woxuankai/spatialalignmentnetwork

code for Multi-Modal MRI Reconstruction Assisted with Spatial Alignment Network

spatialalignmentnetwork's Issues

about Prerequisites

作者您好，请问能否告知您的实验环境和相关配置呢？万分感谢

训练过程问题

作者您好，代码里的随机种子和时间相关，那您的结果是进行了多次实验取平均了吗？如果是的话，具体是多少次呢？

t1_t2_paired_6875_{train,val,test}.csv

   Hello, according to the CSV file you gave me, I have some problems in selecting the T1T2 data pair of the training set. 
   For example, the training data pair you gave me has folder 794098270851, but actually the DICOM format of folder 794098270851 has not been converted into NII files of T1 and T2. Therefore, NII cannot be changed into H5 file. 
   May I ask how to solve this error? Looking forward to your reply, thank you!

关于环境

请问运行环境是什么，比如pytorch的版本

The problem of GPU

Hello, I use NVIDIA GeForce GTX 1080 Ti to run your first data set, and found that it takes a very long time, showing that it took 100 days. Could you tell me the type of gpu you used and how long it takes you? Thank you!

The parameter 'range' is deprecated since 0.12 and will be removed in 0.14. Please use 'value_range' instead.
"The parameter 'range' is deprecated since 0.12 and will be removed in 0.14. "
epoch: 0%|

关于数据集

作者你好，
请问除了MRI DICOM Brain数据集，你还有没有尝试用其他数据集训练，或者你还知不知道还有没有MR的数据集存在这样模态之间配对不准确的问题？我想用其他数据集做做试验。

数据问题

作者您好，冒昧打扰您了，我想问一下，你在文档中所说的解压数据文件“Unzip fastMRI brain DICOM to fastMRI_brain_DICOM folder.”的那个数据在哪里呢，没有看见压缩包呀？

checkpoints are missing

hello,

So nice of you to share your work.
Could you also provide the checkpoints in T1_4xRandomT2_GANOnly and T1_4xRandomT2_GANOnly as mentioned in README.md?

thx a lot.

Best Regards,
kingaza