Fix user warnings from different reasons

DeDoDe/encoder.py

@@ -6,7 +6,10 @@
 class VGG19(nn.Module):
     def __init__(self, pretrained=False, amp = False, amp_dtype = torch.float16) -> None:
         super().__init__()
-        self.layers = nn.ModuleList(tvm.vgg19_bn(pretrained=pretrained).features[:40])
+        vgg_kwargs = {}
+        if not pretrained:
+            vgg_kwargs["weights"] = None
+        self.layers = nn.ModuleList(tvm.vgg19_bn(**vgg_kwargs).features[:40])
         # Maxpool layers: 6, 13, 26, 39
         self.amp = amp
         self.amp_dtype = amp_dtype
@@ -25,12 +28,15 @@ def forward(self, x, **kwargs):
 class VGG(nn.Module):
     def __init__(self, size = "19", pretrained=False, amp = False, amp_dtype = torch.float16) -> None:
         super().__init__()
+        vgg_kwargs = {}
+        if not pretrained:
+            vgg_kwargs["weights"] = None
         if size == "11":
-            self.layers = nn.ModuleList(tvm.vgg11_bn(pretrained=pretrained).features[:22])
+            self.layers = nn.ModuleList(tvm.vgg11_bn(**vgg_kwargs).features[:22])
         elif size == "13": 
-            self.layers = nn.ModuleList(tvm.vgg13_bn(pretrained=pretrained).features[:28])
+            self.layers = nn.ModuleList(tvm.vgg13_bn(**vgg_kwargs).features[:28])
         elif size == "19": 
-            self.layers = nn.ModuleList(tvm.vgg19_bn(pretrained=pretrained).features[:40])
+            self.layers = nn.ModuleList(tvm.vgg19_bn(**vgg_kwargs).features[:40])
         # Maxpool layers: 6, 13, 26, 39
         self.amp = amp
         self.amp_dtype = amp_dtype

DeDoDe/transformer/layers/attention.py

@@ -9,6 +9,7 @@
 #   https://github.com/rwightman/pytorch-image-models/tree/master/timm/models/vision_transformer.py
 
 import logging
+import warnings
 
 from torch import Tensor
 from torch import nn
@@ -22,7 +23,7 @@
 
     XFORMERS_AVAILABLE = True
 except ImportError:
-    logger.warning("xFormers not available")
+    warnings.warn("xFormers not available")
     XFORMERS_AVAILABLE = False
 
 

DeDoDe/transformer/layers/block.py

@@ -10,6 +10,7 @@
 
 import logging
 from typing import Callable, List, Any, Tuple, Dict
+import warnings
 
 import torch
 from torch import nn, Tensor
@@ -29,7 +30,7 @@
 
     XFORMERS_AVAILABLE = True
 except ImportError:
-    logger.warning("xFormers not available")
+    warnings.warn("xFormers not available")
     XFORMERS_AVAILABLE = False
 
 

DeDoDe/utils.py

@@ -69,12 +69,13 @@ def estimate_pose(kpts0, kpts1, K0, K1, norm_thresh, conf=0.99999):
 
 def get_grid(B,H,W, device = get_best_device()):
     x1_n = torch.meshgrid(
-    *[
-        torch.linspace(
-            -1 + 1 / n, 1 - 1 / n, n, device=device
-        )
-        for n in (B, H, W)
-    ]
+        *[
+            torch.linspace(
+                -1 + 1 / n, 1 - 1 / n, n, device=device
+            )
+            for n in (B, H, W)
+        ],
+        indexing = "ij",
     )
     x1_n = torch.stack((x1_n[2], x1_n[1]), dim=-1).reshape(B, H * W, 2)
     return x1_n
@@ -217,7 +218,8 @@ def get_gt_warp(depth1, depth2, T_1to2, K1, K2, depth_interpolation_mode = 'bili
                     -1 + 1 / n, 1 - 1 / n, n, device=depth1.device
                 )
                 for n in (B, H, W)
-            ]
+            ],
+            indexing = "ij",
         )
         x1_n = torch.stack((x1_n[2], x1_n[1]), dim=-1).reshape(B, H * W, 2)
         mask, x2 = warp_kpts(
@@ -670,7 +672,7 @@ def homog_transform(Homog, x):
     return y
 
 def get_homog_warp(Homog, H, W, device = get_best_device()):
-    grid = torch.meshgrid(torch.linspace(-1+1/H,1-1/H,H, device = device), torch.linspace(-1+1/W,1-1/W,W, device = device))
+    grid = torch.meshgrid(torch.linspace(-1+1/H,1-1/H,H, device = device), torch.linspace(-1+1/W,1-1/W,W, device = device), indexing = "ij")
 
     x_A = torch.stack((grid[1], grid[0]), dim = -1)[None]
     x_A_to_B = homog_transform(Homog, x_A)