A data_dict is a dictionary containing properties of a batched point cloud. It should contain the following properties for PTv3: 1. "feat": feature of point cloud 2. "grid_coord": discrete coordinate after grid sampling (voxelization) or "coord" + "grid_size"
(self, data_dict)
| 964 | self.dec.add(module=dec, name=f"dec{s}") |
| 965 | |
| 966 | def forward(self, data_dict): |
| 967 | """ |
| 968 | A data_dict is a dictionary containing properties of a batched point cloud. |
| 969 | It should contain the following properties for PTv3: |
| 970 | 1. "feat": feature of point cloud |
| 971 | 2. "grid_coord": discrete coordinate after grid sampling (voxelization) or "coord" + "grid_size" |
| 972 | 3. "offset" or "batch": https://github.com/Pointcept/Pointcept?tab=readme-ov-file#offset |
| 973 | """ |
| 974 | point = Point(data_dict) |
| 975 | point.serialization(order=self.order, shuffle_orders=self.shuffle_orders) |
| 976 | point.sparsify() |
| 977 | |
| 978 | point = self.embedding(point) |
| 979 | point = self.enc(point) |
| 980 | if not self.cls_mode: |
| 981 | point = self.dec(point) |
| 982 | return point |
nothing calls this directly
no test coverage detected