Fix issues with CaDDN (#542)

* Download DeepLabV3 if not available yet

* Fix dtype issue

Fix RuntimeError: expected backend CPU and dtype Float but got backend CPU and dtype Long

pcdet/models/backbones_3d/vfe/image_vfe_modules/f2v/frustum_grid_generator.py

@@ -17,7 +17,7 @@ class FrustumGridGenerator(nn.Module):
         """
         super().__init__()
         self.dtype = torch.float32
-        self.grid_size = torch.as_tensor(grid_size)
+        self.grid_size = torch.as_tensor(grid_size, dtype=self.dtype)
         self.pc_range = pc_range
         self.out_of_bounds_val = -2
         self.disc_cfg = disc_cfg

pcdet/models/backbones_3d/vfe/image_vfe_modules/ffn/ddn/ddn_template.py

 from collections import OrderedDict
+from pathlib import Path
+from torch import hub
 
 import numpy as np
 import torch
@@ -56,6 +58,15 @@ class DDNTemplate(nn.Module):
         # Update weights
         if self.pretrained_path is not None:
             model_dict = model.state_dict()
+            
+            # Download pretrained model if not available yet
+            checkpoint_path = Path(self.pretrained_path)
+            if not checkpoint_path.exists():
+                checkpoint = checkpoint_path.name
+                save_dir = checkpoint_path.parent
+                save_dir.mkdir(parents=True)
+                url = f'https://download.pytorch.org/models/{checkpoint}'
+                hub.load_state_dict_from_url(url, save_dir)
 
             # Get pretrained state dict
             pretrained_dict = torch.load(self.pretrained_path)