Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ 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
+nuscenes/semseg-ptv3_dino-g/train.log filter=lfs diff=lfs merge=lfs -text

nuscenes/semseg-ptv3_dino-g/config.py

@@ -0,0 +1,261 @@
+weight = None
+resume = False
+evaluate = True
+test_only = False
+seed = 38690432
+save_path = 'exp/nuscenes/semseg-ptv3_dino-g'
+wandb_project = 'semseg_nuscenes'
+num_worker = 16
+batch_size = 12
+batch_size_val = None
+batch_size_test = None
+epoch = 50
+eval_epoch = 50
+clip_grad = None
+sync_bn = False
+enable_amp = True
+empty_cache = False
+empty_cache_per_epoch = False
+find_unused_parameters = False
+mix_prob = 0.8
+param_dicts = [dict(keyword='img_enc|block', lr=0.0002)]
+hooks = [
+    dict(type='CheckpointLoader'),
+    dict(type='IterationTimer', warmup_iter=2),
+    dict(type='InformationWriter'),
+    dict(type='SemSegEvaluator'),
+    dict(type='CheckpointSaver', save_freq=None),
+    dict(type='PreciseEvaluator', test_last=False)
+]
+train = dict(type='DefaultTrainer')
+test = dict(type='SemSegTester', verbose=True)
+model = dict(
+    type='DefaultSegmentorV2',
+    num_classes=16,
+    backbone_out_channels=64,
+    backbone=dict(
+        type='PT-v3m1-image',
+        in_channels=4,
+        order=['z', 'z-trans', 'hilbert', 'hilbert-trans'],
+        stride=(2, 2, 2, 2),
+        enc_depths=(2, 2, 2, 6, 2),
+        enc_channels=(32, 64, 128, 256, 512),
+        enc_num_head=(2, 4, 8, 16, 32),
+        enc_patch_size=(1024, 1024, 1024, 1024, 1024),
+        dec_depths=(2, 2, 2, 2),
+        dec_channels=(64, 64, 128, 256),
+        dec_num_head=(4, 4, 8, 16),
+        dec_patch_size=(1024, 1024, 1024, 1024),
+        mlp_ratio=4,
+        qkv_bias=True,
+        qk_scale=None,
+        init_values=None,
+        attn_drop=0.0,
+        proj_drop=0.0,
+        drop_path=0.3,
+        shuffle_orders=True,
+        pre_norm=True,
+        enable_rpe=False,
+        enable_flash=True,
+        upcast_attention=False,
+        upcast_softmax=False,
+        cls_mode=False,
+        pdnorm_bn=False,
+        pdnorm_ln=False,
+        pdnorm_decouple=True,
+        pdnorm_adaptive=False,
+        pdnorm_affine=True,
+        pdnorm_conditions=('nuScenes', 'SemanticKITTI', 'Waymo'),
+        dinov2='giant',
+        dinov2_missing_embedding=False,
+        dinov2_drop=0.0,
+        dinov2_lora=0,
+        dinov2_frozen=True),
+    criteria=[
+        dict(type='CrossEntropyLoss', loss_weight=1.0, ignore_index=-1),
+        dict(
+            type='LovaszLoss',
+            mode='multiclass',
+            loss_weight=1.0,
+            ignore_index=-1)
+    ])
+optimizer = dict(type='AdamW', lr=0.002, weight_decay=0.005)
+scheduler = dict(
+    type='OneCycleLR',
+    max_lr=[0.002, 0.0002],
+    pct_start=0.04,
+    anneal_strategy='cos',
+    div_factor=10.0,
+    final_div_factor=100.0)
+dataset_type = 'NuScenesDataset'
+data_root = 'data/nuscenes'
+ignore_index = -1
+names = [
+    'barrier', 'bicycle', 'bus', 'car', 'construction_vehicle', 'motorcycle',
+    'pedestrian', 'traffic_cone', 'trailer', 'truck', 'driveable_surface',
+    'other_flat', 'sidewalk', 'terrain', 'manmade', 'vegetation'
+]
+data = dict(
+    num_classes=16,
+    ignore_index=-1,
+    names=[
+        'barrier', 'bicycle', 'bus', 'car', 'construction_vehicle',
+        'motorcycle', 'pedestrian', 'traffic_cone', 'trailer', 'truck',
+        'driveable_surface', 'other_flat', 'sidewalk', 'terrain', 'manmade',
+        'vegetation'
+    ],
+    train=dict(
+        type='NuScenesDataset',
+        split='train',
+        data_root='data/nuscenes',
+        with_images=True,
+        transform=[
+            dict(type='ImageResize', size=[378, 672]),
+            dict(
+                type='ImageColorJitter',
+                brightness=0.4,
+                contrast=0.4,
+                saturation=0.2,
+                hue=0.1),
+            dict(type='ImageRandomHorizontalFlip'),
+            dict(type='ImageNormalize'),
+            dict(
+                type='RandomRotate',
+                angle=[-1, 1],
+                axis='z',
+                center=[0, 0, 0],
+                p=0.5),
+            dict(type='RandomScale', scale=[0.9, 1.1]),
+            dict(type='RandomFlip', p=0.5),
+            dict(type='RandomJitter', sigma=0.005, clip=0.02),
+            dict(
+                type='GridSample',
+                grid_size=0.05,
+                hash_type='fnv',
+                mode='train',
+                keys=('coord', 'strength', 'segment', 'image_coord',
+                      'image_mask'),
+                return_grid_coord=True),
+            dict(type='ToTensor'),
+            dict(
+                type='Collect',
+                keys=('coord', 'grid_coord', 'segment', 'image', 'image_coord',
+                      'image_mask'),
+                feat_keys=('coord', 'strength'))
+        ],
+        test_mode=False,
+        ignore_index=-1,
+        loop=1),
+    val=dict(
+        type='NuScenesDataset',
+        split='val',
+        data_root='data/nuscenes',
+        with_images=True,
+        transform=[
+            dict(type='ImageResize', size=[378, 672]),
+            dict(type='ImageNormalize'),
+            dict(
+                type='GridSample',
+                grid_size=0.05,
+                hash_type='fnv',
+                mode='train',
+                keys=('coord', 'strength', 'segment', 'image_coord',
+                      'image_mask'),
+                return_grid_coord=True),
+            dict(type='ToTensor'),
+            dict(
+                type='Collect',
+                keys=('coord', 'grid_coord', 'segment', 'image', 'image_coord',
+                      'image_mask'),
+                feat_keys=('coord', 'strength'))
+        ],
+        test_mode=False,
+        ignore_index=-1),
+    test=dict(
+        type='NuScenesDataset',
+        split='val',
+        data_root='data/nuscenes',
+        with_images=True,
+        transform=[
+            dict(type='ImageResize', size=[378, 672]),
+            dict(type='Copy', keys_dict=dict(segment='origin_segment')),
+            dict(
+                type='GridSample',
+                grid_size=0.025,
+                hash_type='fnv',
+                mode='train',
+                keys=('coord', 'strength', 'segment', 'image_coord',
+                      'image_mask'),
+                return_inverse=True)
+        ],
+        test_mode=True,
+        test_cfg=dict(
+            voxelize=dict(
+                type='GridSample',
+                grid_size=0.05,
+                hash_type='fnv',
+                mode='test',
+                return_grid_coord=True,
+                keys=('coord', 'strength', 'image_coord', 'image_mask')),
+            crop=None,
+            post_transform=[
+                dict(type='ImageNormalize'),
+                dict(type='ToTensor'),
+                dict(
+                    type='Collect',
+                    keys=('coord', 'grid_coord', 'index', 'image',
+                          'image_coord', 'image_mask'),
+                    feat_keys=('coord', 'strength'))
+            ],
+            aug_transform=[[{
+                'type': 'RandomScale',
+                'scale': [0.9, 0.9]
+            }], [{
+                'type': 'RandomScale',
+                'scale': [0.95, 0.95]
+            }], [{
+                'type': 'RandomScale',
+                'scale': [1, 1]
+            }], [{
+                'type': 'RandomScale',
+                'scale': [1.05, 1.05]
+            }], [{
+                'type': 'RandomScale',
+                'scale': [1.1, 1.1]
+            }],
+                           [{
+                               'type': 'RandomScale',
+                               'scale': [0.9, 0.9]
+                           }, {
+                               'type': 'RandomFlip',
+                               'p': 1
+                           }],
+                           [{
+                               'type': 'RandomScale',
+                               'scale': [0.95, 0.95]
+                           }, {
+                               'type': 'RandomFlip',
+                               'p': 1
+                           }],
+                           [{
+                               'type': 'RandomScale',
+                               'scale': [1, 1]
+                           }, {
+                               'type': 'RandomFlip',
+                               'p': 1
+                           }],
+                           [{
+                               'type': 'RandomScale',
+                               'scale': [1.05, 1.05]
+                           }, {
+                               'type': 'RandomFlip',
+                               'p': 1
+                           }],
+                           [{
+                               'type': 'RandomScale',
+                               'scale': [1.1, 1.1]
+                           }, {
+                               'type': 'RandomFlip',
+                               'p': 1
+                           }]]),
+        ignore_index=-1))

nuscenes/semseg-ptv3_dino-g/events.out.tfevents.1731419596.n23g0016.hpc.itc.rwth-aachen.de

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57a9b7d94e4fbee4a5103bff83c776e29e45a2061c36a7feede432a2ac035196
+size 11464457

nuscenes/semseg-ptv3_dino-g/model/model_best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ac78b93852fb8444498a473a6ea91e91616dd222a620fb49ba2e7d2912e0a99
+size 563967344

nuscenes/semseg-ptv3_dino-g/model/model_last.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18431860100702507669d418d34d2913c513cd5c81b579d1ceb17543afe8bfd6
+size 563967344

nuscenes/semseg-ptv3_dino-g/train.log

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9be3187a6a54ab2963a102e73c85b8fd7bc9e47f530ca4208473780ab64b5ae
+size 41951454