TypeError: not all arguments converted during string formatting

[niftk]

TypeError: not all arguments converted during string formatting

[wyli]

Hi @CourantNg for some reason the string (in sampler_resize.py) is parsed from the config file as a user parameter, could you please post the relevant config file section?

Thanks for your quick reply. Following is my config.

[image]
path_to_search = 
filename_contains = volume
filename_not_contains = segmentation
spatial_window_size = (128, 128, 64)
interp_order = 1
axcodes=(A, R, S)

[label]
path_to_search = 
filename_contains = segmentation
filename_not_contains = volume
spatial_window_size = (128, 128, 64)
interp_order = 0
axcodes=(A, R, S)

############################## system configuration sections
[SYSTEM]
cuda_devices = ""
num_threads = 2
num_gpus = 1
model_dir = 

[NETWORK]
name = vnet
activation_function = prelu
batch_size = 2

# volume level preprocessing
volume_padding_size = 0
window_sampling = resize
normalisation = False
whitening = False

queue_length = 6


[TRAINING]
sample_per_volume = 4
lr = 0.00002
loss_type = Dice
starting_iter = 0
save_every_n = 400
max_iter = 10001
max_checkpoints = 20
tensorboard_every_n = 10

validation_every_n = 50
exclude_fraction_for_validation = 0.025
exclude_fraction_for_inference = 0.025

[INFERENCE]
border = (5, 5, 5)
save_seg_dir = 
output_interp_order = 0
spatial_window_size = (0, 0, 3)


############################ custom configuration sections
[SEGMENTATION]
image = image
label = label
output_prob = False
num_classes = 2
label_normalisation = True
min_numb_labels = 2
min_sampling_ratio = 0.000001

[wyli]

This bit is suspicious, , was it added for debugging?

E:\Yang\niftynet\engine\sampler_resize.py", line 60, in __init__
    " [-1 for dynamic size]", self.window.shapes, "(in sampler_resize.py)")

removing , "(in sampler_resize.py)" should work.

Dear @wyli
You are right.
Thank you very much for your time and help.