在统一采样器中如何生成尺寸?我尝试调试图像大小,似乎它可以在某些迭代中使用,而对其他迭代则无效。任何想法如何解决此问题。我的配置如下:
[自定义]
num_classes:14
output_prob:是
label_normalisation:是
softmax:正确
最小采样率:0
强制性标签:(0,1)
rand_samples:0
min_numb_labels:1
proba_connect:正确
evaluation_units:前景
图片:(“图片”,)
标签:('label',)
重量:()
采样器:()
名称:net_segment
[CONFIG_FILE]
[图像]
csv_file:
搜索路径:/ home / ubuntu / med_deacthalon / Task_all_same_names / imagesTr_1
文件名包含:()
文件名不包含:('肺',)
interp_order:3
loader:无
pixdim:(1.0,1.0,1.0)
axcodes :(“ A”,“ R”,“ S”)
spatial_window_size:(51,51,51)
[LABEL]
-csv_file:
搜索路径:/ home / ubuntu / med_deacthalon / Task_all_same_names / labelsTr_1
文件名包含:()
文件名不包含:('肺',)
interp_order:3
loader:无
pixdim:(1.0,1.0,1.0)
axcodes :(“ A”,“ R”,“ S”)
spatial_window_size:(9,9,9)
[系统]
cuda_devices:“”
num_threads:2
num_gpus:1
model_dir:/ home / ubuntu / models_nifty / deepmedic / all_task_same_name_rename_labels
dataset_split_file:./dataset_split.csv
动作:训练
[网络]
名称:Deepmedic
激活功能:relu
批处理大小:32
衰变:0.0
reg_type:L2
volume_padding_size:(21,21,21)
volume_padding_mode:最小
window_sampling:统一
队列长度:128
multimod_foreground_type:和
histogram_ref_file:histogram_standardisation_alltask.txt
norm_type:百分位数
临界值:(0.01,0.99)
前景类型:otsu_plus
归一化:错误
美白:是
normalise_foreground_only:正确
weight_initializer:he_normal
bias_initializer:零
keep_prob:1.0
weight_initializer_args:{}
bias_initializer_args:{}
[培训]
优化器:亚当
sample_per_volume:32
rotation_angle:(-10.0,10.0)
rotation_angle_x:()
rotation_angle_y:()
rotation_angle_z:()
scaling_percentage:(-10.0,10.0)
random_flipping_axes:-1
do_elastic_deformation:False
num_ctrl_points:4
deformation_sigma:15
比例变形:0.5
lr:0.001
loss_type:骰子
starting_iter:0
save_every_n:45
tensorboard_every_n:20
max_iter:10
max_checkpoints:20
validation_every_n:-1
validation_max_iter:1
exclude_fraction_for_validation:0.0
exclude_fraction_for_inference:0.0
[推断]
spatial_window_size:(57,57,57)
inference_iter:-1
dataset_to_infer:
save_seg_dir:./deepmedic/alltask_newname
output_postfix:_niftynet_out
output_interp_order:0
边界:(36,36,36)
CRITICAL:niftynet: Don't know how to generate sampling locations: Spatial dimensions of the grouped input sources are not consistent. {(477, 451, 187), (391, 369, 147)}
Exception in thread Thread-2:
Traceback (most recent call last):
File "/home/ubuntu/anaconda3/envs/python3/lib/python3.6/threading.py", line 916, in _bootstrap_inner
self.run()
File "/home/ubuntu/anaconda3/envs/python3/lib/python3.6/threading.py", line 864, in run
self._target(*self._args, **self._kwargs)
File "/home/ubuntu/niftynet/NiftyNet/niftynet/engine/image_window_buffer.py", line 148, in _push
for output_dict in self():
File "/home/ubuntu/niftynet/NiftyNet/niftynet/engine/sampler_uniform.py", line 81, in layer_op
self.window.n_samples)
File "/home/ubuntu/niftynet/NiftyNet/niftynet/engine/sampler_uniform.py", line 151, in _spatial_coordinates_generator
_infer_spatial_size(img_sizes, win_sizes)
File "/home/ubuntu/niftynet/NiftyNet/niftynet/engine/sampler_uniform.py", line 238, in _infer_spatial_size
raise NotImplementedError
NotImplementedError
答案 0 :(得分:0)
此问题已解决:https://github.com/NifTK/NiftyNet/issues/170
在摘要中,当在配置文件中设置pixdim
时,图像和标签的标题中应存储相同的体素间距。