 8,
+  "bbox_embedder_cls": "magicdrive.networks.bbox_embedder.ContinuousBBoxWithTextEmbedding",
+  "bbox_embedder_param": {
+    "class_token_dim": 1024,
+    "embedder_num_freq": 4,
+    "minmax_normalize": false,
+    "mode": "all-xyz",
+    "n_classes": 10,
+    "proj_dims": [
+      1024,
+      512,
+      512,
+      1024
+    ],
+    "trainable_class_token": false,
+    "use_text_encoder_init": true
+  },
+  "block_out_channels": [
+    64,
+    128,
+    256,
+    512
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conditioning_embedding_out_channels": [
+    16,
+    32,
+    96,
+    256
+  ],
+  "conv_down": true,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "conv_up": true,
+  "cross_attention_dim": 1024,
+  "cross_attention_norm": null,
+  "down_block_types": [
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "drop_cond_ratio": 0.25,
+  "dual_cross_attention": false,
+  "encoder_hid_dim": null,
+  "encoder_hid_dim_type": null,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 16,
+  "is_rollout": true,
+  "layers_per_block": 2,
+  "map_embedder_cls": null,
+  "map_embedder_param": null,
+  "map_size_w_aux_box": [
+    24,
+    216,
+    216
+  ],
+  "map_size_wo_aux_box": [
+    16,
+    216,
+    216
+  ],
+  "map_with_aux_box": false,
+  "mid_block_only_cross_attention": null,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock2DCrossAttn",
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 16,
+  "projection_class_embeddings_input_dim": null,
+  "resnet_out_scale_factor": 1.0,
+  "resnet_skip_time_act": false,
+  "resnet_time_scale_shift": "default",
+  "sample_size": 116,
+  "time_cond_proj_dim": null,
+  "time_embedding_act_fn": null,
+  "time_embedding_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "tri_size": [
+    100,
+    100,
+    16
+  ],
+  "tri_z_down": false,
+  "up_block_types": [
+    "UpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
+  ],
+  "upcast_attention": false,
+  "use_linear_projection": false,
+  "use_uncond_map": "zero"
+}

occ_rawbox_100_100_16_v3_hdmapv2_t5_epoch630/unet_occ/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c254bb79844033c32503a381c7648c27a480e17edd0a6b0b67c003110ffd6dd
+size 1854967641