Himashi Amanda Peiris

icon indicating a website link https://himashi92.github.io/ icon indicating an email [email protected]

icon company Monash University icon location Clayton icon location Postdoctoral Research Fellow @ Monash Biomedical Imaging | Former Senior Software Engineer @ Pearson

Results 6 comments of Himashi Amanda Peiris

Still overflowing GPU VRAM with reduced batch size

Hi, you cannot load both models at once for training. Can you check whether any pids are listed under your GPU? Try this command : sudo fuser -v /dev/nvidia0

Still overflowing GPU VRAM with reduced batch size

What is the dataset you are trying to train here? If its brats, then input channel size is 4.

Still overflowing GPU VRAM with reduced batch size

Hi, the current version of VTUNet can only handle the input size of Cx128x128x128. Therefore, the network architecture should be modified accordingly to process the input with the size of...

task 4 issue

what is the input size of your scan? It should be 128x128x128. At the moment, VT-UNet can only process an input size of 128x128x128.

Request for Baseline Training Code

All the reproducible hyperparameters and the model weights are mentioned in the repo. To reproduce LA 10%, use this: python train_cobionet_semi.py --dataset_name LA --labelnum 8 --lamda 0.7 --consistency 1.0 --mu...

Request for Baseline Training Code

The code for this specific experiment is not in this repo. However, you can modify the code given here to train VNet using only 10% of LA data: https://github.com/himashi92/Co-BioNet/blob/main/code/train_cobionet_sup.py.