brats17

brats17 copied to clipboard

OOM about training with brats17 data

1

hello, I have some problem about this code, please help me. I use the GPU: v100 16G. but when I try to train the net with brats17 data by your...

Endousan

It seems we are not able to find labels for Brats 2015 dataset from the website https://www.smir.ch/BRATS/Start2015, though the download link claimed so, would you help to point us to...

kevinmtian

unable to locate util package for evaluation.py

arthavit

ERROR (nifti_image_write_hdr_img2): cannot open output file

11

Hello Everyone, I'm facing this error right now.. INFO:niftynet: Restoring parameters from model17/msnet_en32sg_20000.ckpt INFO:niftynet: Restoring parameters from model17/msnet_en32cr_20000.ckpt Data load, 100.0% finished ** ERROR (nifti_image_write_hdr_img2): cannot open output file '/result17/HGG/Brats17_CBICA_ATX_1.nii.gz'...

zaabahmad

Hi, I am trying to train model with brats18 data. If you could please help debugging an error. The error generates while executing train.py file at line "dice = loss.eval(feed_dict...

jeyflyingbird

loss is not declined

1

hello~I'm a student in Tianjin University,recently i am trying to train the model,but i find that the loss is not declined, just between 0.15 and 0.3,so could you tell me...

Echosanmao

net1 output is null

2

Gaojun211

Hi, I have a highly imbalanced data-distributon(3-4% of total voxels represent the object of interest) for my problem. I would like to know how to enable Niftynet's [_balanced_ window sampler](https://github.com/NifTK/NiftyNet/blob/dev/niftynet/engine/sampler_balanced_v2.py)...

pranavkantgaur

I noticed that numerous files overlap in your train_all/test_all config files- wondering if I'm missing something in how these configs are processed but I would think you should not train...

fhenrywells

I am looking to use some of the pretrained models on 2018 data and was wondering if there was a way to procedurally generate a "config18" directory with the relevant...

fhenrywells