Issues
- 0
Why are DP and DDP used together?
#211 opened by ailinxiqingcheng - 1
- 1
对于模型v1,使用的损失函数是loss = (losses["loss"] * weights + losses['loss_cal'] * 10).mean()?以及在segmentation_sample.py文件中,debug时,输出的第一张图片(ss)是怎么的来的,以及使用的是前面的损失函数吗? 输出的最后一张图片(c)是怎么的来的 是使用loss_cal的损失函数计算得来的吗? 希望有人能解答一下
#209 opened by LiuYM00 - 3
這個github是MedSegDiff-V2還是v1?
#188 opened by henrychou1233 - 10
- 1
- 4
Help!RuntimeError: a leaf Variable that requires grad is being used in an in-place operation.
#203 opened by doqizuo - 0
btcvloader
#207 opened by dulicui742 - 0
How to use dim
#206 opened by Hxxxz0 - 1
sample得到预测的噪声后怎么去除噪声呢?
#205 opened by qxxfd - 1
为什么训练出现了完全噪声的结果,如图所示的原图和标签的训练对,20000+图片对(单通道灰度图)训练200epoch得到如图结果,sample_step=30,是否有朋友知道是什么原因呢
#204 opened by qxxfd - 3
- 0
Brats dataset bug!!!
#202 opened by 1580827935 - 1
What are these losses calculated separately
#176 opened by QijingGJ - 5
Sample result is not a binary mask
#170 opened by YasuoFly - 13
assert x_t.shape == eps.shape
#201 opened by 2039551625 - 0
Poor performance using dpm-solver
#200 opened by wk19990213 - 7
BRATS model generating only black Samples
#155 opened by qahsinahk - 0
Error while loading saved checkpoint in V1
#193 opened by pedropesserl - 0
Errors while sampling with custom dataset
#199 opened by pedropesserl - 0
Model Efficiency in the v2 paper
#198 opened by YunduanLou - 1
help! run the model on my own dataset, visdom show nothing. and have this erro: WARNING:root:Setting up a new session... [W socket.cpp:697] [c10d] The client socket has failed to connect to [DESKTOP-BP72V]:55472 (system error: 10049 - �����������У�������ĵ�ַ��Ч��).
#186 opened by fayeGou - 0
- 0
- 1
Meaning of the high_way parameter
#187 opened by chinmay5 - 2
The generated image is black
#192 opened by haoli-maker - 4
which .pt should I use to sample?
#182 opened by take-off-xyz - 1
How do I stop running code
#191 opened by helianggit - 3
Where is the code of SS-former
#181 opened by Yutong0429Hou - 0
About resuming training process
#190 opened by kaneyxx - 0
- 2
some question about V1 version
#177 opened by tensorctn - 3
How long will you use to train the model?
#183 opened by sonnystalin - 1
- 0
How to generate images of intermediate diffusion steps?
#162 opened by zxk72 - 0
DDIT dataloaders / config
#175 opened by JonnyScream - 0
Why is the effect of using dpm-solver not good?
#174 opened by zxk72 - 1
这个程序的训练模块要运行多少时间?
#173 opened by LaneXu - 0
pretraining model
#171 opened by Taocunguo - 0
- 1
how to deal with StopIteration
#167 opened by Sillyyk - 1
three model.pt
#168 opened by Sillyyk - 1
- 2
- 0
for Dice score = 0
#163 opened by CNGaoWenbo - 0
May I ask how the BRTAS data is organized, especially slice, and the model I trained after decompressing the original data finally gets all black samples
#161 opened by Taocunguo - 2
CUDA out of memory caused by using dpm-solver
#157 opened by VLAD-KONATA - 2
Dataloader about BraTS2020
#156 opened by zzwFograinbow - 1
BRATS sampling Assertion error
#154 opened by qahsinahk - 4
Image Size 256 problem/ Evaluation problem
#153 opened by qahsinahk