update docs

Differential Revision: D18047206

fbshipit-source-id: 387070b12a292bd7fbd53b15318f7528a8748652

INSTALL.md

@@ -55,3 +55,10 @@ Note: you may need to rebuild detectron2 after reinstalling a different build of
 	not match the version of cuda you are running with.
 	This happens sometimes when using anaconda.
 
++ "Not compiled with GPU support": make sure
+	```
+	python -c 'import torch; from torch.utils.cpp_extension import CUDA_HOME; print(torch.cuda.is_available(), CUDA_HOME)'
+	```
+	print valid outputs at the time you build detectron2.
+
++ "invalid device function": you build detectron2 with one version of CUDA but run it with a different version.

detectron2/modeling/box_regression.py

@@ -76,7 +76,7 @@ def apply_deltas(self, deltas, boxes):
                 box transformations for the single box boxes[i].
             boxes (Tensor): boxes to transform, of shape (N, 4)
         """
-        assert torch.isfinite(deltas).all().item()
+        assert torch.isfinite(deltas).all().item(), "Box regression deltas become infinite or NaN!"
         boxes = boxes.to(deltas.dtype)
 
         widths = boxes[:, 2] - boxes[:, 0]
@@ -179,7 +179,7 @@ def apply_deltas(self, deltas, boxes):
             boxes (Tensor): boxes to transform, of shape (N, 5)
         """
         assert deltas.shape[1] == 5 and boxes.shape[1] == 5
-        assert torch.isfinite(deltas).all().item()
+        assert torch.isfinite(deltas).all().item(), "Box regression deltas become infinite or NaN!"
 
         boxes = boxes.to(deltas.dtype)
 

docs/tutorials/configs.md

@@ -23,6 +23,8 @@ cfg.merge_from_file("my_cfg.yaml")   # load values from a file
 cfg.merge_from_list(["MODEL.WEIGHTS", "weights.pth"])   # can also load values from a list of str
 ```
 
+To see a list of available configs in detectron2, see [Config References](../modules/config.html#config-references)
+
 
 ### Best Practice with Configs
 

docs/tutorials/models.md

@@ -7,8 +7,10 @@ from detectron2.modeling import build_model
 model = build_model(cfg)  # returns a torch.nn.Module
 ```
 
-To load a existing checkpoint to the model, use
+To load an existing checkpoint to the model, use
 `DetectionCheckpointer(model).load(file_path)`.
+Detectron2 recognizes models in pytorch's `.pth` format, as well as the `.pkl` files
+in our model zoo.
 
 You can use a model by just `outputs = model(inputs)`.
 Next, we explain the inputs/outputs format used by the builtin models in detectron2.