GCNet

GCNet copied to clipboard

re-training 3dmatch

1

I tried retraining 3dmatch database, and get following log for val after about 20 hours, seems loss and recall not improved significantly, is this right?  Thanks

x1y9

About creating my own data set

2

I want to add my own dataset based on the 3DMatch dataset. How do I create it? In 3DMatch data file, the training set and the test set are both...

jiedia

qsisi

Hi, Thanks for sharing the codebase for your work. I am trying to train the network on custom data, but I got the following error： /pytorch/aten/src/ATen/native/cuda/Loss.cu:102: operator(): block: [1,0,0], thread:...

stepwhal

The trainning error

2

dear sir when I trainning the model ,i meet the error. as follows. @zhulf0804 @qsisi  can you know how to solve it? Thanks

Kris-Xu96

result=o3d.pipelines.registration.registration_ransac_based_on_feature_matching

4

result=o3d.pipelines.registration.registration_ransac_based_on_feature_matching( TypeError: registration_ransac_based_on_feature_matching(): incompatible function arguments. The following argument types are supported:

Zaizai123-beep

How long does it take to retrain the 3D match dataset

1

I retrained the 3D match dataset and it takes about 3 hours per epoch on RTX 4090. Is it normal for me to need about 5 days to complete the...

wzs-go

Why is my training speed on the 4090 not as fast as that on the 3090?

WeridCloud