Thank you for your wonderful work! I am running td3d on my dataset,

Not running,about samsunglabs/td3d

Comments (3)

linwei0763 commented on June 13, 2024

Here is the log.

2023-08-03 05:56:37,714 - mmdet - INFO - Environment info:

sys.platform: linux
Python: 3.8.17 (default, Jul 5 2023, 21:04:15) [GCC 11.2.0]
CUDA available: True
GPU 0: NVIDIA A40
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 11.1, V11.1.105
GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
PyTorch: 1.9.1+cu111
PyTorch compiling details: PyTorch built with:

GCC 7.3
C++ Version: 201402
Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications
Intel(R) MKL-DNN v2.1.2 (Git Hash 98be7e8afa711dc9b66c8ff3504129cb82013cdb)
OpenMP 201511 (a.k.a. OpenMP 4.5)
NNPACK is enabled
CPU capability usage: AVX2
CUDA Runtime 11.1
NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86
CuDNN 8.0.5
Magma 2.5.2
Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.1, CUDNN_VERSION=8.0.5, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.9.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON,

TorchVision: 0.10.1+cu111
OpenCV: 4.8.0
MMCV: 1.6.0
MMCV Compiler: GCC 7.5
MMCV CUDA Compiler: 11.1
MMDetection: 2.24.1
MMSegmentation: 0.24.1
MMDetection3D: 1.0.0rc3+fd4b4d4
spconv2.0: False

2023-08-03 05:56:38,730 - mmdet - INFO - Distributed training: False
2023-08-03 05:56:39,700 - mmdet - INFO - Config:
voxel_size = 0.02
padding = 0.08
n_points = 100000
class_names = ('ceiling', 'floor', 'wall', 'beam', 'column', 'window', 'door',
'table', 'chair', 'sofa', 'bookcase', 'board', 'clutter')
model = dict(
type='TD3DInstanceSegmentor',
voxel_size=0.02,
backbone=dict(
type='MinkResNet',
in_channels=3,
depth=34,
norm='batch',
return_stem=True,
stride=1),
neck=dict(
type='NgfcTinySegmentationNeck',
in_channels=(64, 128, 256, 512),
out_channels=128),
head=dict(
type='TD3DInstanceHead',
in_channels=128,
n_reg_outs=6,
n_classes=13,
n_levels=4,
padding=0.08,
voxel_size=0.02,
unet=dict(type='MinkUNet14B', in_channels=32, out_channels=14, D=3),
first_assigner=dict(
type='S3DISAssigner',
top_pts_threshold=6,
label2level=[3, 3, 3, 3, 2, 2, 2, 2, 1, 2, 2, 1, 1]),
second_assigner=dict(type='MaxIoU3DAssigner', threshold=0.25),
roi_extractor=dict(
type='Mink3DRoIExtractor',
voxel_size=0.02,
padding=0.08,
min_pts_threshold=10)),
train_cfg=dict(num_rois=1),
test_cfg=dict(
nms_pre=100, iou_thr=0.2, score_thr=0.15, binary_score_thr=0.2))
optimizer = dict(type='AdamW', lr=0.001, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=10, norm_type=2))
lr_config = dict(policy='step', warmup=None, step=[28, 32])
runner = dict(type='EpochBasedRunner', max_epochs=33)
custom_hooks = [dict(type='EmptyCacheHook', after_iter=True)]
checkpoint_config = dict(interval=1, max_keep_ckpts=50)
log_config = dict(interval=50, hooks=[dict(type='TextLoggerHook')])
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/td3d_is_s3dis-3d-5class'
load_from = None
resume_from = None
workflow = [('train', 1)]
dataset_type = 'S3DISInstanceSegDataset'
data_root = './data/s3dis/'
train_area = [2]
test_area = 1
train_pipeline = [
dict(
type='LoadPointsFromFile',
coord_type='DEPTH',
shift_height=False,
use_color=True,
load_dim=6,
use_dim=[0, 1, 2, 3, 4, 5]),
dict(type='LoadAnnotations3D', with_mask_3d=True, with_seg_3d=True),
dict(type='PointSample', num_points=100000),
dict(
type='PointSegClassMappingV2',
valid_cat_ids=(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12),
max_cat_id=13),
dict(
type='RandomFlip3D',
sync_2d=False,
flip_ratio_bev_horizontal=0.5,
flip_ratio_bev_vertical=0.5),
dict(
type='GlobalRotScaleTrans',
rot_range=[0, 0],
scale_ratio_range=[0.95, 1.05],
translation_std=[0.1, 0.1, 0.1],
shift_height=False),
dict(type='BboxRecalculation'),
dict(type='NormalizePointsColor', color_mean=None),
dict(
type='DefaultFormatBundle3D',
class_names=('ceiling', 'floor', 'wall', 'beam', 'column', 'window',
'door', 'table', 'chair', 'sofa', 'bookcase', 'board',
'clutter')),
dict(
type='Collect3D',
keys=[
'points', 'gt_bboxes_3d', 'gt_labels_3d', 'pts_semantic_mask',
'pts_instance_mask'
])
]
test_pipeline = [
dict(
type='LoadPointsFromFile',
coord_type='DEPTH',
shift_height=False,
use_color=True,
load_dim=6,
use_dim=[0, 1, 2, 3, 4, 5]),
dict(
type='MultiScaleFlipAug3D',
img_scale=(1333, 800),
pts_scale_ratio=1,
flip=False,
transforms=[
dict(type='NormalizePointsColor', color_mean=None),
dict(
type='DefaultFormatBundle3D',
class_names=('ceiling', 'floor', 'wall', 'beam', 'column',
'window', 'door', 'table', 'chair', 'sofa',
'bookcase', 'board', 'clutter'),
with_label=False),
dict(type='Collect3D', keys=['points'])
])
]
data = dict(
samples_per_gpu=4,
workers_per_gpu=6,
train=dict(
type='RepeatDataset',
times=13,
dataset=dict(
type='ConcatDataset',
datasets=[
dict(
type='S3DISInstanceSegDataset',
data_root='./data/s3dis/',
ann_file='./data/s3dis/s3dis_infos_Area_2.pkl',
pipeline=[
dict(
type='LoadPointsFromFile',
coord_type='DEPTH',
shift_height=False,
use_color=True,
load_dim=6,
use_dim=[0, 1, 2, 3, 4, 5]),
dict(
type='LoadAnnotations3D',
with_mask_3d=True,
with_seg_3d=True),
dict(type='PointSample', num_points=100000),
dict(
type='PointSegClassMappingV2',
valid_cat_ids=(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10,
11, 12),
max_cat_id=13),
dict(
type='RandomFlip3D',
sync_2d=False,
flip_ratio_bev_horizontal=0.5,
flip_ratio_bev_vertical=0.5),
dict(
type='GlobalRotScaleTrans',
rot_range=[0, 0],
scale_ratio_range=[0.95, 1.05],
translation_std=[0.1, 0.1, 0.1],
shift_height=False),
dict(type='BboxRecalculation'),
dict(type='NormalizePointsColor', color_mean=None),
dict(
type='DefaultFormatBundle3D',
class_names=('ceiling', 'floor', 'wall', 'beam',
'column', 'window', 'door', 'table',
'chair', 'sofa', 'bookcase', 'board',
'clutter')),
dict(
type='Collect3D',
keys=[
'points', 'gt_bboxes_3d', 'gt_labels_3d',
'pts_semantic_mask', 'pts_instance_mask'
])
],
filter_empty_gt=True,
classes=('ceiling', 'floor', 'wall', 'beam', 'column',
'window', 'door', 'table', 'chair', 'sofa',
'bookcase', 'board', 'clutter'),
box_type_3d='Depth')
],
separate_eval=False)),
val=dict(
type='S3DISInstanceSegDataset',
data_root='./data/s3dis/',
ann_file='./data/s3dis/s3dis_infos_Area_1.pkl',
pipeline=[
dict(
type='LoadPointsFromFile',
coord_type='DEPTH',
shift_height=False,
use_color=True,
load_dim=6,
use_dim=[0, 1, 2, 3, 4, 5]),
dict(
type='MultiScaleFlipAug3D',
img_scale=(1333, 800),
pts_scale_ratio=1,
flip=False,
transforms=[
dict(type='NormalizePointsColor', color_mean=None),
dict(
type='DefaultFormatBundle3D',
class_names=('ceiling', 'floor', 'wall', 'beam',
'column', 'window', 'door', 'table',
'chair', 'sofa', 'bookcase', 'board',
'clutter'),
with_label=False),
dict(type='Collect3D', keys=['points'])
])
],
filter_empty_gt=False,
classes=('ceiling', 'floor', 'wall', 'beam', 'column', 'window',
'door', 'table', 'chair', 'sofa', 'bookcase', 'board',
'clutter'),
test_mode=True,
box_type_3d='Depth'),
test=dict(
type='S3DISInstanceSegDataset',
data_root='./data/s3dis/',
ann_file='./data/s3dis/s3dis_infos_Area_1.pkl',
pipeline=[
dict(
type='LoadPointsFromFile',
coord_type='DEPTH',
shift_height=False,
use_color=True,
load_dim=6,
use_dim=[0, 1, 2, 3, 4, 5]),
dict(
type='MultiScaleFlipAug3D',
img_scale=(1333, 800),
pts_scale_ratio=1,
flip=False,
transforms=[
dict(type='NormalizePointsColor', color_mean=None),
dict(
type='DefaultFormatBundle3D',
class_names=('ceiling', 'floor', 'wall', 'beam',
'column', 'window', 'door', 'table',
'chair', 'sofa', 'bookcase', 'board',
'clutter'),
with_label=False),
dict(type='Collect3D', keys=['points'])
])
],
filter_empty_gt=False,
classes=('ceiling', 'floor', 'wall', 'beam', 'column', 'window',
'door', 'table', 'chair', 'sofa', 'bookcase', 'board',
'clutter'),
test_mode=True,
box_type_3d='Depth'))
gpu_ids = [0]

2023-08-03 05:56:39,701 - mmdet - INFO - Set random seed to 0, deterministic: False
Name of parameter - Initialization information

backbone.conv1.kernel - torch.Size([27, 3, 64]):
Initialized by user-defined init_weights in MinkResNet

backbone.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.conv1.kernel - torch.Size([27, 64, 64]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer1.0.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.conv2.kernel - torch.Size([27, 64, 64]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer1.0.norm2.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.norm2.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.downsample.0.kernel - torch.Size([1, 64, 64]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer1.0.downsample.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.downsample.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.conv1.kernel - torch.Size([27, 64, 64]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer1.1.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.conv2.kernel - torch.Size([27, 64, 64]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer1.1.norm2.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.norm2.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.conv1.kernel - torch.Size([27, 64, 64]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer1.2.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.conv2.kernel - torch.Size([27, 64, 64]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer1.2.norm2.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.norm2.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.conv1.kernel - torch.Size([27, 64, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.0.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.conv2.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.0.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.downsample.0.kernel - torch.Size([1, 64, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.0.downsample.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.downsample.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.conv1.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.1.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.conv2.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.1.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.conv1.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.2.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.conv2.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.2.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.conv1.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.3.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.conv2.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer2.3.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.conv1.kernel - torch.Size([27, 128, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.0.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.conv2.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.0.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.downsample.0.kernel - torch.Size([1, 128, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.0.downsample.1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.downsample.1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.conv1.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.1.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.conv2.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.1.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.conv1.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.2.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.conv2.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.2.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.conv1.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.3.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.conv2.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.3.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.conv1.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.4.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.conv2.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.4.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.conv1.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.5.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.conv2.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer3.5.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.conv1.kernel - torch.Size([27, 256, 512]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer4.0.norm1.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.norm1.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.conv2.kernel - torch.Size([27, 512, 512]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer4.0.norm2.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.norm2.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.downsample.0.kernel - torch.Size([1, 256, 512]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer4.0.downsample.1.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.downsample.1.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.conv1.kernel - torch.Size([27, 512, 512]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer4.1.norm1.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.norm1.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.conv2.kernel - torch.Size([27, 512, 512]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer4.1.norm2.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.norm2.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.conv1.kernel - torch.Size([27, 512, 512]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer4.2.norm1.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.norm1.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.conv2.kernel - torch.Size([27, 512, 512]):
Initialized by user-defined init_weights in MinkResNet

backbone.layer4.2.norm2.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.norm2.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_0.0.kernel - torch.Size([27, 64, 64]):
Initialized by user-defined init_weights in NgfcTinySegmentationNeck

neck.lateral_block_0.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_0.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_0.0.kernel - torch.Size([27, 64, 128]):
Initialized by user-defined init_weights in NgfcTinySegmentationNeck

neck.out_block_0.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_0.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_1.0.kernel - torch.Size([27, 128, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_1.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_1.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_1.0.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in NgfcTinySegmentationNeck

neck.lateral_block_1.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_1.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_1.0.kernel - torch.Size([27, 128, 128]):
Initialized by user-defined init_weights in NgfcTinySegmentationNeck

neck.out_block_1.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_1.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_2.0.kernel - torch.Size([27, 256, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_2.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_2.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_2.0.kernel - torch.Size([27, 256, 256]):
Initialized by user-defined init_weights in NgfcTinySegmentationNeck

neck.lateral_block_2.1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_2.1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_2.0.kernel - torch.Size([27, 256, 128]):
Initialized by user-defined init_weights in NgfcTinySegmentationNeck

neck.out_block_2.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_2.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_3.0.kernel - torch.Size([27, 512, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_3.1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_3.1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_3.0.kernel - torch.Size([27, 512, 128]):
Initialized by user-defined init_weights in NgfcTinySegmentationNeck

neck.out_block_3.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_3.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.upsample_st_4.0.kernel - torch.Size([27, 128, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.upsample_st_4.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.upsample_st_4.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.conv_32_ch.0.kernel - torch.Size([27, 64, 32]):
Initialized by user-defined init_weights in NgfcTinySegmentationNeck

neck.conv_32_ch.1.bn.weight - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.conv_32_ch.1.bn.bias - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.conv0p1s1.kernel - torch.Size([125, 32, 32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn0.bn.weight - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn0.bn.bias - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.conv1p1s2.kernel - torch.Size([8, 32, 32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn1.bn.weight - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn1.bn.bias - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block1.0.conv1.kernel - torch.Size([27, 32, 32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block1.0.norm1.bn.weight - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block1.0.norm1.bn.bias - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block1.0.conv2.kernel - torch.Size([27, 32, 32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block1.0.norm2.bn.weight - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block1.0.norm2.bn.bias - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.conv2p2s2.kernel - torch.Size([8, 32, 32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn2.bn.weight - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn2.bn.bias - torch.Size([32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.conv1.kernel - torch.Size([27, 32, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.conv2.kernel - torch.Size([27, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.norm2.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.norm2.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.downsample.0.kernel - torch.Size([32, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.downsample.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block2.0.downsample.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.conv3p4s2.kernel - torch.Size([8, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn3.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn3.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.conv1.kernel - torch.Size([27, 64, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.downsample.0.kernel - torch.Size([64, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.downsample.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block3.0.downsample.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.conv4p8s2.kernel - torch.Size([8, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn4.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bn4.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.conv1.kernel - torch.Size([27, 128, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.conv2.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.downsample.0.kernel - torch.Size([128, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.downsample.1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block4.0.downsample.1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.convtr4p16s2.kernel - torch.Size([8, 256, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bntr4.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bntr4.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.conv1.kernel - torch.Size([27, 256, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.downsample.0.kernel - torch.Size([256, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.downsample.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block5.0.downsample.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.convtr5p8s2.kernel - torch.Size([8, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bntr5.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bntr5.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.conv1.kernel - torch.Size([27, 192, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.downsample.0.kernel - torch.Size([192, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.downsample.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block6.0.downsample.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.convtr6p4s2.kernel - torch.Size([8, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bntr6.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bntr6.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.conv1.kernel - torch.Size([27, 160, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.downsample.0.kernel - torch.Size([160, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.downsample.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block7.0.downsample.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.convtr7p2s2.kernel - torch.Size([8, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bntr7.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.bntr7.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.conv1.kernel - torch.Size([27, 160, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.downsample.0.kernel - torch.Size([160, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.downsample.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.block8.0.downsample.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.final.kernel - torch.Size([128, 14]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.unet.final.bias - torch.Size([1, 14]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.reg_conv.kernel - torch.Size([128, 6]):
Initialized by user-defined init_weights in TD3DInstanceHead

head.reg_conv.bias - torch.Size([1, 6]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

head.cls_conv.kernel - torch.Size([128, 13]):
Initialized by user-defined init_weights in TD3DInstanceHead

head.cls_conv.bias - torch.Size([1, 13]):
Initialized by user-defined init_weights in TD3DInstanceHead
Name of parameter - Initialization information

backbone.conv1.kernel - torch.Size([27, 3, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.conv1.kernel - torch.Size([27, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.conv2.kernel - torch.Size([27, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.norm2.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.norm2.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.downsample.0.kernel - torch.Size([1, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.downsample.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.0.downsample.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.conv1.kernel - torch.Size([27, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.conv2.kernel - torch.Size([27, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.norm2.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.1.norm2.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.conv1.kernel - torch.Size([27, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.norm1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.norm1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.conv2.kernel - torch.Size([27, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.norm2.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer1.2.norm2.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.conv1.kernel - torch.Size([27, 64, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.downsample.0.kernel - torch.Size([1, 64, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.downsample.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.0.downsample.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.conv1.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.1.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.conv1.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.2.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.conv1.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.norm1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.norm1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.conv2.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.norm2.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer2.3.norm2.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.conv1.kernel - torch.Size([27, 128, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.conv2.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.downsample.0.kernel - torch.Size([1, 128, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.downsample.1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.0.downsample.1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.conv1.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.conv2.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.1.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.conv1.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.conv2.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.2.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.conv1.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.conv2.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.3.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.conv1.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.conv2.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.4.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.conv1.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.norm1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.norm1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.conv2.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.norm2.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer3.5.norm2.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.conv1.kernel - torch.Size([27, 256, 512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.norm1.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.norm1.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.conv2.kernel - torch.Size([27, 512, 512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.norm2.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.norm2.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.downsample.0.kernel - torch.Size([1, 256, 512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.downsample.1.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.0.downsample.1.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.conv1.kernel - torch.Size([27, 512, 512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.norm1.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.norm1.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.conv2.kernel - torch.Size([27, 512, 512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.norm2.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.1.norm2.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.conv1.kernel - torch.Size([27, 512, 512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.norm1.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.norm1.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.conv2.kernel - torch.Size([27, 512, 512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.norm2.bn.weight - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

backbone.layer4.2.norm2.bn.bias - torch.Size([512]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_0.0.kernel - torch.Size([27, 64, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_0.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_0.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_0.0.kernel - torch.Size([27, 64, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_0.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_0.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_1.0.kernel - torch.Size([27, 128, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_1.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_1.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_1.0.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_1.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_1.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_1.0.kernel - torch.Size([27, 128, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_1.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_1.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_2.0.kernel - torch.Size([27, 256, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_2.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_2.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_2.0.kernel - torch.Size([27, 256, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_2.1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.lateral_block_2.1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_2.0.kernel - torch.Size([27, 256, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_2.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_2.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_3.0.kernel - torch.Size([27, 512, 256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_3.1.bn.weight - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.up_block_3.1.bn.bias - torch.Size([256]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_3.0.kernel - torch.Size([27, 512, 128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_3.1.bn.weight - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.out_block_3.1.bn.bias - torch.Size([128]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.upsample_st_4.0.kernel - torch.Size([27, 128, 64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.upsample_st_4.1.bn.weight - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.upsample_st_4.1.bn.bias - torch.Size([64]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor

neck.conv_32_ch.0.kernel - torch.Size([27, 64, 32]):
The value is the same before and after calling init_weights of TD3DInstanceSegmentor