Commit 9baca57

author

Stefan Knegt

authored

Update README.md

1 parent 3ad908b commit 9baca57Copy full SHA for 9baca57

File tree

+14

-0

lines changed

+14

-0

lines changed

Lines changed: 14 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -37,8 +37,22 @@ for epoch in range(epochs):`
`37`	`37`	`## Train on LIDC Dataset`
`38`	`38`	`One of the datasets used in the original paper is the [LIDC dataset](https://wiki.cancerimagingarchive.net). I've preprocessed this data and stored them in 5 .pickle files which you can [download here](https://drive.google.com/drive/folders/1xKfKCQo8qa6SAr3u7qWNtQjIphIrvmd5?usp=sharing). After downloading the files you can load the data as follows:`
`39`	`39`	```
	`40`	`+import torch`
	`41`	`+from torch.utils.data import DataLoader`
	`42`	`+from torch.utils.data.sampler import SubsetRandomSampler`
`40`	`43`	`from load_LIDC_data import LIDC_IDRI`
	`44`	`+`
`41`	`45`	`dataset = LIDC_IDRI(dataset_location = 'insert_path_here')`
	`46`	`+dataset_size = len(dataset)`
	`47`	`+indices = list(range(dataset_size))`
	`48`	`+split = int(np.floor(test_split * dataset_size))`
	`49`	`+np.random.shuffle(indices)`
	`50`	`+train_indices, test_indices = indices[split:], indices[:split]`
	`51`	`+train_sampler = SubsetRandomSampler(train_indices)`
	`52`	`+test_sampler = SubsetRandomSampler(test_indices)`
	`53`	`+train_loader = DataLoader(dataset, batch_size=batch_size, sampler=train_sampler)`
	`54`	`+test_loader = DataLoader(dataset, batch_size=batch_size, sampler=test_sampler)`
	`55`	`+print("Number of training/test patches:", (len(train_indices),len(test_indices)))`
`42`	`56`	```
`43`	`57`	`Combining this with the training code snippet above, you can start training your own Probabilistic UNet.`
`44`	`58`

Comments

(0)