MKL version fix in CI (#1820)

Summary:
Fix for "undefined symbol: iJIT_NotifyEvent" build issue,

Pull Request resolved: https://github.com/facebookresearch/pytorch3d/pull/1820

Differential Revision: D58685326

.circleci/config.yml

@@ -162,6 +162,90 @@ workflows:
     jobs:
       # - main:
       #     context: DOCKERHUB_TOKEN
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda113
+          context: DOCKERHUB_TOKEN
+          cu_version: cu113
+          name: linux_conda_py38_cu113_pyt1120
+          python_version: '3.8'
+          pytorch_version: 1.12.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py38_cu116_pyt1120
+          python_version: '3.8'
+          pytorch_version: 1.12.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda113
+          context: DOCKERHUB_TOKEN
+          cu_version: cu113
+          name: linux_conda_py38_cu113_pyt1121
+          python_version: '3.8'
+          pytorch_version: 1.12.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py38_cu116_pyt1121
+          python_version: '3.8'
+          pytorch_version: 1.12.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py38_cu116_pyt1130
+          python_version: '3.8'
+          pytorch_version: 1.13.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py38_cu117_pyt1130
+          python_version: '3.8'
+          pytorch_version: 1.13.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py38_cu116_pyt1131
+          python_version: '3.8'
+          pytorch_version: 1.13.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py38_cu117_pyt1131
+          python_version: '3.8'
+          pytorch_version: 1.13.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py38_cu117_pyt200
+          python_version: '3.8'
+          pytorch_version: 2.0.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda118
+          context: DOCKERHUB_TOKEN
+          cu_version: cu118
+          name: linux_conda_py38_cu118_pyt200
+          python_version: '3.8'
+          pytorch_version: 2.0.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py38_cu117_pyt201
+          python_version: '3.8'
+          pytorch_version: 2.0.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda118
+          context: DOCKERHUB_TOKEN
+          cu_version: cu118
+          name: linux_conda_py38_cu118_pyt201
+          python_version: '3.8'
+          pytorch_version: 2.0.1
       - binary_linux_conda:
           conda_docker_image: pytorch/conda-builder:cuda118
           context: DOCKERHUB_TOKEN
@@ -247,33 +331,89 @@ workflows:
           python_version: '3.8'
           pytorch_version: 2.3.1
       - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda118
+          conda_docker_image: pytorch/conda-builder:cuda113
           context: DOCKERHUB_TOKEN
-          cu_version: cu118
+          cu_version: cu113
-          name: linux_conda_py38_cu118_pyt240
+          name: linux_conda_py39_cu113_pyt1120
-          python_version: '3.8'
+          python_version: '3.9'
-          pytorch_version: 2.4.0
+          pytorch_version: 1.12.0
       - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
+          conda_docker_image: pytorch/conda-builder:cuda116
           context: DOCKERHUB_TOKEN
-          cu_version: cu121
+          cu_version: cu116
-          name: linux_conda_py38_cu121_pyt240
+          name: linux_conda_py39_cu116_pyt1120
-          python_version: '3.8'
+          python_version: '3.9'
-          pytorch_version: 2.4.0
+          pytorch_version: 1.12.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda113
+          context: DOCKERHUB_TOKEN
+          cu_version: cu113
+          name: linux_conda_py39_cu113_pyt1121
+          python_version: '3.9'
+          pytorch_version: 1.12.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py39_cu116_pyt1121
+          python_version: '3.9'
+          pytorch_version: 1.12.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py39_cu116_pyt1130
+          python_version: '3.9'
+          pytorch_version: 1.13.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py39_cu117_pyt1130
+          python_version: '3.9'
+          pytorch_version: 1.13.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py39_cu116_pyt1131
+          python_version: '3.9'
+          pytorch_version: 1.13.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py39_cu117_pyt1131
+          python_version: '3.9'
+          pytorch_version: 1.13.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py39_cu117_pyt200
+          python_version: '3.9'
+          pytorch_version: 2.0.0
       - binary_linux_conda:
           conda_docker_image: pytorch/conda-builder:cuda118
           context: DOCKERHUB_TOKEN
           cu_version: cu118
-          name: linux_conda_py38_cu118_pyt241
+          name: linux_conda_py39_cu118_pyt200
-          python_version: '3.8'
+          python_version: '3.9'
-          pytorch_version: 2.4.1
+          pytorch_version: 2.0.0
       - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
+          conda_docker_image: pytorch/conda-builder:cuda117
           context: DOCKERHUB_TOKEN
-          cu_version: cu121
+          cu_version: cu117
-          name: linux_conda_py38_cu121_pyt241
+          name: linux_conda_py39_cu117_pyt201
-          python_version: '3.8'
+          python_version: '3.9'
-          pytorch_version: 2.4.1
+          pytorch_version: 2.0.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda118
+          context: DOCKERHUB_TOKEN
+          cu_version: cu118
+          name: linux_conda_py39_cu118_pyt201
+          python_version: '3.9'
+          pytorch_version: 2.0.1
       - binary_linux_conda:
           conda_docker_image: pytorch/conda-builder:cuda118
           context: DOCKERHUB_TOKEN
@@ -359,33 +499,89 @@ workflows:
           python_version: '3.9'
           pytorch_version: 2.3.1
       - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda118
+          conda_docker_image: pytorch/conda-builder:cuda113
           context: DOCKERHUB_TOKEN
-          cu_version: cu118
+          cu_version: cu113
-          name: linux_conda_py39_cu118_pyt240
+          name: linux_conda_py310_cu113_pyt1120
-          python_version: '3.9'
+          python_version: '3.10'
-          pytorch_version: 2.4.0
+          pytorch_version: 1.12.0
       - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
+          conda_docker_image: pytorch/conda-builder:cuda116
           context: DOCKERHUB_TOKEN
-          cu_version: cu121
+          cu_version: cu116
-          name: linux_conda_py39_cu121_pyt240
+          name: linux_conda_py310_cu116_pyt1120
-          python_version: '3.9'
+          python_version: '3.10'
-          pytorch_version: 2.4.0
+          pytorch_version: 1.12.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda113
+          context: DOCKERHUB_TOKEN
+          cu_version: cu113
+          name: linux_conda_py310_cu113_pyt1121
+          python_version: '3.10'
+          pytorch_version: 1.12.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py310_cu116_pyt1121
+          python_version: '3.10'
+          pytorch_version: 1.12.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py310_cu116_pyt1130
+          python_version: '3.10'
+          pytorch_version: 1.13.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py310_cu117_pyt1130
+          python_version: '3.10'
+          pytorch_version: 1.13.0
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda116
+          context: DOCKERHUB_TOKEN
+          cu_version: cu116
+          name: linux_conda_py310_cu116_pyt1131
+          python_version: '3.10'
+          pytorch_version: 1.13.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py310_cu117_pyt1131
+          python_version: '3.10'
+          pytorch_version: 1.13.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda117
+          context: DOCKERHUB_TOKEN
+          cu_version: cu117
+          name: linux_conda_py310_cu117_pyt200
+          python_version: '3.10'
+          pytorch_version: 2.0.0
       - binary_linux_conda:
           conda_docker_image: pytorch/conda-builder:cuda118
           context: DOCKERHUB_TOKEN
           cu_version: cu118
-          name: linux_conda_py39_cu118_pyt241
+          name: linux_conda_py310_cu118_pyt200
-          python_version: '3.9'
+          python_version: '3.10'
-          pytorch_version: 2.4.1
+          pytorch_version: 2.0.0
       - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
+          conda_docker_image: pytorch/conda-builder:cuda117
           context: DOCKERHUB_TOKEN
-          cu_version: cu121
+          cu_version: cu117
-          name: linux_conda_py39_cu121_pyt241
+          name: linux_conda_py310_cu117_pyt201
-          python_version: '3.9'
+          python_version: '3.10'
-          pytorch_version: 2.4.1
+          pytorch_version: 2.0.1
+      - binary_linux_conda:
+          conda_docker_image: pytorch/conda-builder:cuda118
+          context: DOCKERHUB_TOKEN
+          cu_version: cu118
+          name: linux_conda_py310_cu118_pyt201
+          python_version: '3.10'
+          pytorch_version: 2.0.1
       - binary_linux_conda:
           conda_docker_image: pytorch/conda-builder:cuda118
           context: DOCKERHUB_TOKEN
@@ -470,34 +666,6 @@ workflows:
           name: linux_conda_py310_cu121_pyt231
           python_version: '3.10'
           pytorch_version: 2.3.1
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda118
-          context: DOCKERHUB_TOKEN
-          cu_version: cu118
-          name: linux_conda_py310_cu118_pyt240
-          python_version: '3.10'
-          pytorch_version: 2.4.0
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
-          context: DOCKERHUB_TOKEN
-          cu_version: cu121
-          name: linux_conda_py310_cu121_pyt240
-          python_version: '3.10'
-          pytorch_version: 2.4.0
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda118
-          context: DOCKERHUB_TOKEN
-          cu_version: cu118
-          name: linux_conda_py310_cu118_pyt241
-          python_version: '3.10'
-          pytorch_version: 2.4.1
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
-          context: DOCKERHUB_TOKEN
-          cu_version: cu121
-          name: linux_conda_py310_cu121_pyt241
-          python_version: '3.10'
-          pytorch_version: 2.4.1
       - binary_linux_conda:
           conda_docker_image: pytorch/conda-builder:cuda118
           context: DOCKERHUB_TOKEN
@@ -582,34 +750,6 @@ workflows:
           name: linux_conda_py311_cu121_pyt231
           python_version: '3.11'
           pytorch_version: 2.3.1
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda118
-          context: DOCKERHUB_TOKEN
-          cu_version: cu118
-          name: linux_conda_py311_cu118_pyt240
-          python_version: '3.11'
-          pytorch_version: 2.4.0
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
-          context: DOCKERHUB_TOKEN
-          cu_version: cu121
-          name: linux_conda_py311_cu121_pyt240
-          python_version: '3.11'
-          pytorch_version: 2.4.0
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda118
-          context: DOCKERHUB_TOKEN
-          cu_version: cu118
-          name: linux_conda_py311_cu118_pyt241
-          python_version: '3.11'
-          pytorch_version: 2.4.1
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
-          context: DOCKERHUB_TOKEN
-          cu_version: cu121
-          name: linux_conda_py311_cu121_pyt241
-          python_version: '3.11'
-          pytorch_version: 2.4.1
       - binary_linux_conda:
           conda_docker_image: pytorch/conda-builder:cuda118
           context: DOCKERHUB_TOKEN
@@ -652,34 +792,6 @@ workflows:
           name: linux_conda_py312_cu121_pyt231
           python_version: '3.12'
           pytorch_version: 2.3.1
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda118
-          context: DOCKERHUB_TOKEN
-          cu_version: cu118
-          name: linux_conda_py312_cu118_pyt240
-          python_version: '3.12'
-          pytorch_version: 2.4.0
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
-          context: DOCKERHUB_TOKEN
-          cu_version: cu121
-          name: linux_conda_py312_cu121_pyt240
-          python_version: '3.12'
-          pytorch_version: 2.4.0
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda118
-          context: DOCKERHUB_TOKEN
-          cu_version: cu118
-          name: linux_conda_py312_cu118_pyt241
-          python_version: '3.12'
-          pytorch_version: 2.4.1
-      - binary_linux_conda:
-          conda_docker_image: pytorch/conda-builder:cuda121
-          context: DOCKERHUB_TOKEN
-          cu_version: cu121
-          name: linux_conda_py312_cu121_pyt241
-          python_version: '3.12'
-          pytorch_version: 2.4.1
       - binary_linux_conda_cuda:
           name: testrun_conda_cuda_py310_cu117_pyt201
           context: DOCKERHUB_TOKEN

.circleci/regenerate.py

@@ -19,14 +19,18 @@ from packaging import version
 # The CUDA versions which have pytorch conda packages available for linux for each
 # version of pytorch.
 CONDA_CUDA_VERSIONS = {
+    "1.12.0": ["cu113", "cu116"],
+    "1.12.1": ["cu113", "cu116"],
+    "1.13.0": ["cu116", "cu117"],
+    "1.13.1": ["cu116", "cu117"],
+    "2.0.0": ["cu117", "cu118"],
+    "2.0.1": ["cu117", "cu118"],
     "2.1.0": ["cu118", "cu121"],
     "2.1.1": ["cu118", "cu121"],
     "2.1.2": ["cu118", "cu121"],
     "2.2.0": ["cu118", "cu121"],
     "2.2.2": ["cu118", "cu121"],
     "2.3.1": ["cu118", "cu121"],
-    "2.4.0": ["cu118", "cu121"],
-    "2.4.1": ["cu118", "cu121"],
 }
 
 

INSTALL.md

@@ -8,10 +8,11 @@
 The core library is written in PyTorch. Several components have underlying implementation in CUDA for improved performance. A subset of these components have CPU implementations in C++/PyTorch. It is advised to use PyTorch3D with GPU support in order to use all the features.
 
 - Linux or macOS or Windows
-- Python
+- Python 3.8, 3.9 or 3.10
-- PyTorch 2.1.0, 2.1.1, 2.1.2, 2.2.0, 2.2.1, 2.2.2, 2.3.0, 2.3.1, 2.4.0 or 2.4.1.
+- PyTorch 1.12.0, 1.12.1, 1.13.0, 2.0.0, 2.0.1, 2.1.0, 2.1.1, 2.1.2, 2.2.0, 2.2.1, 2.2.2, 2.3.0 or 2.3.1.
 - torchvision that matches the PyTorch installation. You can install them together as explained at pytorch.org to make sure of this.
 - gcc & g++ ≥ 4.9
+- [fvcore](https://github.com/facebookresearch/fvcore)
 - [ioPath](https://github.com/facebookresearch/iopath)
 - If CUDA is to be used, use a version which is supported by the corresponding pytorch version and at least version 9.2.
 - If CUDA older than 11.7 is to be used and you are building from source, the CUB library must be available. We recommend version 1.10.0.
@@ -21,7 +22,7 @@ The runtime dependencies can be installed by running:
 conda create -n pytorch3d python=3.9
 conda activate pytorch3d
 conda install pytorch=1.13.0 torchvision pytorch-cuda=11.6 -c pytorch -c nvidia
-conda install -c iopath iopath
+conda install -c fvcore -c iopath -c conda-forge fvcore iopath
 ```
 
 For the CUB build time dependency, which you only need if you have CUDA older than 11.7, if you are using conda, you can continue with
@@ -48,7 +49,6 @@ For developing on top of PyTorch3D or contributing, you will need to run the lin
 - tdqm
 - jupyter
 - imageio
-- fvcore
 - plotly
 - opencv-python
 
@@ -59,7 +59,6 @@ conda install jupyter
 pip install scikit-image matplotlib imageio plotly opencv-python
 
 # Tests/Linting
-conda install -c fvcore -c conda-forge fvcore
 pip install black usort flake8 flake8-bugbear flake8-comprehensions
 ```
 
@@ -98,7 +97,7 @@ version_str="".join([
     torch.version.cuda.replace(".",""),
     f"_pyt{pyt_version_str}"
 ])
-!pip install iopath
+!pip install fvcore iopath
 !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html
 ```
 

dev/run_tutorials.sh

@@ -23,7 +23,7 @@ conda init bash
 source ~/.bashrc
 conda create -y -n myenv python=3.8 matplotlib ipython ipywidgets nbconvert
 conda activate myenv
-conda install -y -c iopath iopath
+conda install -y -c fvcore -c iopath -c conda-forge fvcore iopath
 conda install -y -c pytorch pytorch=1.6.0 cudatoolkit=10.1 torchvision
 conda install -y -c pytorch3d-nightly pytorch3d
 pip install plotly scikit-image

docs/requirements.txt

@@ -5,6 +5,7 @@ sphinx_rtd_theme
 sphinx_markdown_tables
 numpy
 iopath
+fvcore
 https://download.pytorch.org/whl/cpu/torchvision-0.15.2%2Bcpu-cp311-cp311-linux_x86_64.whl
 https://download.pytorch.org/whl/cpu/torch-2.0.1%2Bcpu-cp311-cp311-linux_x86_64.whl
 omegaconf

docs/tutorials/bundle_adjustment.ipynb

@@ -96,7 +96,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/camera_position_optimization_with_differentiable_rendering.ipynb

@@ -83,7 +83,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/dataloaders_ShapeNetCore_R2N2.ipynb

@@ -58,7 +58,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/deform_source_mesh_to_target_mesh.ipynb

@@ -97,7 +97,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/fit_simple_neural_radiance_field.ipynb

@@ -63,7 +63,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/fit_textured_mesh.ipynb

@@ -75,7 +75,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/fit_textured_volume.ipynb

@@ -54,7 +54,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/implicitron_config_system.ipynb

@@ -85,7 +85,7 @@
         "        torch.version.cuda.replace(\".\",\"\"),\n",
         "        f\"_pyt{pyt_version_str}\"\n",
         "    ])\n",
-        "    !pip install iopath\n",
+        "    !pip install fvcore iopath\n",
         "    if sys.platform.startswith(\"linux\"):\n",
         "        print(\"Trying to install wheel for PyTorch3D\")\n",
         "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/implicitron_volumes.ipynb

@@ -79,7 +79,7 @@
         "        torch.version.cuda.replace(\".\",\"\"),\n",
         "        f\"_pyt{pyt_version_str}\"\n",
         "    ])\n",
-        "    !pip install iopath\n",
+        "    !pip install fvcore iopath\n",
         "    if sys.platform.startswith(\"linux\"):\n",
         "        print(\"Trying to install wheel for PyTorch3D\")\n",
         "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/render_colored_points.ipynb

@@ -57,7 +57,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/render_densepose.ipynb

@@ -64,7 +64,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

docs/tutorials/render_textured_meshes.ipynb

@@ -80,7 +80,7 @@
     "        torch.version.cuda.replace(\".\",\"\"),\n",
     "        f\"_pyt{pyt_version_str}\"\n",
     "    ])\n",
-    "    !pip install iopath\n",
+    "    !pip install fvcore iopath\n",
     "    if sys.platform.startswith(\"linux\"):\n",
     "        print(\"Trying to install wheel for PyTorch3D\")\n",
     "        !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",

packaging/build_conda.py

@@ -123,7 +123,7 @@ def do_build(start_args: List[str]):
     if test_flag is not None:
         args.append(test_flag)
 
-    args.extend(["-c", "bottler", "-c", "iopath", "-c", "conda-forge"])
+    args.extend(["-c", "bottler", "-c", "fvcore", "-c", "iopath", "-c", "conda-forge"])
     args.append("--no-anaconda-upload")
     args.extend(["--python", os.environ["PYTHON_VERSION"]])
     args.append("packaging/pytorch3d")

packaging/linux_wheels/README.md

@@ -26,6 +26,6 @@ version_str="".join([
     torch.version.cuda.replace(".",""),
     f"_pyt{pyt_version_str}"
 ])
-!pip install iopath
+!pip install fvcore iopath
 !pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html
 ```

packaging/linux_wheels/inside.sh

@@ -144,7 +144,7 @@ do
             conda activate "$tag"
             # shellcheck disable=SC2086
             conda install -y -c pytorch $extra_channel "pytorch=$pytorch_version" "$cudatools=$CUDA_TAG"
-            pip install iopath
+            pip install fvcore iopath
             echo "python version" "$python_version" "pytorch version" "$pytorch_version" "cuda version" "$cu_version" "tag" "$tag"
 
             rm -rf dist

packaging/pytorch3d/meta.yaml

@@ -22,6 +22,7 @@ requirements:
     - python
     - numpy >=1.11
     - torchvision >=0.5
+    - fvcore
     - iopath
     {{ environ.get('CONDA_PYTORCH_CONSTRAINT') }}
     {{ environ.get('CONDA_CUDATOOLKIT_CONSTRAINT') }}

projects/implicitron_trainer/experiment.py

@@ -99,7 +99,7 @@ except ModuleNotFoundError:
 no_accelerate = os.environ.get("PYTORCH3D_NO_ACCELERATE") is not None
 
 
-class Experiment(Configurable):
+class Experiment(Configurable):  # pyre-ignore: 13
     """
     This class is at the top level of Implicitron's config hierarchy. Its
     members are high-level components necessary for training an implicit rende-
@@ -120,16 +120,12 @@ class Experiment(Configurable):
             will be saved here.
     """
 
-    # pyre-fixme[13]: Attribute `data_source` is never initialized.
     data_source: DataSourceBase
     data_source_class_type: str = "ImplicitronDataSource"
-    # pyre-fixme[13]: Attribute `model_factory` is never initialized.
     model_factory: ModelFactoryBase
     model_factory_class_type: str = "ImplicitronModelFactory"
-    # pyre-fixme[13]: Attribute `optimizer_factory` is never initialized.
     optimizer_factory: OptimizerFactoryBase
     optimizer_factory_class_type: str = "ImplicitronOptimizerFactory"
-    # pyre-fixme[13]: Attribute `training_loop` is never initialized.
     training_loop: TrainingLoopBase
     training_loop_class_type: str = "ImplicitronTrainingLoop"
 

projects/implicitron_trainer/impl/model_factory.py

@@ -45,7 +45,7 @@ class ModelFactoryBase(ReplaceableBase):
 
 
 @registry.register
-class ImplicitronModelFactory(ModelFactoryBase):
+class ImplicitronModelFactory(ModelFactoryBase):  # pyre-ignore [13]
     """
     A factory class that initializes an implicit rendering model.
 
@@ -61,7 +61,6 @@ class ImplicitronModelFactory(ModelFactoryBase):
 
     """
 
-    # pyre-fixme[13]: Attribute `model` is never initialized.
     model: ImplicitronModelBase
     model_class_type: str = "GenericModel"
     resume: bool = True

projects/implicitron_trainer/impl/training_loop.py

@@ -30,13 +30,13 @@ from .utils import seed_all_random_engines
 logger = logging.getLogger(__name__)
 
 
+# pyre-fixme[13]: Attribute `evaluator` is never initialized.
 class TrainingLoopBase(ReplaceableBase):
     """
     Members:
         evaluator: An EvaluatorBase instance, used to evaluate training results.
     """
 
-    # pyre-fixme[13]: Attribute `evaluator` is never initialized.
     evaluator: Optional[EvaluatorBase]
     evaluator_class_type: Optional[str] = "ImplicitronEvaluator"
 

pytorch3d/__init__.py

@@ -6,4 +6,4 @@
 
 # pyre-unsafe
 
-__version__ = "0.7.8"
+__version__ = "0.7.6"

pytorch3d/csrc/ext.cpp

@@ -7,15 +7,11 @@
  */
 
 // clang-format off
-#if !defined(USE_ROCM)
 #include "./pulsar/global.h" // Include before <torch/extension.h>.
-#endif
 #include <torch/extension.h>
 // clang-format on
-#if !defined(USE_ROCM)
 #include "./pulsar/pytorch/renderer.h"
 #include "./pulsar/pytorch/tensor_util.h"
-#endif
 #include "ball_query/ball_query.h"
 #include "blending/sigmoid_alpha_blend.h"
 #include "compositing/alpha_composite.h"
@@ -103,8 +99,6 @@ PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
   m.def("marching_cubes", &MarchingCubes);
 
   // Pulsar.
-  // Pulsar not enabled on AMD.
-#if !defined(USE_ROCM)
 #ifdef PULSAR_LOGGING_ENABLED
   c10::ShowLogInfoToStderr();
 #endif
@@ -189,5 +183,4 @@ PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
   m.attr("MAX_UINT") = py::int_(MAX_UINT);
   m.attr("MAX_USHORT") = py::int_(MAX_USHORT);
   m.attr("PULSAR_MAX_GRAD_SPHERES") = py::int_(MAX_GRAD_SPHERES);
-#endif
 }

pytorch3d/csrc/rasterize_meshes/rasterize_meshes.cu

@@ -144,7 +144,7 @@ __device__ void CheckPixelInsideFace(
   const bool zero_face_area =
       (face_area <= kEpsilon && face_area >= -1.0f * kEpsilon);
 
-  if (zmax < 0 || (cull_backfaces && back_face) || outside_bbox ||
+  if (zmax < 0 || cull_backfaces && back_face || outside_bbox ||
       zero_face_area) {
     return;
   }

pytorch3d/csrc/utils/float_math.cuh

@@ -18,8 +18,6 @@ const auto vEpsilon = 1e-8;
 
 // Common functions and operators for float2.
 
-// Complex arithmetic is already defined for AMD.
-#if !defined(USE_ROCM)
 __device__ inline float2 operator-(const float2& a, const float2& b) {
   return make_float2(a.x - b.x, a.y - b.y);
 }
@@ -43,7 +41,6 @@ __device__ inline float2 operator*(const float2& a, const float2& b) {
 __device__ inline float2 operator*(const float a, const float2& b) {
   return make_float2(a * b.x, a * b.y);
 }
-#endif
 
 __device__ inline float FloatMin3(const float a, const float b, const float c) {
   return fminf(a, fminf(b, c));

pytorch3d/csrc/utils/warp_reduce.cuh

@@ -23,51 +23,37 @@ WarpReduceMin(scalar_t* min_dists, int64_t* min_idxs, const size_t tid) {
     min_idxs[tid] = min_idxs[tid + 32];
     min_dists[tid] = min_dists[tid + 32];
   }
-// AMD does not use explicit syncwarp and instead automatically inserts memory
-// fences during compilation.
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   // s = 16
   if (min_dists[tid] > min_dists[tid + 16]) {
     min_idxs[tid] = min_idxs[tid + 16];
     min_dists[tid] = min_dists[tid + 16];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   // s = 8
   if (min_dists[tid] > min_dists[tid + 8]) {
     min_idxs[tid] = min_idxs[tid + 8];
     min_dists[tid] = min_dists[tid + 8];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   // s = 4
   if (min_dists[tid] > min_dists[tid + 4]) {
     min_idxs[tid] = min_idxs[tid + 4];
     min_dists[tid] = min_dists[tid + 4];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   // s = 2
   if (min_dists[tid] > min_dists[tid + 2]) {
     min_idxs[tid] = min_idxs[tid + 2];
     min_dists[tid] = min_dists[tid + 2];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   // s = 1
   if (min_dists[tid] > min_dists[tid + 1]) {
     min_idxs[tid] = min_idxs[tid + 1];
     min_dists[tid] = min_dists[tid + 1];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
 }
 
 template <typename scalar_t>
@@ -79,42 +65,30 @@ __device__ void WarpReduceMax(
     dists[tid] = dists[tid + 32];
     dists_idx[tid] = dists_idx[tid + 32];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   if (dists[tid] < dists[tid + 16]) {
     dists[tid] = dists[tid + 16];
     dists_idx[tid] = dists_idx[tid + 16];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   if (dists[tid] < dists[tid + 8]) {
     dists[tid] = dists[tid + 8];
     dists_idx[tid] = dists_idx[tid + 8];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   if (dists[tid] < dists[tid + 4]) {
     dists[tid] = dists[tid + 4];
     dists_idx[tid] = dists_idx[tid + 4];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   if (dists[tid] < dists[tid + 2]) {
     dists[tid] = dists[tid + 2];
     dists_idx[tid] = dists_idx[tid + 2];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
   if (dists[tid] < dists[tid + 1]) {
     dists[tid] = dists[tid + 1];
     dists_idx[tid] = dists_idx[tid + 1];
   }
-#if !defined(USE_ROCM)
   __syncwarp();
-#endif
 }

pytorch3d/implicitron/dataset/data_source.py

@@ -41,7 +41,7 @@ class DataSourceBase(ReplaceableBase):
 
 
 @registry.register
-class ImplicitronDataSource(DataSourceBase):
+class ImplicitronDataSource(DataSourceBase):  # pyre-ignore[13]
     """
     Represents the data used in Implicitron. This is the only implementation
     of DataSourceBase provided.
@@ -52,11 +52,8 @@ class ImplicitronDataSource(DataSourceBase):
         data_loader_map_provider_class_type: identifies type for data_loader_map_provider.
     """
 
-    # pyre-fixme[13]: Attribute `dataset_map_provider` is never initialized.
     dataset_map_provider: DatasetMapProviderBase
-    # pyre-fixme[13]: Attribute `dataset_map_provider_class_type` is never initialized.
     dataset_map_provider_class_type: str
-    # pyre-fixme[13]: Attribute `data_loader_map_provider` is never initialized.
     data_loader_map_provider: DataLoaderMapProviderBase
     data_loader_map_provider_class_type: str = "SequenceDataLoaderMapProvider"
 

pytorch3d/implicitron/dataset/frame_data.py

@@ -276,7 +276,6 @@ class FrameData(Mapping[str, Any]):
             image_size_hw=tuple(self.effective_image_size_hw),  # pyre-ignore
         )
         crop_bbox_xywh = bbox_xyxy_to_xywh(clamp_bbox_xyxy)
-        self.crop_bbox_xywh = crop_bbox_xywh
 
         if self.fg_probability is not None:
             self.fg_probability = crop_around_box(
@@ -435,7 +434,7 @@ class FrameData(Mapping[str, Any]):
             # TODO: don't store K; enforce working in NDC space
             return join_cameras_as_batch(batch)
         else:
-            return torch.utils.data.dataloader.default_collate(batch)
+            return torch.utils.data._utils.collate.default_collate(batch)
 
 
 FrameDataSubtype = TypeVar("FrameDataSubtype", bound=FrameData)

pytorch3d/implicitron/dataset/json_index_dataset_map_provider.py

@@ -66,7 +66,7 @@ _NEED_CONTROL: Tuple[str, ...] = (
 
 
 @registry.register
-class JsonIndexDatasetMapProvider(DatasetMapProviderBase):
+class JsonIndexDatasetMapProvider(DatasetMapProviderBase):  # pyre-ignore [13]
     """
     Generates the training / validation and testing dataset objects for
     a dataset laid out on disk like Co3D, with annotations in json files.
@@ -95,7 +95,6 @@ class JsonIndexDatasetMapProvider(DatasetMapProviderBase):
         path_manager_factory_class_type: The class type of `path_manager_factory`.
     """
 
-    # pyre-fixme[13]: Attribute `category` is never initialized.
     category: str
     task_str: str = "singlesequence"
     dataset_root: str = _CO3D_DATASET_ROOT
@@ -105,10 +104,8 @@ class JsonIndexDatasetMapProvider(DatasetMapProviderBase):
     test_restrict_sequence_id: int = -1
     assert_single_seq: bool = False
     only_test_set: bool = False
-    # pyre-fixme[13]: Attribute `dataset` is never initialized.
     dataset: JsonIndexDataset
     dataset_class_type: str = "JsonIndexDataset"
-    # pyre-fixme[13]: Attribute `path_manager_factory` is never initialized.
     path_manager_factory: PathManagerFactory
     path_manager_factory_class_type: str = "PathManagerFactory"
 

pytorch3d/implicitron/dataset/json_index_dataset_map_provider_v2.py

@@ -56,7 +56,7 @@ logger = logging.getLogger(__name__)
 
 
 @registry.register
-class JsonIndexDatasetMapProviderV2(DatasetMapProviderBase):
+class JsonIndexDatasetMapProviderV2(DatasetMapProviderBase):  # pyre-ignore [13]
     """
     Generates the training, validation, and testing dataset objects for
     a dataset laid out on disk like CO3Dv2, with annotations in gzipped json files.
@@ -171,9 +171,7 @@ class JsonIndexDatasetMapProviderV2(DatasetMapProviderBase):
         path_manager_factory_class_type: The class type of `path_manager_factory`.
     """
 
-    # pyre-fixme[13]: Attribute `category` is never initialized.
     category: str
-    # pyre-fixme[13]: Attribute `subset_name` is never initialized.
     subset_name: str
     dataset_root: str = _CO3DV2_DATASET_ROOT
 
@@ -185,10 +183,8 @@ class JsonIndexDatasetMapProviderV2(DatasetMapProviderBase):
     n_known_frames_for_test: int = 0
 
     dataset_class_type: str = "JsonIndexDataset"
-    # pyre-fixme[13]: Attribute `dataset` is never initialized.
     dataset: JsonIndexDataset
 
-    # pyre-fixme[13]: Attribute `path_manager_factory` is never initialized.
     path_manager_factory: PathManagerFactory
     path_manager_factory_class_type: str = "PathManagerFactory"
 

pytorch3d/implicitron/dataset/rendered_mesh_dataset_map_provider.py

@@ -32,7 +32,7 @@ from .utils import DATASET_TYPE_KNOWN
 
 
 @registry.register
-class RenderedMeshDatasetMapProvider(DatasetMapProviderBase):
+class RenderedMeshDatasetMapProvider(DatasetMapProviderBase):  # pyre-ignore [13]
     """
     A simple single-scene dataset based on PyTorch3D renders of a mesh.
     Provides `num_views` renders of the mesh as train, with no val
@@ -76,7 +76,6 @@ class RenderedMeshDatasetMapProvider(DatasetMapProviderBase):
     resolution: int = 128
     use_point_light: bool = True
     gpu_idx: Optional[int] = 0
-    # pyre-fixme[13]: Attribute `path_manager_factory` is never initialized.
     path_manager_factory: PathManagerFactory
     path_manager_factory_class_type: str = "PathManagerFactory"
 

pytorch3d/implicitron/dataset/single_sequence_dataset.py

@@ -83,6 +83,7 @@ class SingleSceneDataset(DatasetBase, Configurable):
         return self.eval_batches
 
 
+# pyre-fixme[13]: Uninitialized attribute
 class SingleSceneDatasetMapProviderBase(DatasetMapProviderBase):
     """
     Base for provider of data for one scene from LLFF or blender datasets.
@@ -99,11 +100,8 @@ class SingleSceneDatasetMapProviderBase(DatasetMapProviderBase):
             testing frame.
     """
 
-    # pyre-fixme[13]: Attribute `base_dir` is never initialized.
     base_dir: str
-    # pyre-fixme[13]: Attribute `object_name` is never initialized.
     object_name: str
-    # pyre-fixme[13]: Attribute `path_manager_factory` is never initialized.
     path_manager_factory: PathManagerFactory
     path_manager_factory_class_type: str = "PathManagerFactory"
     n_known_frames_for_test: Optional[int] = None

pytorch3d/implicitron/dataset/utils.py

@@ -348,7 +348,6 @@ def adjust_camera_to_image_scale_(
     camera: PerspectiveCameras,
     original_size_wh: torch.Tensor,
     new_size_wh: torch.LongTensor,
-    # pyre-fixme[7]: Expected `PerspectiveCameras` but got implicit return value of `None`.
 ) -> PerspectiveCameras:
     focal_length_px, principal_point_px = _convert_ndc_to_pixels(
         camera.focal_length[0],
@@ -368,7 +367,7 @@ def adjust_camera_to_image_scale_(
         image_size_wh_output,
     )
     camera.focal_length = focal_length_scaled[None]
-    camera.principal_point = principal_point_scaled[None]
+    camera.principal_point = principal_point_scaled[None]  # pyre-ignore
 
 
 # NOTE this cache is per-worker; they are implemented as processes.

pytorch3d/implicitron/models/generic_model.py

@@ -65,7 +65,7 @@ logger = logging.getLogger(__name__)
 
 
 @registry.register
-class GenericModel(ImplicitronModelBase):
+class GenericModel(ImplicitronModelBase):  # pyre-ignore: 13
     """
     GenericModel is a wrapper for the neural implicit
     rendering and reconstruction pipeline which consists
@@ -226,42 +226,34 @@ class GenericModel(ImplicitronModelBase):
 
     # ---- global encoder settings
     global_encoder_class_type: Optional[str] = None
-    # pyre-fixme[13]: Attribute `global_encoder` is never initialized.
     global_encoder: Optional[GlobalEncoderBase]
 
     # ---- raysampler
     raysampler_class_type: str = "AdaptiveRaySampler"
-    # pyre-fixme[13]: Attribute `raysampler` is never initialized.
     raysampler: RaySamplerBase
 
     # ---- renderer configs
     renderer_class_type: str = "MultiPassEmissionAbsorptionRenderer"
-    # pyre-fixme[13]: Attribute `renderer` is never initialized.
     renderer: BaseRenderer
 
     # ---- image feature extractor settings
     # (This is only created if view_pooler is enabled)
-    # pyre-fixme[13]: Attribute `image_feature_extractor` is never initialized.
     image_feature_extractor: Optional[FeatureExtractorBase]
     image_feature_extractor_class_type: Optional[str] = None
     # ---- view pooler settings
     view_pooler_enabled: bool = False
-    # pyre-fixme[13]: Attribute `view_pooler` is never initialized.
     view_pooler: Optional[ViewPooler]
 
     # ---- implicit function settings
     implicit_function_class_type: str = "NeuralRadianceFieldImplicitFunction"
     # This is just a model, never constructed.
     # The actual implicit functions live in self._implicit_functions
-    # pyre-fixme[13]: Attribute `implicit_function` is never initialized.
     implicit_function: ImplicitFunctionBase
 
     # ----- metrics
-    # pyre-fixme[13]: Attribute `view_metrics` is never initialized.
     view_metrics: ViewMetricsBase
     view_metrics_class_type: str = "ViewMetrics"
 
-    # pyre-fixme[13]: Attribute `regularization_metrics` is never initialized.
     regularization_metrics: RegularizationMetricsBase
     regularization_metrics_class_type: str = "RegularizationMetrics"
 

pytorch3d/implicitron/models/global_encoder/global_encoder.py

@@ -59,13 +59,12 @@ class GlobalEncoderBase(ReplaceableBase):
 
 # TODO: probabilistic embeddings?
 @registry.register
-class SequenceAutodecoder(GlobalEncoderBase, torch.nn.Module):
+class SequenceAutodecoder(GlobalEncoderBase, torch.nn.Module):  # pyre-ignore: 13
     """
     A global encoder implementation which provides an autodecoder encoding
     of the frame's sequence identifier.
     """
 
-    # pyre-fixme[13]: Attribute `autodecoder` is never initialized.
     autodecoder: Autodecoder
 
     def __post_init__(self):

pytorch3d/implicitron/models/implicit_function/decoding_functions.py

@@ -244,6 +244,7 @@ class MLPWithInputSkips(Configurable, torch.nn.Module):
 
 
 @registry.register
+# pyre-fixme[13]: Attribute `network` is never initialized.
 class MLPDecoder(DecoderFunctionBase):
     """
     Decoding function which uses `MLPWithIputSkips` to convert the embedding to output.
@@ -271,7 +272,6 @@ class MLPDecoder(DecoderFunctionBase):
 
     input_dim: int = 3
     param_groups: Dict[str, str] = field(default_factory=lambda: {})
-    # pyre-fixme[13]: Attribute `network` is never initialized.
     network: MLPWithInputSkips
 
     def __post_init__(self):

pytorch3d/implicitron/models/implicit_function/scene_representation_networks.py

@@ -318,11 +318,10 @@ class SRNRaymarchHyperNet(Configurable, torch.nn.Module):
 
 
 @registry.register
+# pyre-fixme[13]: Uninitialized attribute
 class SRNImplicitFunction(ImplicitFunctionBase, torch.nn.Module):
     latent_dim: int = 0
-    # pyre-fixme[13]: Attribute `raymarch_function` is never initialized.
     raymarch_function: SRNRaymarchFunction
-    # pyre-fixme[13]: Attribute `pixel_generator` is never initialized.
     pixel_generator: SRNPixelGenerator
 
     def __post_init__(self):
@@ -367,6 +366,7 @@ class SRNImplicitFunction(ImplicitFunctionBase, torch.nn.Module):
 
 
 @registry.register
+# pyre-fixme[13]: Uninitialized attribute
 class SRNHyperNetImplicitFunction(ImplicitFunctionBase, torch.nn.Module):
     """
     This implicit function uses a hypernetwork to generate the
@@ -377,9 +377,7 @@ class SRNHyperNetImplicitFunction(ImplicitFunctionBase, torch.nn.Module):
 
     latent_dim_hypernet: int = 0
     latent_dim: int = 0
-    # pyre-fixme[13]: Attribute `hypernet` is never initialized.
     hypernet: SRNRaymarchHyperNet
-    # pyre-fixme[13]: Attribute `pixel_generator` is never initialized.
     pixel_generator: SRNPixelGenerator
 
     def __post_init__(self):

pytorch3d/implicitron/models/implicit_function/voxel_grid.py

@@ -805,6 +805,7 @@ class VMFactorizedVoxelGrid(VoxelGridBase):
         )
 
 
+# pyre-fixme[13]: Attribute `voxel_grid` is never initialized.
 class VoxelGridModule(Configurable, torch.nn.Module):
     """
     A wrapper torch.nn.Module for the VoxelGrid classes, which
@@ -844,7 +845,6 @@ class VoxelGridModule(Configurable, torch.nn.Module):
     """
 
     voxel_grid_class_type: str = "FullResolutionVoxelGrid"
-    # pyre-fixme[13]: Attribute `voxel_grid` is never initialized.
     voxel_grid: VoxelGridBase
 
     extents: Tuple[float, float, float] = (2.0, 2.0, 2.0)

pytorch3d/implicitron/models/implicit_function/voxel_grid_implicit_function.py

@@ -39,6 +39,7 @@ enable_get_default_args(HarmonicEmbedding)
 
 
 @registry.register
+# pyre-ignore[13]
 class VoxelGridImplicitFunction(ImplicitFunctionBase, torch.nn.Module):
     """
     This implicit function consists of two streams, one for the density calculation and one
@@ -144,11 +145,9 @@ class VoxelGridImplicitFunction(ImplicitFunctionBase, torch.nn.Module):
     """
 
     # ---- voxel grid for density
-    # pyre-fixme[13]: Attribute `voxel_grid_density` is never initialized.
     voxel_grid_density: VoxelGridModule
 
     # ---- voxel grid for color
-    # pyre-fixme[13]: Attribute `voxel_grid_color` is never initialized.
     voxel_grid_color: VoxelGridModule
 
     # ---- harmonic embeddings density
@@ -164,12 +163,10 @@ class VoxelGridImplicitFunction(ImplicitFunctionBase, torch.nn.Module):
 
     # ---- decoder function for density
     decoder_density_class_type: str = "MLPDecoder"
-    # pyre-fixme[13]: Attribute `decoder_density` is never initialized.
     decoder_density: DecoderFunctionBase
 
     # ---- decoder function for color
     decoder_color_class_type: str = "MLPDecoder"
-    # pyre-fixme[13]: Attribute `decoder_color` is never initialized.
     decoder_color: DecoderFunctionBase
 
     # ---- cuda streams

pytorch3d/implicitron/models/overfit_model.py

@@ -69,7 +69,7 @@ IMPLICIT_FUNCTION_ARGS_TO_REMOVE: List[str] = [
 
 
 @registry.register
-class OverfitModel(ImplicitronModelBase):
+class OverfitModel(ImplicitronModelBase):  # pyre-ignore: 13
     """
     OverfitModel is a wrapper for the neural implicit
     rendering and reconstruction pipeline which consists
@@ -198,34 +198,27 @@ class OverfitModel(ImplicitronModelBase):
 
     # ---- global encoder settings
     global_encoder_class_type: Optional[str] = None
-    # pyre-fixme[13]: Attribute `global_encoder` is never initialized.
     global_encoder: Optional[GlobalEncoderBase]
 
     # ---- raysampler
     raysampler_class_type: str = "AdaptiveRaySampler"
-    # pyre-fixme[13]: Attribute `raysampler` is never initialized.
     raysampler: RaySamplerBase
 
     # ---- renderer configs
     renderer_class_type: str = "MultiPassEmissionAbsorptionRenderer"
-    # pyre-fixme[13]: Attribute `renderer` is never initialized.
     renderer: BaseRenderer
 
     # ---- implicit function settings
     share_implicit_function_across_passes: bool = False
     implicit_function_class_type: str = "NeuralRadianceFieldImplicitFunction"
-    # pyre-fixme[13]: Attribute `implicit_function` is never initialized.
     implicit_function: ImplicitFunctionBase
     coarse_implicit_function_class_type: Optional[str] = None
-    # pyre-fixme[13]: Attribute `coarse_implicit_function` is never initialized.
     coarse_implicit_function: Optional[ImplicitFunctionBase]
 
     # ----- metrics
-    # pyre-fixme[13]: Attribute `view_metrics` is never initialized.
     view_metrics: ViewMetricsBase
     view_metrics_class_type: str = "ViewMetrics"
 
-    # pyre-fixme[13]: Attribute `regularization_metrics` is never initialized.
     regularization_metrics: RegularizationMetricsBase
     regularization_metrics_class_type: str = "RegularizationMetrics"
 

pytorch3d/implicitron/models/renderer/multipass_ea.py

@@ -18,7 +18,9 @@ from .raymarcher import RaymarcherBase
 
 
 @registry.register
-class MultiPassEmissionAbsorptionRenderer(BaseRenderer, torch.nn.Module):
+class MultiPassEmissionAbsorptionRenderer(  # pyre-ignore: 13
+    BaseRenderer, torch.nn.Module
+):
     """
     Implements the multi-pass rendering function, in particular,
     with emission-absorption ray marching used in NeRF [1]. First, it evaluates
@@ -84,7 +86,6 @@ class MultiPassEmissionAbsorptionRenderer(BaseRenderer, torch.nn.Module):
     """
 
     raymarcher_class_type: str = "EmissionAbsorptionRaymarcher"
-    # pyre-fixme[13]: Attribute `raymarcher` is never initialized.
     raymarcher: RaymarcherBase
 
     n_pts_per_ray_fine_training: int = 64

pytorch3d/implicitron/models/renderer/ray_point_refiner.py

@@ -16,6 +16,8 @@ from pytorch3d.renderer.implicit.sample_pdf import sample_pdf
 
 
 @expand_args_fields
+# pyre-fixme[13]: Attribute `n_pts_per_ray` is never initialized.
+# pyre-fixme[13]: Attribute `random_sampling` is never initialized.
 class RayPointRefiner(Configurable, torch.nn.Module):
     """
     Implements the importance sampling of points along rays.
@@ -43,9 +45,7 @@ class RayPointRefiner(Configurable, torch.nn.Module):
             for Anti-Aliasing Neural Radiance Fields." ICCV 2021.
     """
 
-    # pyre-fixme[13]: Attribute `n_pts_per_ray` is never initialized.
     n_pts_per_ray: int
-    # pyre-fixme[13]: Attribute `random_sampling` is never initialized.
     random_sampling: bool
     add_input_samples: bool = True
     blurpool_weights: bool = False

pytorch3d/implicitron/models/renderer/sdf_renderer.py

@@ -24,10 +24,9 @@ from .rgb_net import RayNormalColoringNetwork
 
 
 @registry.register
-class SignedDistanceFunctionRenderer(BaseRenderer, torch.nn.Module):
+class SignedDistanceFunctionRenderer(BaseRenderer, torch.nn.Module):  # pyre-ignore[13]
     render_features_dimensions: int = 3
     object_bounding_sphere: float = 1.0
-    # pyre-fixme[13]: Attribute `ray_tracer` is never initialized.
     ray_tracer: RayTracing
     ray_normal_coloring_network_args: DictConfig = get_default_args_field(
         RayNormalColoringNetwork

pytorch3d/implicitron/models/utils.py

@@ -121,6 +121,7 @@ def weighted_sum_losses(
         return None
     loss = sum(losses_weighted)
     assert torch.is_tensor(loss)
+    # pyre-fixme[7]: Expected `Optional[Tensor]` but got `int`.
     return loss
 
 

pytorch3d/implicitron/models/view_pooler/view_pooler.py

@@ -16,6 +16,7 @@ from .feature_aggregator import FeatureAggregatorBase
 from .view_sampler import ViewSampler
 
 
+# pyre-ignore: 13
 class ViewPooler(Configurable, torch.nn.Module):
     """
     Implements sampling of image-based features at the 2d projections of a set
@@ -34,10 +35,8 @@ class ViewPooler(Configurable, torch.nn.Module):
             from a set of source images. FeatureAggregator executes step (4) above.
     """
 
-    # pyre-fixme[13]: Attribute `view_sampler` is never initialized.
     view_sampler: ViewSampler
     feature_aggregator_class_type: str = "AngleWeightedReductionFeatureAggregator"
-    # pyre-fixme[13]: Attribute `feature_aggregator` is never initialized.
     feature_aggregator: FeatureAggregatorBase
 
     def __post_init__(self):

pytorch3d/implicitron/tools/point_cloud_utils.py

@@ -156,6 +156,7 @@ def render_point_cloud_pytorch3d(
     cumprod = torch.cat((torch.ones_like(cumprod[..., :1]), cumprod[..., :-1]), dim=-1)
     depths = (weights * cumprod * fragments.zbuf).sum(dim=-1)
     # add the rendering mask
+    # pyre-fixme[6]: For 1st param expected `Tensor` but got `float`.
     render_mask = -torch.prod(1.0 - weights, dim=-1) + 1.0
 
     # cat depths and render mask

pytorch3d/io/mtl_io.py

@@ -409,7 +409,6 @@ def _parse_mtl(
     texture_files = {}
     material_name = ""
 
-    # pyre-fixme[9]: f has type `str`; used as `IO[typing.Any]`.
     with _open_file(f, path_manager, "r") as f:
         for line in f:
             tokens = line.strip().split()

pytorch3d/io/obj_io.py

@@ -649,7 +649,8 @@ def _load_obj(
         # Create an array of strings of material names for each face.
         # If faces_materials_idx == -1 then that face doesn't have a material.
         idx = faces_materials_idx.cpu().numpy()
-        face_material_names = np.array([""] + material_names)[idx + 1]  # (F,)
+        face_material_names = np.array(material_names)[idx]  # (F,)
+        face_material_names[idx == -1] = ""
 
         # Construct the atlas.
         texture_atlas = make_mesh_texture_atlas(
@@ -755,13 +756,10 @@ def save_obj(
     output_path = Path(f)
 
     # Save the .obj file
-    # pyre-fixme[9]: f has type `Union[Path, str]`; used as `IO[typing.Any]`.
     with _open_file(f, path_manager, "w") as f:
         if save_texture:
             # Add the header required for the texture info to be loaded correctly
             obj_header = "\nmtllib {0}.mtl\nusemtl mesh\n\n".format(output_path.stem)
-            # pyre-fixme[16]: Item `Path` of `Union[Path, str]` has no attribute
-            #  `write`.
             f.write(obj_header)
         _save(
             f,

pytorch3d/loss/chamfer.py

@@ -27,10 +27,8 @@ def _validate_chamfer_reduction_inputs(
     """
     if batch_reduction is not None and batch_reduction not in ["mean", "sum"]:
         raise ValueError('batch_reduction must be one of ["mean", "sum"] or None')
-    if point_reduction is not None and point_reduction not in ["mean", "sum", "max"]:
+    if point_reduction is not None and point_reduction not in ["mean", "sum"]:
-        raise ValueError(
+        raise ValueError('point_reduction must be one of ["mean", "sum"] or None')
-            'point_reduction must be one of ["mean", "sum", "max"] or None'
-        )
     if point_reduction is None and batch_reduction is not None:
         raise ValueError("Batch reduction must be None if point_reduction is None")
 
@@ -82,6 +80,7 @@ def _chamfer_distance_single_direction(
     x_normals,
     y_normals,
     weights,
+    batch_reduction: Union[str, None],
     point_reduction: Union[str, None],
     norm: int,
     abs_cosine: bool,
@@ -104,6 +103,11 @@ def _chamfer_distance_single_direction(
             raise ValueError("weights cannot be negative.")
         if weights.sum() == 0.0:
             weights = weights.view(N, 1)
+            if batch_reduction in ["mean", "sum"]:
+                return (
+                    (x.sum((1, 2)) * weights).sum() * 0.0,
+                    (x.sum((1, 2)) * weights).sum() * 0.0,
+                )
             return ((x.sum((1, 2)) * weights) * 0.0, (x.sum((1, 2)) * weights) * 0.0)
 
     cham_norm_x = x.new_zeros(())
@@ -131,10 +135,7 @@ def _chamfer_distance_single_direction(
         if weights is not None:
             cham_norm_x *= weights.view(N, 1)
 
-    if point_reduction == "max":
+    if point_reduction is not None:
-        assert not return_normals
-        cham_x = cham_x.max(1).values  # (N,)
-    elif point_reduction is not None:
         # Apply point reduction
         cham_x = cham_x.sum(1)  # (N,)
         if return_normals:
@@ -145,34 +146,22 @@ def _chamfer_distance_single_direction(
             if return_normals:
                 cham_norm_x /= x_lengths_clamped
 
+        if batch_reduction is not None:
+            # batch_reduction == "sum"
+            cham_x = cham_x.sum()
+            if return_normals:
+                cham_norm_x = cham_norm_x.sum()
+            if batch_reduction == "mean":
+                div = weights.sum() if weights is not None else max(N, 1)
+                cham_x /= div
+                if return_normals:
+                    cham_norm_x /= div
+
     cham_dist = cham_x
     cham_normals = cham_norm_x if return_normals else None
     return cham_dist, cham_normals
 
 
-def _apply_batch_reduction(
-    cham_x, cham_norm_x, weights, batch_reduction: Union[str, None]
-):
-    if batch_reduction is None:
-        return (cham_x, cham_norm_x)
-    # batch_reduction == "sum"
-    N = cham_x.shape[0]
-    cham_x = cham_x.sum()
-    if cham_norm_x is not None:
-        cham_norm_x = cham_norm_x.sum()
-    if batch_reduction == "mean":
-        if weights is None:
-            div = max(N, 1)
-        elif weights.sum() == 0.0:
-            div = 1
-        else:
-            div = weights.sum()
-        cham_x /= div
-        if cham_norm_x is not None:
-            cham_norm_x /= div
-    return (cham_x, cham_norm_x)
-
-
 def chamfer_distance(
     x,
     y,
@@ -208,8 +197,7 @@ def chamfer_distance(
         batch_reduction: Reduction operation to apply for the loss across the
             batch, can be one of ["mean", "sum"] or None.
         point_reduction: Reduction operation to apply for the loss across the
-            points, can be one of ["mean", "sum", "max"] or None. Using "max" leads to the
+            points, can be one of ["mean", "sum"] or None.
-            Hausdorff distance.
         norm: int indicates the norm used for the distance. Supports 1 for L1 and 2 for L2.
         single_directional: If False (default), loss comes from both the distance between
             each point in x and its nearest neighbor in y and each point in y and its nearest
@@ -239,10 +227,6 @@ def chamfer_distance(
 
     if not ((norm == 1) or (norm == 2)):
         raise ValueError("Support for 1 or 2 norm.")
-
-    if point_reduction == "max" and (x_normals is not None or y_normals is not None):
-        raise ValueError('Normals must be None if point_reduction is "max"')
-
     x, x_lengths, x_normals = _handle_pointcloud_input(x, x_lengths, x_normals)
     y, y_lengths, y_normals = _handle_pointcloud_input(y, y_lengths, y_normals)
 
@@ -254,13 +238,13 @@ def chamfer_distance(
         x_normals,
         y_normals,
         weights,
+        batch_reduction,
         point_reduction,
         norm,
         abs_cosine,
     )
     if single_directional:
-        loss = cham_x
+        return cham_x, cham_norm_x
-        loss_normals = cham_norm_x
     else:
         cham_y, cham_norm_y = _chamfer_distance_single_direction(
             y,
@@ -270,23 +254,17 @@ def chamfer_distance(
             y_normals,
             x_normals,
             weights,
+            batch_reduction,
             point_reduction,
             norm,
             abs_cosine,
         )
-        if point_reduction == "max":
+        if point_reduction is not None:
-            loss = torch.maximum(cham_x, cham_y)
+            return (
-            loss_normals = None
+                cham_x + cham_y,
-        elif point_reduction is not None:
+                (cham_norm_x + cham_norm_y) if cham_norm_x is not None else None,
-            loss = cham_x + cham_y
+            )
-            if cham_norm_x is not None:
+        return (
-                loss_normals = cham_norm_x + cham_norm_y
+            (cham_x, cham_y),
-            else:
+            (cham_norm_x, cham_norm_y) if cham_norm_x is not None else None,
-                loss_normals = None
+        )
-        else:
-            loss = (cham_x, cham_y)
-            if cham_norm_x is not None:
-                loss_normals = (cham_norm_x, cham_norm_y)
-            else:
-                loss_normals = None
-    return _apply_batch_reduction(loss, loss_normals, weights, batch_reduction)

pytorch3d/ops/points_to_volumes.py

@@ -617,7 +617,6 @@ def _splat_points_to_volumes(
                 w = wX * wY * wZ
 
                 # valid - binary indicators of votes that fall into the volume
-                # pyre-fixme[16]: `int` has no attribute `long`.
                 valid = (
                     (0 <= X_)
                     * (X_ < grid_sizes_xyz[:, None, 0:1])
@@ -636,19 +635,14 @@ def _splat_points_to_volumes(
                 idx_valid = idx * valid + rand_idx * (1 - valid)
                 w_valid = w * valid.type_as(w)
                 if mask is not None:
-                    # pyre-fixme[6]: For 1st argument expected `Tensor` but got `int`.
                     w_valid = w_valid * mask.type_as(w)[:, :, None]
 
                 # scatter add casts the votes into the weight accumulator
                 # and the feature accumulator
-                # pyre-fixme[6]: For 3rd argument expected `Tensor` but got
-                #  `Union[int, Tensor]`.
                 volume_densities.scatter_add_(1, idx_valid, w_valid)
 
                 # reshape idx_valid -> (minibatch, feature_dim, n_points)
                 idx_valid = idx_valid.view(ba, 1, n_points).expand_as(points_features)
-                # pyre-fixme[16]: Item `int` of `Union[int, Tensor]` has no
-                #  attribute `view`.
                 w_valid = w_valid.view(ba, 1, n_points)
 
                 # volume_features of shape (minibatch, feature_dim, n_voxels)
@@ -730,7 +724,6 @@ def _round_points_to_volumes(
     # valid - binary indicators of votes that fall into the volume
     # pyre-fixme[9]: grid_sizes has type `LongTensor`; used as `Tensor`.
     grid_sizes = grid_sizes.type_as(XYZ)
-    # pyre-fixme[16]: `int` has no attribute `long`.
     valid = (
         (0 <= X)
         * (X < grid_sizes_xyz[:, None, 0:1])

pytorch3d/ops/utils.py

@@ -143,6 +143,8 @@ def convert_pointclouds_to_tensor(pcl: Union[torch.Tensor, "Pointclouds"]):
     elif torch.is_tensor(pcl):
         X = pcl
         num_points = X.shape[1] * torch.ones(  # type: ignore
+            # pyre-fixme[16]: Item `Pointclouds` of `Union[Pointclouds, Tensor]` has
+            #  no attribute `shape`.
             X.shape[0],
             device=X.device,
             dtype=torch.int64,

pytorch3d/renderer/__init__.py

@@ -6,8 +6,6 @@
 
 # pyre-unsafe
 
-import torch
-
 from .blending import (
     BlendParams,
     hard_rgb_blend,
@@ -76,13 +74,9 @@ from .points import (
     PointsRasterizationSettings,
     PointsRasterizer,
     PointsRenderer,
+    PulsarPointsRenderer,
     rasterize_points,
 )
-
-# Pulsar is not enabled on amd.
-if not torch.version.hip:
-    from .points import PulsarPointsRenderer
-
 from .splatter_blend import SplatterBlender
 from .utils import (
     convert_to_tensors_and_broadcast,

pytorch3d/renderer/blending.py

@@ -212,15 +212,15 @@ def softmax_rgb_blend(
 
     # Reshape to be compatible with (N, H, W, K) values in fragments
     if torch.is_tensor(zfar):
+        # pyre-fixme[16]
         zfar = zfar[:, None, None, None]
     if torch.is_tensor(znear):
+        # pyre-fixme[16]: Item `float` of `Union[float, Tensor]` has no attribute
+        #  `__getitem__`.
         znear = znear[:, None, None, None]
 
-    # pyre-fixme[6]: Expected `float` but got `Union[float, Tensor]`
     z_inv = (zfar - fragments.zbuf) / (zfar - znear) * mask
-    # pyre-fixme[6]: Expected `Tensor` but got `float`
     z_inv_max = torch.max(z_inv, dim=-1).values[..., None].clamp(min=eps)
-    # pyre-fixme[6]: Expected `Tensor` but got `float`
     weights_num = prob_map * torch.exp((z_inv - z_inv_max) / blend_params.gamma)
 
     # Also apply exp normalize trick for the background color weight.

pytorch3d/renderer/cameras.py

@@ -1782,6 +1782,8 @@ def get_ndc_to_screen_transform(
     K = torch.zeros((cameras._N, 4, 4), device=cameras.device, dtype=torch.float32)
     if not torch.is_tensor(image_size):
         image_size = torch.tensor(image_size, device=cameras.device)
+    # pyre-fixme[16]: Item `List` of `Union[List[typing.Any], Tensor, Tuple[Any,
+    #  ...]]` has no attribute `view`.
     image_size = image_size.view(-1, 2)  # of shape (1 or B)x2
     height, width = image_size.unbind(1)
 

pytorch3d/renderer/mesh/clip.py

@@ -497,7 +497,6 @@ def clip_faces(
     faces_case3 = face_verts_unclipped[case3_unclipped_idx]
 
     # index (0, 1, or 2) of the vertex in front of the clipping plane
-    # pyre-fixme[61]: `faces_clipped_verts` is undefined, or not always defined.
     p1_face_ind = torch.where(~faces_clipped_verts[case3_unclipped_idx])[1]
 
     # Solve for the points p4, p5 that intersect the clipping plane
@@ -541,7 +540,6 @@ def clip_faces(
     faces_case4 = face_verts_unclipped[case4_unclipped_idx]
 
     # index (0, 1, or 2) of the vertex behind the clipping plane
-    # pyre-fixme[61]: `faces_clipped_verts` is undefined, or not always defined.
     p1_face_ind = torch.where(faces_clipped_verts[case4_unclipped_idx])[1]
 
     # Solve for the points p4, p5 that intersect the clipping plane

pytorch3d/renderer/points/__init__.py

@@ -6,13 +6,8 @@
 
 # pyre-unsafe
 
-import torch
-
 from .compositor import AlphaCompositor, NormWeightedCompositor
-
+from .pulsar.unified import PulsarPointsRenderer
-# Pulsar not enabled on amd.
-if not torch.version.hip:
-    from .pulsar.unified import PulsarPointsRenderer
 
 from .rasterize_points import rasterize_points
 from .rasterizer import PointsRasterizationSettings, PointsRasterizer

pytorch3d/renderer/utils.py

@@ -270,8 +270,6 @@ class TensorProperties(nn.Module):
                         # to have the same shape as the input tensor.
                         new_dims = len(tensor_dims) - len(idx_dims)
                         new_shape = idx_dims + (1,) * new_dims
-                        # pyre-fixme[58]: `+` is not supported for operand types
-                        # `Tuple[int]` and `torch._C.Size`
                         expand_dims = (-1,) + tensor_dims[1:]
                         _batch_idx = _batch_idx.view(*new_shape)
                         _batch_idx = _batch_idx.expand(*expand_dims)

pytorch3d/transforms/rotation_conversions.py

@@ -97,10 +97,7 @@ def _sqrt_positive_part(x: torch.Tensor) -> torch.Tensor:
     """
     ret = torch.zeros_like(x)
     positive_mask = x > 0
-    if torch.is_grad_enabled():
     ret[positive_mask] = torch.sqrt(x[positive_mask])
-    else:
-        ret = torch.where(positive_mask, torch.sqrt(x), ret)
     return ret
 
 

pytorch3d/vis/plotly_vis.py

@@ -369,7 +369,6 @@ def plot_scene(
         # update camera viewpoint if provided
         if viewpoints_eye_at_up_world is not None:
             # Use camera params for batch index or the first camera if only one provided.
-            # pyre-fixme[61]: `n_viewpoint_cameras` is undefined, or not always defined.
             viewpoint_idx = min(n_viewpoint_cameras - 1, subplot_idx)
 
             eye, at, up = (i[viewpoint_idx] for i in viewpoints_eye_at_up_world)
@@ -628,7 +627,7 @@ def _add_struct_from_batch(
 
 
 def _add_mesh_trace(
-    fig: go.Figure,
+    fig: go.Figure,  # pyre-ignore[11]
     meshes: Meshes,
     trace_name: str,
     subplot_idx: int,
@@ -674,7 +673,6 @@ def _add_mesh_trace(
     verts[~verts_used] = verts_center
 
     row, col = subplot_idx // ncols + 1, subplot_idx % ncols + 1
-    # pyre-fixme[16]: `Figure` has no attribute `add_trace`.
     fig.add_trace(
         go.Mesh3d(
             x=verts[:, 0],
@@ -741,7 +739,6 @@ def _add_pointcloud_trace(
 
     row = subplot_idx // ncols + 1
     col = subplot_idx % ncols + 1
-    # pyre-fixme[16]: `Figure` has no attribute `add_trace`.
     fig.add_trace(
         go.Scatter3d(
             x=verts[:, 0],
@@ -803,7 +800,6 @@ def _add_camera_trace(
     x, y, z = all_cam_wires.detach().cpu().numpy().T.astype(float)
 
     row, col = subplot_idx // ncols + 1, subplot_idx % ncols + 1
-    # pyre-fixme[16]: `Figure` has no attribute `add_trace`.
     fig.add_trace(
         go.Scatter3d(x=x, y=y, z=z, marker={"size": 1}, name=trace_name),
         row=row,
@@ -898,7 +894,6 @@ def _add_ray_bundle_trace(
         ray_lines = torch.cat((ray_lines, nan_tensor, ray_line))
     x, y, z = ray_lines.detach().cpu().numpy().T.astype(float)
     row, col = subplot_idx // ncols + 1, subplot_idx % ncols + 1
-    # pyre-fixme[16]: `Figure` has no attribute `add_trace`.
     fig.add_trace(
         go.Scatter3d(
             x=x,
@@ -993,7 +988,7 @@ def _gen_fig_with_subplots(
 def _update_axes_bounds(
     verts_center: torch.Tensor,
     max_expand: float,
-    current_layout: go.Scene,
+    current_layout: go.Scene,  # pyre-ignore[11]
 ) -> None:  # pragma: no cover
     """
     Takes in the vertices' center point and max spread, and the current plotly figure
@@ -1010,7 +1005,6 @@ def _update_axes_bounds(
 
     # Ensure that within a subplot, the bounds capture all traces
     old_xrange, old_yrange, old_zrange = (
-        # pyre-fixme[16]: `Scene` has no attribute `__getitem__`.
         current_layout["xaxis"]["range"],
         current_layout["yaxis"]["range"],
         current_layout["zaxis"]["range"],
@@ -1029,7 +1023,6 @@ def _update_axes_bounds(
     xaxis = {"range": x_range}
     yaxis = {"range": y_range}
     zaxis = {"range": z_range}
-    # pyre-fixme[16]: `Scene` has no attribute `update`.
     current_layout.update({"xaxis": xaxis, "yaxis": yaxis, "zaxis": zaxis})
 
 

setup.py

@@ -153,7 +153,7 @@ setup(
     )
     + [trainer],
     package_dir={trainer: "projects/implicitron_trainer"},
-    install_requires=["iopath"],
+    install_requires=["fvcore", "iopath"],
     extras_require={
         "all": ["matplotlib", "tqdm>4.29.0", "imageio", "ipywidgets"],
         "dev": ["flake8", "usort"],

tests/test_chamfer.py

@@ -847,85 +847,6 @@ class TestChamfer(TestCaseMixin, unittest.TestCase):
             loss, loss_norm, pred_loss[0], pred_loss_norm[0], p1, p11, p2, p22
         )
 
-    def test_chamfer_point_reduction_max(self):
-        """
-        Compare output of vectorized chamfer loss with naive implementation
-        for point_reduction = "max" and batch_reduction = None.
-        """
-        N, P1, P2 = 7, 10, 18
-        device = get_random_cuda_device()
-        points_normals = TestChamfer.init_pointclouds(N, P1, P2, device)
-        p1 = points_normals.p1
-        p2 = points_normals.p2
-        weights = points_normals.weights
-        p11 = p1.detach().clone()
-        p22 = p2.detach().clone()
-        p11.requires_grad = True
-        p22.requires_grad = True
-
-        pred_loss, unused_pred_loss_norm = TestChamfer.chamfer_distance_naive(
-            p1, p2, x_normals=None, y_normals=None
-        )
-
-        loss, loss_norm = chamfer_distance(
-            p11,
-            p22,
-            x_normals=None,
-            y_normals=None,
-            weights=weights,
-            batch_reduction=None,
-            point_reduction="max",
-        )
-        pred_loss_max = torch.maximum(
-            pred_loss[0].max(1).values, pred_loss[1].max(1).values
-        )
-        pred_loss_max *= weights
-        self.assertClose(loss, pred_loss_max)
-
-        self.assertIsNone(loss_norm)
-
-        # Check gradients
-        self._check_gradients(loss, loss_norm, pred_loss_max, None, p1, p11, p2, p22)
-
-    def test_single_directional_chamfer_point_reduction_max(self):
-        """
-        Compare output of vectorized single directional chamfer loss with naive implementation
-        for point_reduction = "max" and batch_reduction = None.
-        """
-        N, P1, P2 = 7, 10, 18
-        device = get_random_cuda_device()
-        points_normals = TestChamfer.init_pointclouds(N, P1, P2, device)
-        p1 = points_normals.p1
-        p2 = points_normals.p2
-        weights = points_normals.weights
-        p11 = p1.detach().clone()
-        p22 = p2.detach().clone()
-        p11.requires_grad = True
-        p22.requires_grad = True
-
-        pred_loss, unused_pred_loss_norm = TestChamfer.chamfer_distance_naive(
-            p1, p2, x_normals=None, y_normals=None
-        )
-
-        loss, loss_norm = chamfer_distance(
-            p11,
-            p22,
-            x_normals=None,
-            y_normals=None,
-            weights=weights,
-            batch_reduction=None,
-            point_reduction="max",
-            single_directional=True,
-        )
-        pred_loss_max = pred_loss[0].max(1).values
-        pred_loss_max *= weights
-        self.assertClose(loss, pred_loss_max)
-
-        self.assertIsNone(loss_norm)
-
-        # Check gradients
-        self._check_gradients(loss, loss_norm, pred_loss_max, None, p1, p11, p2, p22)
-
     def _check_gradients(
         self,
         loss,
@@ -1099,9 +1020,9 @@ class TestChamfer(TestCaseMixin, unittest.TestCase):
         with self.assertRaisesRegex(ValueError, "batch_reduction must be one of"):
             chamfer_distance(p1, p2, weights=weights, batch_reduction="max")
 
-        # Error when point_reduction is not in ["mean", "sum", "max"] or None.
+        # Error when point_reduction is not in ["mean", "sum"] or None.
         with self.assertRaisesRegex(ValueError, "point_reduction must be one of"):
-            chamfer_distance(p1, p2, weights=weights, point_reduction="min")
+            chamfer_distance(p1, p2, weights=weights, point_reduction="max")
 
     def test_incorrect_weights(self):
         N, P1, P2 = 16, 64, 128