Hi
I installed the latest branch of BiaPy from github and ran a workflow for detection.
I got the following error:
[06:59:18.885498] ##############################
[06:59:18.889505] # PREPARE TRAIN GENERATORS #
[06:59:18.890497] ##############################
[06:59:18.892500] Initializing train data generator . . .
[06:59:18.908179] Checking which channel of the mask needs normalization . . .
[06:59:20.994035] Normalization config used for X: {'type': 'div', 'orig_dtype': dtype('uint16'), 'reduced_uint16': 1}
[06:59:20.995028] Normalization config used for Y: as_mask
[06:59:20.999033] Initializing val data generator . . .
Traceback (most recent call last):
File "C:\Users\rajasekhar.p\BiaPy\main.py", line 51, in <module>
_biapy.run_job()
File "C:\Users\rajasekhar.p\BiaPy\biapy\_biapy.py", line 412, in run_job
self.train()
File "C:\Users\rajasekhar.p\BiaPy\biapy\_biapy.py", line 152, in train
self.workflow.train()
File "C:\Users\rajasekhar.p\BiaPy\biapy\engine\base_workflow.py", line 492, in train
self.prepare_train_generators()
File "C:\Users\rajasekhar.p\BiaPy\biapy\engine\base_workflow.py", line 315, in prepare_train_generators
self.num_training_steps_per_epoch = create_train_val_augmentors(self.cfg, self.X_train, self.Y_train,
File "C:\Users\rajasekhar.p\BiaPy\biapy\data\generators\__init__.py", line 214, in create_train_val_augmentors
val_generator = f_name(**dic)
File "C:\Users\rajasekhar.p\BiaPy\biapy\data\generators\pair_data_3D_generator.py", line 24, in __init__
super().__init__(**kwars)
File "C:\Users\rajasekhar.p\BiaPy\biapy\data\generators\pair_base_data_generator.py", line 416, in __init__
if _X.ndim != (self.ndim+2):
AttributeError: 'NoneType' object has no attribute 'ndim'
This was fixed when I installed from the branch associated with the released version of 3.3.15. I am using a Windows 11 PC.
I used the default 3d_detection.yml configuration file from Biapy and only added following augmentation:
Thank you so much for catching the error. I made a mistake in recent changes on that branch. The problem now is solved in the last commit. Feel free to close the issue!
Hi I installed the latest branch of BiaPy from github and ran a workflow for detection.
I got the following error:
This was fixed when I installed from the branch associated with the released version of 3.3.15. I am using a Windows 11 PC. I used the default 3d_detection.yml configuration file from Biapy and only added following augmentation:
Cheers Pradeep