ChangeMamba

ChangeMamba copied to clipboard

something about "infer_Mamba.SCD.py"

1

Hello, I would like to use the BDA model for prediction, but I only see 'infer_MambaBCD.py' ,'infer_MambaSCD.py 'available. Will there be an update to include 'infer_MambaBDA.py'? Thank you very much."

XiaoBaiHhy

SiameseKPConv 和MambaCD对比

3

非常感谢您的指导，成功复现MambaCD而且在2D变化检测达到sota，因为自己刚接触这块领域，关注到在3D提出了SiameseKPConv ，不知道是否针对这两个模型进行过对比呢

pingcong

Questions about issue 45

1

Hi, I'm having the same problem as [issue 45](https://github.com/ChenHongruixuan/MambaCD/issues/45) when I'm working on an SCD task, but I don't see a solution. Firstly, the problem arises on the premise that:...

heikeyuhuajia

compile bug

2

could you please provide your version of python there are some problems when use kernels RuntimeError: The detected CUDA version (11.7) mismatches the version that was used to compile PyTorch...

ZhiyiHe1997

使用MambaBCD_Small_LEVIRCD进行infer结果无图像

4

'--resume', \MambaBCD_Small_LEVIRCD+_F1_0.8825.pth' '--cfg', \MambaCD-master\changedetection\configs\vssm1\vssm_small_224.yaml' 已添加，但是输出结果经过np.argmax(output_1, axis=1)后变成了全0，所有结果都是这样 不添加预训练参数的话，可以生成类似于随机的黑白图像

Si1ive

hello, is the num_classes fixed in MambaBCD, I changed it to 3 or more classes, the last layer keep output '_**(main_clf): Conv2d(128, 2, kernel_size=(1, 1), stride=(1, 1)_**' . I also...

ZhiyiHe1997

Hi author, Sorry to bother you again, but I haven't received a reply from the SMNet author yet, so could you please share the code you implemented? Thank you very...

sasa550

Hello, we were wondering if the pre-trained weights for the BDA task would be released? Thank you!

ruzcko

随机不确定

3

有没有人在训练时遇到每次训练结果不一样的情况，在一开始已经设定了随机种子固定，调试发现，模型在第一次更新参数前是完全一致的，相同的输出产生相同的预测、loss，但是经过一次参数更新后，模型参数和预测就不一样了，这是哪里还有随机的部分没被固定？ def set_seed(random_seed): random.seed(random_seed) # NumPy np.random.seed(random_seed) # PyTorch torch.manual_seed(random_seed) # if torch.cuda.is_available(): torch.cuda.manual_seed(random_seed) torch.cuda.manual_seed_all(random_seed) torch.backends.cudnn.deterministic = True torch.backends.cudnn.benchmark = False

Lyan-ing

Hi @ChenHongruixuan , may i know if the MambaBDA weights will be released soon? Thank you. Ref. to [Issue74](https://github.com/ChenHongruixuan/MambaCD/issues/74)

JoyLinWQ