init commit

.gitignore

@@ -3,6 +3,13 @@ __pycache__/
 *.py[cod]
 *$py.class
 *.ipynb
+amlt/
+mmdetection/
+mmdetection3d/
+data
+ckpts
+work_dirs
+.amltignore
 
 # C extensions
 *.so

projects/configs/petrv2/hybrid_petrv2_vovnet_gridmask_p4_800x320_lambda1_group4_t1800.py

@@ -0,0 +1,252 @@
+_base_ = [
+    '../../../mmdetection3d/configs/_base_/datasets/nus-3d.py',
+    '../../../mmdetection3d/configs/_base_/default_runtime.py'
+]
+backbone_norm_cfg = dict(type='LN', requires_grad=True)
+plugin=True
+plugin_dir='projects/mmdet3d_plugin/'
+
+# If point cloud range is changed, the models should also change their point
+# cloud range accordingly
+point_cloud_range = [-51.2, -51.2, -5.0, 51.2, 51.2, 3.0]
+voxel_size = [0.2, 0.2, 8]
+img_norm_cfg = dict(
+    mean=[103.530, 116.280, 123.675], std=[57.375, 57.120, 58.395], to_rgb=False)
+# For nuScenes we usually do 10-class detection
+class_names = [
+    'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier',
+    'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'
+]
+input_modality = dict(
+    use_lidar=True,
+    use_camera=True,
+    use_radar=False,
+    use_map=False,
+    use_external=False)
+model = dict(
+    type='Petr3D',
+    use_grid_mask=True,
+    img_backbone=dict(
+        type='VoVNetCP', ###use checkpoint to save memory
+        spec_name='V-99-eSE',
+        norm_eval=True,
+        frozen_stages=-1,
+        input_ch=3,
+        out_features=('stage4','stage5',)),
+    img_neck=dict(
+        type='CPFPN',  ###remove unused parameters 
+        in_channels=[768, 1024],
+        out_channels=256,
+        num_outs=2),
+    pts_bbox_head=dict(
+        type='HybridPETRv2Head',
+        num_classes=10,
+        in_channels=256,
+        num_query=2700,
+        num_queries_one2one=900,
+        k_one2many=4,
+        LID=True,
+        with_position=True,
+        with_multiview=True,
+        with_fpe=True,
+        with_time=True,
+        with_multi=True,
+        position_range=[-61.2, -61.2, -10.0, 61.2, 61.2, 10.0],
+        code_weights = [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
+        transformer=dict(
+            type='PETRTransformer',
+            decoder=dict(
+                type='PETRTransformerDecoder',
+                return_intermediate=True,
+                num_layers=6,
+                transformerlayers=dict(
+                    type='PETRTransformerDecoderLayer',
+                    attn_cfgs=[
+                        dict(
+                            type='MultiheadAttention',
+                            embed_dims=256,
+                            num_heads=8,
+                            dropout=0.1),
+                        dict(
+                            type='PETRMultiheadAttention',
+                            embed_dims=256,
+                            num_heads=8,
+                            dropout=0.1),
+                        ],
+                    feedforward_channels=2048,
+                    ffn_dropout=0.1,
+                    with_cp=True,  ###use checkpoint to save memory
+                    operation_order=('self_attn', 'norm', 'cross_attn', 'norm',
+                                     'ffn', 'norm')),
+            )),
+        bbox_coder=dict(
+            type='NMSFreeCoder',
+            post_center_range=[-61.2, -61.2, -10.0, 61.2, 61.2, 10.0],
+            pc_range=point_cloud_range,
+            max_num=300,
+            voxel_size=voxel_size,
+            num_classes=10), 
+        positional_encoding=dict(
+            type='SinePositionalEncoding3D', num_feats=128, normalize=True),
+        loss_cls=dict(
+            type='FocalLoss',
+            use_sigmoid=True,
+            gamma=2.0,
+            alpha=0.25,
+            loss_weight=2.0),
+        loss_bbox=dict(type='L1Loss', loss_weight=0.25),
+        loss_iou=dict(type='GIoULoss', loss_weight=0.0)),
+    # model training and testing settings
+    train_cfg=dict(pts=dict(
+        grid_size=[512, 512, 1],
+        voxel_size=voxel_size,
+        point_cloud_range=point_cloud_range,
+        out_size_factor=4,
+        assigner=dict(
+            type='HungarianAssigner3D',
+            cls_cost=dict(type='FocalLossCost', weight=2.0),
+            reg_cost=dict(type='BBox3DL1Cost', weight=0.25),
+            iou_cost=dict(type='IoUCost', weight=0.0), # Fake cost. This is just to make it compatible with DETR head. 
+            pc_range=point_cloud_range))))
+
+dataset_type = 'CustomNuScenesDataset'
+data_root = 'data/nuscenes/'
+
+file_client_args = dict(backend='disk')
+
+db_sampler = dict(
+    data_root=data_root,
+    info_path=data_root + 'nuscenes_dbinfos_train.pkl',
+    rate=1.0,
+    prepare=dict(
+        filter_by_difficulty=[-1],
+        filter_by_min_points=dict(
+            car=5,
+            truck=5,
+            bus=5,
+            trailer=5,
+            construction_vehicle=5,
+            traffic_cone=5,
+            barrier=5,
+            motorcycle=5,
+            bicycle=5,
+            pedestrian=5)),
+    classes=class_names,
+    sample_groups=dict(
+        car=2,
+        truck=3,
+        construction_vehicle=7,
+        bus=4,
+        trailer=6,
+        barrier=2,
+        motorcycle=6,
+        bicycle=6,
+        pedestrian=2,
+        traffic_cone=2),
+    points_loader=dict(
+        type='LoadPointsFromFile',
+        coord_type='LIDAR',
+        load_dim=5,
+        use_dim=[0, 1, 2, 3, 4],
+        file_client_args=file_client_args))
+ida_aug_conf = {
+        "resize_lim": (0.47, 0.625),
+        "final_dim": (320, 800),
+        "bot_pct_lim": (0.0, 0.0),
+        "rot_lim": (0.0, 0.0),
+        "H": 900,
+        "W": 1600,
+        "rand_flip": True,
+    }
+train_pipeline = [
+    dict(type='LoadMultiViewImageFromFiles', to_float32=True),
+    dict(type='LoadMultiViewImageFromMultiSweepsFiles', sweeps_num=1, to_float32=True, pad_empty_sweeps=True, test_mode=False, sweep_range=[3,27]),
+    dict(type='LoadAnnotations3D', with_bbox_3d=True, with_label_3d=True, with_attr_label=False),
+    dict(type='ObjectRangeFilter', point_cloud_range=point_cloud_range),
+    dict(type='ObjectNameFilter', classes=class_names),
+    dict(type='ResizeCropFlipImage', data_aug_conf = ida_aug_conf, training=True),
+    dict(type='GlobalRotScaleTransImage',
+            rot_range=[-0.3925, 0.3925],
+            translation_std=[0, 0, 0],
+            scale_ratio_range=[0.95, 1.05],
+            reverse_angle=True,
+            training=True
+            ),
+    dict(type='NormalizeMultiviewImage', **img_norm_cfg),
+    dict(type='PadMultiViewImage', size_divisor=32),
+    dict(type='DefaultFormatBundle3D', class_names=class_names),
+    dict(type='Collect3D', keys=['gt_bboxes_3d', 'gt_labels_3d', 'img'],
+            meta_keys=('filename', 'ori_shape', 'img_shape', 'lidar2img', 'intrinsics', 'extrinsics',
+                'pad_shape', 'scale_factor', 'flip', 'box_mode_3d', 'box_type_3d',
+                'img_norm_cfg', 'sample_idx', 'timestamp'))
+]
+test_pipeline = [
+    dict(type='LoadMultiViewImageFromFiles', to_float32=True),
+    dict(type='LoadMultiViewImageFromMultiSweepsFiles', sweeps_num=1, to_float32=True, pad_empty_sweeps=True, sweep_range=[3,27]),
+    dict(type='ResizeCropFlipImage', data_aug_conf = ida_aug_conf, training=False),
+    dict(type='NormalizeMultiviewImage', **img_norm_cfg),
+    dict(type='PadMultiViewImage', size_divisor=32),
+    dict(
+        type='MultiScaleFlipAug3D',
+        img_scale=(1333, 800),
+        pts_scale_ratio=1,
+        flip=False,
+        transforms=[
+            dict(
+                type='DefaultFormatBundle3D',
+                class_names=class_names,
+                with_label=False),
+            dict(type='Collect3D', keys=['img'],
+            meta_keys=('filename', 'ori_shape', 'img_shape', 'lidar2img', 'intrinsics', 'extrinsics',
+                'pad_shape', 'scale_factor', 'flip', 'box_mode_3d', 'box_type_3d',
+                'img_norm_cfg', 'sample_idx', 'timestamp'))
+        ])
+]
+
+data = dict(
+    samples_per_gpu=1,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        data_root=data_root,
+        ann_file=data_root + 'mmdet3d_nuscenes_30f_infos_train.pkl',
+        pipeline=train_pipeline,
+        classes=class_names,
+        modality=input_modality,
+        test_mode=False,
+        use_valid_flag=True,
+        # we use box_type_3d='LiDAR' in kitti and nuscenes dataset
+        # and box_type_3d='Depth' in sunrgbd and scannet dataset.
+        box_type_3d='LiDAR'),
+    val=dict(type=dataset_type, pipeline=test_pipeline, ann_file=data_root + 'mmdet3d_nuscenes_30f_infos_val.pkl', classes=class_names, modality=input_modality),
+    test=dict(type=dataset_type, pipeline=test_pipeline, ann_file=data_root + 'mmdet3d_nuscenes_30f_infos_val.pkl', classes=class_names, modality=input_modality))
+    # test=dict(type=dataset_type, pipeline=test_pipeline, ann_file=data_root + 'mmdet3d_nuscenes_30f_infos_test.pkl', classes=class_names, modality=input_modality))
+
+
+optimizer = dict(
+    type='AdamW', 
+    lr=2e-4,
+    paramwise_cfg=dict(
+        custom_keys={
+            'img_backbone': dict(lr_mult=0.1),
+        }),
+    weight_decay=0.01)
+
+optimizer_config = dict(type='Fp16OptimizerHook', loss_scale=512., grad_clip=dict(max_norm=35, norm_type=2))
+
+# learning policy
+lr_config = dict(
+    policy='CosineAnnealing',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=1.0 / 3,
+    min_lr_ratio=1e-3,
+    )
+total_epochs = 24
+evaluation = dict(interval=6, pipeline=test_pipeline)
+find_unused_parameters=False #### when use checkpoint, find_unused_parameters must be False
+checkpoint_config = dict(interval=1, max_keep_ckpts=12, create_symlink=False)
+runner = dict(type='EpochBasedRunner', max_epochs=total_epochs)
+load_from='ckpts/fcos3d_vovnet_imgbackbone-remapped.pth'
+resume_from=None
+