Extract finding directories for test data

Summary: Make common functions for finding directories where test data is found, instead of lots of tests using their own `__file__`  while trying to get ./tests/data and the tutorials data.

Reviewed By: nikhilaravi

Differential Revision: D27633701

fbshipit-source-id: 1467bb6018cea16eba3cab097d713116d51071e9

tests/common_testing.py

@@ -9,6 +9,20 @@ import torch
 from PIL import Image
 
 
+def get_tests_dir() -> Path:
+    """
+    Returns Path for the directory containing this file.
+    """
+    return Path(__file__).resolve().parent
+
+
+def get_pytorch3d_dir() -> Path:
+    """
+    Returns Path for the root PyTorch3D directory.
+    """
+    return get_tests_dir().parent
+
+
 def load_rgb_image(filename: str, data_dir: Union[str, Path]):
     filepath = data_dir / filename
     with Image.open(filepath) as raw_image:

tests/test_build.py

@@ -3,7 +3,8 @@ import json
 import os
 import unittest
 from collections import Counter
-from pathlib import Path
+
+from common_testing import get_pytorch3d_dir, get_tests_dir
 
 
 # This file groups together tests which look at the code without running it.
@@ -16,7 +17,7 @@ class TestBuild(unittest.TestCase):
     def test_name_clash(self):
         # For setup.py, all translation units need distinct names, so we
         # cannot have foo.cu and foo.cpp, even in different directories.
-        test_dir = Path(__file__).resolve().parent
+        test_dir = get_tests_dir()
         source_dir = test_dir.parent / "pytorch3d"
 
         stems = []
@@ -30,7 +31,7 @@ class TestBuild(unittest.TestCase):
 
     @unittest.skipIf(in_conda_build, "In conda build")
     def test_copyright(self):
-        test_dir = Path(__file__).resolve().parent
+        test_dir = get_tests_dir()
         root_dir = test_dir.parent
 
         extensions = ("py", "cu", "cuh", "cpp", "h", "hpp", "sh")
@@ -63,8 +64,8 @@ class TestBuild(unittest.TestCase):
     @unittest.skipIf(in_conda_build, "In conda build")
     def test_valid_ipynbs(self):
         # Check that the ipython notebooks are valid json
-        test_dir = Path(__file__).resolve().parent
-        tutorials_dir = test_dir.parent / "docs" / "tutorials"
+        root_dir = get_pytorch3d_dir()
+        tutorials_dir = root_dir / "docs" / "tutorials"
         tutorials = sorted(tutorials_dir.glob("*.ipynb"))
 
         for tutorial in tutorials:

tests/test_io_obj.py

@@ -8,7 +8,11 @@ from pathlib import Path
 from tempfile import NamedTemporaryFile
 
 import torch
-from common_testing import TestCaseMixin
+from common_testing import (
+    TestCaseMixin,
+    get_pytorch3d_dir,
+    get_tests_dir,
+)
 from iopath.common.file_io import PathManager
 from pytorch3d.io import IO, load_obj, load_objs_as_meshes, save_obj
 from pytorch3d.io.mtl_io import (
@@ -475,7 +479,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
         self.assertEqual(actual_file, expected_file)
 
     def test_load_mtl(self):
-        DATA_DIR = Path(__file__).resolve().parent.parent / "docs/tutorials/data"
+        DATA_DIR = get_pytorch3d_dir() / "docs/tutorials/data"
         obj_filename = "cow_mesh/cow.obj"
         filename = os.path.join(DATA_DIR, obj_filename)
         verts, faces, aux = load_obj(filename)
@@ -559,7 +563,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
     def test_load_mtl_texture_atlas_compare_softras(self):
         # Load saved texture atlas created with SoftRas.
         device = torch.device("cuda:0")
-        DATA_DIR = Path(__file__).resolve().parent.parent
+        DATA_DIR = get_pytorch3d_dir()
         obj_filename = DATA_DIR / "docs/tutorials/data/cow_mesh/cow.obj"
         expected_atlas_fname = DATA_DIR / "tests/data/cow_texture_atlas_softras.pt"
 
@@ -590,7 +594,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
         self.assertClose(expected_atlas, aux.texture_atlas, atol=5e-5)
 
     def test_load_mtl_noload(self):
-        DATA_DIR = Path(__file__).resolve().parent.parent / "docs/tutorials/data"
+        DATA_DIR = get_pytorch3d_dir() / "docs/tutorials/data"
         obj_filename = "cow_mesh/cow.obj"
         filename = os.path.join(DATA_DIR, obj_filename)
         verts, faces, aux = load_obj(filename, load_textures=False)
@@ -628,7 +632,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
         self.assertTrue(aux.verts_uvs is None)
 
     def test_load_obj_mlt_no_image(self):
-        DATA_DIR = Path(__file__).resolve().parent / "data"
+        DATA_DIR = get_tests_dir() / "data"
         obj_filename = "obj_mtl_no_image/model.obj"
         filename = os.path.join(DATA_DIR, obj_filename)
         R = 8
@@ -657,7 +661,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
         self.assertEqual(list(aux.material_colors.keys()), ["material_1"])
 
     def test_load_obj_missing_texture(self):
-        DATA_DIR = Path(__file__).resolve().parent / "data"
+        DATA_DIR = get_tests_dir() / "data"
         obj_filename = "missing_files_obj/model.obj"
         filename = os.path.join(DATA_DIR, obj_filename)
         with self.assertWarnsRegex(UserWarning, "Texture file does not exist"):
@@ -672,7 +676,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
         self.assertTrue(torch.allclose(faces.verts_idx, expected_faces))
 
     def test_load_obj_missing_texture_noload(self):
-        DATA_DIR = Path(__file__).resolve().parent / "data"
+        DATA_DIR = get_tests_dir() / "data"
         obj_filename = "missing_files_obj/model.obj"
         filename = os.path.join(DATA_DIR, obj_filename)
         verts, faces, aux = load_obj(filename, load_textures=False)
@@ -688,7 +692,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
         self.assertTrue(aux.texture_images is None)
 
     def test_load_obj_missing_mtl(self):
-        DATA_DIR = Path(__file__).resolve().parent / "data"
+        DATA_DIR = get_tests_dir() / "data"
         obj_filename = "missing_files_obj/model2.obj"
         filename = os.path.join(DATA_DIR, obj_filename)
         with self.assertWarnsRegex(UserWarning, "Mtl file does not exist"):
@@ -703,7 +707,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
         self.assertTrue(torch.allclose(faces.verts_idx, expected_faces))
 
     def test_load_obj_missing_mtl_noload(self):
-        DATA_DIR = Path(__file__).resolve().parent / "data"
+        DATA_DIR = get_tests_dir() / "data"
         obj_filename = "missing_files_obj/model2.obj"
         filename = os.path.join(DATA_DIR, obj_filename)
         verts, faces, aux = load_obj(filename, load_textures=False)
@@ -760,7 +764,7 @@ class TestMeshObjIO(TestCaseMixin, unittest.TestCase):
                         mesh.textures.atlas_padded(), mesh3.textures.atlas_padded()
                     )
 
-        DATA_DIR = Path(__file__).resolve().parent.parent / "docs/tutorials/data"
+        DATA_DIR = get_pytorch3d_dir() / "docs/tutorials/data"
         obj_filename = DATA_DIR / "cow_mesh/cow.obj"
 
         mesh = load_objs_as_meshes([obj_filename])

tests/test_marching_cubes.py

@@ -2,10 +2,9 @@
 import os
 import pickle
 import unittest
-from pathlib import Path
 
 import torch
-from common_testing import TestCaseMixin
+from common_testing import TestCaseMixin, get_tests_dir
 from pytorch3d.ops.marching_cubes import marching_cubes_naive
 
 
@@ -641,7 +640,7 @@ class TestMarchingCubes(TestCaseMixin, unittest.TestCase):
             volume, isolevel=64, return_local_coords=False
         )
 
-        DATA_DIR = Path(__file__).resolve().parent / "data"
+        DATA_DIR = get_tests_dir() / "data"
         data_filename = "test_marching_cubes_data/sphere_level64.pickle"
         filename = os.path.join(DATA_DIR, data_filename)
         with open(filename, "rb") as file:
@@ -677,7 +676,7 @@ class TestMarchingCubes(TestCaseMixin, unittest.TestCase):
             volume = volume.permute(0, 3, 2, 1)  # (B, D, H, W)
             verts, faces = marching_cubes_naive(volume, isolevel=0.001)
 
-            DATA_DIR = Path(__file__).resolve().parent / "data"
+            DATA_DIR = get_tests_dir() / "data"
             data_filename = "test_marching_cubes_data/double_ellipsoid.pickle"
             filename = os.path.join(DATA_DIR, data_filename)
             with open(filename, "rb") as file:

tests/test_points_alignment.py

@@ -1,11 +1,10 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All rights reserved.
 
 import unittest
-from pathlib import Path
 
 import numpy as np
 import torch
-from common_testing import TestCaseMixin
+from common_testing import TestCaseMixin, get_tests_dir
 from pytorch3d.ops import points_alignment
 from pytorch3d.structures.pointclouds import Pointclouds
 from pytorch3d.transforms import rotation_conversions
@@ -40,7 +39,7 @@ class TestICP(TestCaseMixin, unittest.TestCase):
         super().setUp()
         torch.manual_seed(42)
         np.random.seed(42)
-        trimesh_results_path = Path(__file__).resolve().parent / "data/icp_data.pth"
+        trimesh_results_path = get_tests_dir() / "data/icp_data.pth"
         self.trimesh_results = torch.load(trimesh_results_path)
 
     @staticmethod

tests/test_r2n2.py

@@ -9,7 +9,7 @@ from pathlib import Path
 
 import numpy as np
 import torch
-from common_testing import TestCaseMixin, load_rgb_image
+from common_testing import TestCaseMixin, load_rgb_image, get_tests_dir
 from PIL import Image
 from pytorch3d.datasets import (
     R2N2,
@@ -37,7 +37,7 @@ VOXELS_REL_PATH = "ShapeNetVox"
 
 
 DEBUG = False
-DATA_DIR = Path(__file__).resolve().parent / "data"
+DATA_DIR = get_tests_dir() / "data"
 
 
 class TestR2N2(TestCaseMixin, unittest.TestCase):

tests/test_rasterize_rectangle_images.py

@@ -2,11 +2,15 @@
 
 import unittest
 from itertools import product
-from pathlib import Path
 
 import numpy as np
 import torch
-from common_testing import TestCaseMixin, load_rgb_image
+from common_testing import (
+    TestCaseMixin,
+    load_rgb_image,
+    get_pytorch3d_dir,
+    get_tests_dir,
+)
 from PIL import Image
 from pytorch3d.io import load_obj
 from pytorch3d.renderer.cameras import FoVPerspectiveCameras, look_at_view_transform
@@ -42,7 +46,7 @@ from pytorch3d.utils import torus
 
 
 DEBUG = False
-DATA_DIR = Path(__file__).resolve().parent / "data"
+DATA_DIR = get_tests_dir() / "data"
 
 # Verts/Faces for a simple mesh with two faces.
 verts0 = torch.tensor(
@@ -449,7 +453,7 @@ class TestRasterizeRectangleImagesMeshes(TestCaseMixin, unittest.TestCase):
         Test a larger textured mesh is rendered correctly in a non square image.
         """
         device = torch.device("cuda:0")
-        obj_dir = Path(__file__).resolve().parent.parent / "docs/tutorials/data"
+        obj_dir = get_pytorch3d_dir() / "docs/tutorials/data"
         obj_filename = obj_dir / "cow_mesh/cow.obj"
 
         # Load mesh + texture

tests/test_rasterizer.py

@@ -2,10 +2,10 @@
 
 
 import unittest
-from pathlib import Path
 
 import numpy as np
 import torch
+from common_testing import get_tests_dir
 from PIL import Image
 from pytorch3d.renderer.cameras import FoVPerspectiveCameras, look_at_view_transform
 from pytorch3d.renderer.mesh.rasterizer import MeshRasterizer, RasterizationSettings
@@ -17,7 +17,7 @@ from pytorch3d.structures import Pointclouds
 from pytorch3d.utils.ico_sphere import ico_sphere
 
 
-DATA_DIR = Path(__file__).resolve().parent / "data"
+DATA_DIR = get_tests_dir() / "data"
 DEBUG = False  # Set DEBUG to true to save outputs from the tests.
 
 

tests/test_render_meshes.py

@@ -6,11 +6,15 @@ Sanity checks for output images from the renderer.
 """
 import os
 import unittest
-from pathlib import Path
 
 import numpy as np
 import torch
-from common_testing import TestCaseMixin, load_rgb_image
+from common_testing import (
+    TestCaseMixin,
+    load_rgb_image,
+    get_pytorch3d_dir,
+    get_tests_dir,
+)
 from PIL import Image
 from pytorch3d.io import load_obj
 from pytorch3d.renderer.cameras import (
@@ -46,7 +50,7 @@ from pytorch3d.utils.torus import torus
 # If DEBUG=True, save out images generated in the tests for debugging.
 # All saved images have prefix DEBUG_
 DEBUG = False
-DATA_DIR = Path(__file__).resolve().parent / "data"
+DATA_DIR = get_tests_dir() / "data"
 
 
 class TestRenderMeshes(TestCaseMixin, unittest.TestCase):
@@ -384,7 +388,7 @@ class TestRenderMeshes(TestCaseMixin, unittest.TestCase):
         The pupils in the eyes of the cow should always be looking to the left.
         """
         device = torch.device("cuda:0")
-        obj_dir = Path(__file__).resolve().parent.parent / "docs/tutorials/data"
+        obj_dir = get_pytorch3d_dir() / "docs/tutorials/data"
         obj_filename = obj_dir / "cow_mesh/cow.obj"
 
         # Load mesh + texture
@@ -966,7 +970,7 @@ class TestRenderMeshes(TestCaseMixin, unittest.TestCase):
         Also check that the backward pass for texture atlas rendering is differentiable.
         """
         device = torch.device("cuda:0")
-        obj_dir = Path(__file__).resolve().parent.parent / "docs/tutorials/data"
+        obj_dir = get_pytorch3d_dir() / "docs/tutorials/data"
         obj_filename = obj_dir / "cow_mesh/cow.obj"
 
         # Load mesh and texture as a per face texture atlas.

tests/test_render_meshes_clipped.py

@@ -9,12 +9,11 @@ See pytorch3d/renderer/mesh/clip.py for more details about the
 clipping process.
 """
 import unittest
-from pathlib import Path
 
 import imageio
 import numpy as np
 import torch
-from common_testing import TestCaseMixin, load_rgb_image
+from common_testing import TestCaseMixin, load_rgb_image, get_tests_dir
 from pytorch3d.io import save_obj
 from pytorch3d.renderer.cameras import FoVPerspectiveCameras, look_at_view_transform
 from pytorch3d.renderer.lighting import PointLights
@@ -34,7 +33,7 @@ from pytorch3d.structures.meshes import Meshes
 # If DEBUG=True, save out images generated in the tests for debugging.
 # All saved images have prefix DEBUG_
 DEBUG = False
-DATA_DIR = Path(__file__).resolve().parent / "data"
+DATA_DIR = get_tests_dir() / "data"
 
 
 class TestRenderMeshesClipping(TestCaseMixin, unittest.TestCase):

tests/test_render_points.py

@@ -7,11 +7,15 @@ Sanity checks for output images from the pointcloud renderer.
 import unittest
 import warnings
 from os import path
-from pathlib import Path
 
 import numpy as np
 import torch
-from common_testing import TestCaseMixin, load_rgb_image
+from common_testing import (
+    TestCaseMixin,
+    load_rgb_image,
+    get_pytorch3d_dir,
+    get_tests_dir,
+)
 from PIL import Image
 from pytorch3d.renderer.cameras import (
     FoVOrthographicCameras,
@@ -36,7 +40,7 @@ from pytorch3d.utils.ico_sphere import ico_sphere
 # If DEBUG=True, save out images generated in the tests for debugging.
 # All saved images have prefix DEBUG_
 DEBUG = False
-DATA_DIR = Path(__file__).resolve().parent / "data"
+DATA_DIR = get_tests_dir() / "data"
 
 
 class TestRenderPoints(TestCaseMixin, unittest.TestCase):
@@ -216,7 +220,7 @@ class TestRenderPoints(TestCaseMixin, unittest.TestCase):
 
     def test_pointcloud_with_features(self):
         device = torch.device("cuda:0")
-        file_dir = Path(__file__).resolve().parent.parent / "docs/tutorials/data"
+        file_dir = get_pytorch3d_dir() / "docs/tutorials/data"
         pointcloud_filename = file_dir / "PittsburghBridge/pointcloud.npz"
 
         # Note, this file is too large to check in to the repo.

tests/test_sample_points_from_meshes.py

@@ -2,11 +2,15 @@
 
 
 import unittest
-from pathlib import Path
 
 import numpy as np
 import torch
-from common_testing import TestCaseMixin, get_random_cuda_device
+from common_testing import (
+    TestCaseMixin,
+    get_random_cuda_device,
+    get_pytorch3d_dir,
+    get_tests_dir,
+)
 from PIL import Image
 from pytorch3d.io import load_objs_as_meshes
 from pytorch3d.ops import sample_points_from_meshes
@@ -26,7 +30,7 @@ from pytorch3d.utils.ico_sphere import ico_sphere
 # If DEBUG=True, save out images generated in the tests for debugging.
 # All saved images have prefix DEBUG_
 DEBUG = False
-DATA_DIR = Path(__file__).resolve().parent / "data"
+DATA_DIR = get_tests_dir() / "data"
 
 
 class TestSamplePoints(TestCaseMixin, unittest.TestCase):
@@ -261,7 +265,7 @@ class TestSamplePoints(TestCaseMixin, unittest.TestCase):
         Confirm that torch.multinomial does not sample elements which have
         zero probability using a real example of input from a training run.
         """
-        weights = torch.load(Path(__file__).resolve().parent / "weights.pt")
+        weights = torch.load(get_tests_dir() / "weights.pt")
         S = 4096
         num_trials = 100
         for _ in range(0, num_trials):
@@ -378,7 +382,7 @@ class TestSamplePoints(TestCaseMixin, unittest.TestCase):
         # the cow mesh and its texture uv to a pointcloud with texture
 
         device = torch.device("cuda:0")
-        obj_dir = Path(__file__).resolve().parent.parent / "docs/tutorials/data"
+        obj_dir = get_pytorch3d_dir() / "docs/tutorials/data"
         obj_filename = obj_dir / "cow_mesh/cow.obj"
 
         for text_type in ("uv", "atlas"):

tests/test_shapenet_core.py

@@ -4,11 +4,10 @@ Sanity checks for loading ShapeNetCore.
 """
 import os
 import unittest
-from pathlib import Path
 
 import numpy as np
 import torch
-from common_testing import TestCaseMixin, load_rgb_image
+from common_testing import TestCaseMixin, load_rgb_image, get_tests_dir
 from PIL import Image
 from pytorch3d.datasets import ShapeNetCore, collate_batched_meshes
 from pytorch3d.renderer import (
@@ -26,7 +25,7 @@ VERSION = 1
 # If DEBUG=True, save out images generated in the tests for debugging.
 # All saved images have prefix DEBUG_
 DEBUG = False
-DATA_DIR = Path(__file__).resolve().parent / "data"
+DATA_DIR = get_tests_dir() / "data"
 
 
 class TestShapenetCore(TestCaseMixin, unittest.TestCase):