Issues with CacheDataset function

[nguyen-peter]

Hello! I have been trying to create a 3D image segmentation model for ct scans. I have the following transforms:

train_transforms = Compose( [ LoadImaged(keys=["image", "label"]), EnsureChannelFirstd(keys=["image", "label"]), Orientationd(keys=["image", "label"], axcodes="RAS"), Spacingd( keys=["image", "label"], pixdim=(1.5, 1.5, 2.0), mode=("bilinear", "nearest"), ), ScaleIntensityRanged( keys=["image"], a_min=-175, a_max=250, b_min=0.0, b_max=1.0, clip=True, ), CropForegroundd(keys=["image", "label"], source_key="image"), RandCropByPosNegLabeld( keys=["image", "label"], label_key="label", spatial_size=(96, 96, 96), pos=1, neg=1, num_samples=4, image_key="image", image_threshold=0, ), RandFlipd( keys=["image", "label"], spatial_axis=[0], prob=0.10, ), RandFlipd( keys=["image", "label"], spatial_axis=[1], prob=0.10, ), RandFlipd( keys=["image", "label"], spatial_axis=[2], prob=0.10, ), RandRotate90d( keys=["image", "label"], prob=0.10, max_k=3, ), RandShiftIntensityd( keys=["image"], offsets=0.10, prob=0.50, ), ] ) val_transforms = Compose( [ LoadImaged(keys=["image", "label"]), EnsureChannelFirstd(keys=["image", "label"]), Orientationd(keys=["image", "label"], axcodes="RAS"), Spacingd( keys=["image", "label"], pixdim=(1.5, 1.5, 2.0), mode=("bilinear", "nearest"), ), ScaleIntensityRanged(keys=["image"], a_min=-175, a_max=250, b_min=0.0, b_max=1.0, clip=True), CropForegroundd(keys=["image", "label"], source_key="image"), ] ) print("complete")

And I am trying to load my data with the following code:

`data_dir = "/panfs/jay/groups/25/barkerfk/nguy4214/data/"
split_json = "dataset_0.json"

datasets = data_dir + split_json
datalist = load_decathlon_datalist(datasets, True, "training")
val_files = load_decathlon_datalist(datasets, True, "validation")
train_ds = CacheDataset(
data=datalist,
transform=train_transforms,
cache_num=50,
cache_rate=1.0,
num_workers=8,
)
train_loader = DataLoader(train_ds, batch_size=1, shuffle=True, num_workers=8, pin_memory=True)
val_ds = CacheDataset(data=val_files, transform=val_transforms, cache_num=6, cache_rate=1.0, num_workers=4)
val_loader = DataLoader(val_ds, batch_size=1, shuffle=False, num_workers=4, pin_memory=True)

print(train_ds)

print(val_files)
print("done")`

My data is a json that looks like this:

dataset_0.json

It appears that it is struggling at CacheDataset: it typically produces the following output:
Loading dataset: 4%|▍ | 1/23 [00:00<00:07, 2.76it/s]/home/barkerfk/nguy4214/.local/lib/python3.7/site-packages/monai/transforms/spatial/array.py:705: UserWarning: axcodes ('RAS') length is smaller than the number of input spatial dimensions D=2. Orientation: input spatial shape is torch.Size([1060, 1085]), num. channels is 1,please make sure the input is in the channel-first format. f"axcodes ('{self.axcodes}') length is smaller than the number of input spatial dimensions D={sr}.\n" Loading dataset: 52%|█████▏ | 12/23 [00:00<00:00, 18.29it/s]

Thank you for any help

[KumoLiu]

Hi @nguyen-peter, the warning message is from Orientationd. If you only use 2D input, you only need to specify two elements in axcodes.

Thanks!