mmFormer
mmFormer copied to clipboard
[MICCAI 2022] The official code for "mmFormer: Multimodal Medical Transformer for Incomplete Multimodal Learning of Brain Tumor Segmentation"
作者您好,想实现您论文中的代码,发现数据在process.py文件中进行了裁剪处理,并且在网络的输出后,输出大小不一,请问如何恢复到裁剪前240x240x155的数据大小,并保存为nii.gz文件,由于无法统一数据大小,无法对应到原始flair模态上进行可视化处理,并且与ground truth有数据偏移,请问作者如何解决这个问题?
Congratulations on your great work. However, I have a question why did you choose to apply delta for the missing modality in the middle of the network rather than at...
作者您好,我想复现您的mmformer代码,但是我没有看明白您的代码中 mmformer/data/dataset_nii.py里面是怎么把训练集和测试集划分开来的,可以请您给我简单解释一下吗
关于复现结果
请问您开源的代码能够复现出您提供的pth文件的效果吗,因为我测试发现您提供的log和pth的结果不一致,并且我自己复现的结果也是和log中差不多,达不到pth的结果(主要是enhancing的结果差很多),请问您是在其他地方加了什么trick吗?
作者您好,我想请教您论文中的标签和预测结果可视化是怎么做的呢,是用哪些数据和方法进行可视化的呢
想问一下是单卡还是多卡呢,我看论文是单卡17G,我用RTX3090 24G单卡显示爆显存,是为什么呢?能解答一下吗
Traceback (most recent call last): File "/media/jxust/磁盘2/mmFormer-main/mmformer/train.py", line 248, in main() File "/media/jxust/磁盘2/mmFormer-main/mmformer/train.py", line 166, in main fuse_pred, sep_preds, prm_preds = model(x, mask) File "/home/jxust/anaconda3/envs/nnformer/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result...
Traceback (most recent call last): File "/media/jxust/磁盘2/mmFormer-main/mmformer/train.py", line 248, in main() File "/media/jxust/磁盘2/mmFormer-main/mmformer/train.py", line 166, in main fuse_pred, sep_preds, prm_preds = model(x, mask) File "/home/jxust/anaconda3/envs/nnformer/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result...