How to replace your model with my own model to train and test on these data sets?

loicland / superpoint_graph

Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs

MIT License

758 stars 214 forks source link

How to replace your model with my own model to train and test on these data sets? #213

Closed longmalongma closed 4 years ago

longmalongma commented 4 years ago

Thanks for your great work, can you give me some advice on how to replace your model with my own model to train and test on these data sets?for example semantic3d dataset．

LeanderWayne commented 4 years ago

there is instructions in README.md about custom datasets, sementic3d.md about sementic3d dataset. As for sementic3d, the author has already test it, and there is more details in the essay.

loicland commented 4 years ago

Semantic3d is already implemented. For your own data there might be a little bit of work, but I give pointers at the end of the main README as well as a template custom_dataset.py wrapper inn /learning/.