423A35C7/pytorch3d
1
0
Fork 0
mirror of https://github.com/facebookresearch/pytorch3d.git synced 2025-12-14 11:26:24 +08:00
Code Issues Packages Projects Releases Wiki Activity

pulsar build and CI changes

Browse Source 
Summary:
Changes to CI and some minor fixes now that pulsar is part of pytorch3d. Most significantly, add CUB to CI builds.

Make CUB_HOME override the CUB already in cudatoolkit (important for cuda11.0 which uses cub 1.9.9 which pulsar doesn't work well with.
Make imageio available for testing.
Lint fixes.
Fix some test verbosity.
Avoid use of atomicAdd_block on older GPUs.

Reviewed By: nikhilaravi, classner

Differential Revision: D24773716

fbshipit-source-id: 2428356bb2e62735f2bc0c15cbe4cff35b1b24b8
This commit is contained in:
Jeremy Reizenstein
2020-11-10 09:36:29 -08:00
committed by Facebook GitHub Bot
parent 804235b05a
commit d220ee2f66
14 changed files with 88 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
.circleci/config.in.yml
View File

@@ -68,14 +68,26 @@ jobs:
steps:
- checkout
- <<: *setupcuda
- run: pip3 install --progress-bar off wheel matplotlib 'pillow<7'
- run: pip3 install --progress-bar off imageio wheel matplotlib 'pillow<7'
- run: pip3 install --progress-bar off torch torchvision
# - run: conda create -p ~/conda_env python=3.7 numpy
# - run: conda activate ~/conda_env
# - run: conda install -c pytorch pytorch torchvision
- run: pip3 install --progress-bar off 'git+https://github.com/facebookresearch/fvcore'
- run: LD_LIBRARY_PATH=$LD_LIBARY_PATH:/usr/local/cuda-10.2/lib64 python3 setup.py build_ext --inplace
- run:
name: get cub
command: |
cd ..
wget --no-verbose https://github.com/NVIDIA/cub/archive/1.10.0.tar.gz
tar xzf 1.10.0.tar.gz
# This expands to a directory called cub-1.10.0
- run:
name: build
command: |
export LD_LIBRARY_PATH=$LD_LIBARY_PATH:/usr/local/cuda-10.2/lib64
export CUB_HOME=$(realpath ../cub-1.10.0)
python3 setup.py build_ext --inplace
- run: LD_LIBRARY_PATH=$LD_LIBARY_PATH:/usr/local/cuda-10.2/lib64 python -m unittest discover -v -s tests
- run: python3 setup.py bdist_wheel
@@ -89,7 +101,7 @@ jobs:
resource_class: 2xlarge+
steps:
- checkout
- run: packaging/build_wheel.sh
- run: MAX_JOBS=15 packaging/build_wheel.sh
- store_artifacts:
path: dist
- persist_to_workspace:
@@ -109,7 +121,7 @@ jobs:
- checkout
# This is building with cuda but no gpu present,
# so we aren't running the tests.
- run: TEST_FLAG=--no-test packaging/build_conda.sh
- run: MAX_JOBS=15 TEST_FLAG=--no-test packaging/build_conda.sh
- store_artifacts:
path: /opt/conda/conda-bld/linux-64
- persist_to_workspace:
@@ -215,9 +227,9 @@ workflows:
context: DOCKERHUB_TOKEN
{{workflows()}}
- binary_linux_conda_cuda:
name: testrun_conda_cuda_py37_cu101_pyt14
name: testrun_conda_cuda_py36_cu101_pyt14
context: DOCKERHUB_TOKEN
python_version: "3.7"
python_version: "3.6"
pytorch_version: "1.4"
cu_version: "cu101"
- binary_linux_conda_cuda:

26
.circleci/config.yml
View File

@@ -68,14 +68,26 @@ jobs:
steps:
- checkout
- <<: *setupcuda
- run: pip3 install --progress-bar off wheel matplotlib 'pillow<7'
- run: pip3 install --progress-bar off imageio wheel matplotlib 'pillow<7'
- run: pip3 install --progress-bar off torch torchvision
# - run: conda create -p ~/conda_env python=3.7 numpy
# - run: conda activate ~/conda_env
# - run: conda install -c pytorch pytorch torchvision
- run: pip3 install --progress-bar off 'git+https://github.com/facebookresearch/fvcore'
- run: LD_LIBRARY_PATH=$LD_LIBARY_PATH:/usr/local/cuda-10.2/lib64 python3 setup.py build_ext --inplace
- run:
name: get cub
command: |
cd ..
wget --no-verbose https://github.com/NVIDIA/cub/archive/1.10.0.tar.gz
tar xzf 1.10.0.tar.gz
# This expands to a directory called cub-1.10.0
- run:
name: build
command: |
export LD_LIBRARY_PATH=$LD_LIBARY_PATH:/usr/local/cuda-10.2/lib64
export CUB_HOME=$(realpath ../cub-1.10.0)
python3 setup.py build_ext --inplace
- run: LD_LIBRARY_PATH=$LD_LIBARY_PATH:/usr/local/cuda-10.2/lib64 python -m unittest discover -v -s tests
- run: python3 setup.py bdist_wheel
@@ -89,7 +101,7 @@ jobs:
resource_class: 2xlarge+
steps:
- checkout
- run: packaging/build_wheel.sh
- run: MAX_JOBS=15 packaging/build_wheel.sh
- store_artifacts:
path: dist
- persist_to_workspace:
@@ -109,7 +121,7 @@ jobs:
- checkout
# This is building with cuda but no gpu present,
# so we aren't running the tests.
- run: TEST_FLAG=--no-test packaging/build_conda.sh
- run: MAX_JOBS=15 TEST_FLAG=--no-test packaging/build_conda.sh
- store_artifacts:
path: /opt/conda/conda-bld/linux-64
- persist_to_workspace:
@@ -489,6 +501,12 @@ workflows:
python_version: "3.6"
pytorch_version: "1.4"
cu_version: "cu101"
- binary_linux_conda_cuda:
name: testrun_conda_cuda_py37_cu102_pyt160
context: DOCKERHUB_TOKEN
python_version: "3.7"
pytorch_version: '1.6.0'
cu_version: "cu102"
- binary_linux_conda_cuda:
name: testrun_conda_cuda_py37_cu110_pyt170
context: DOCKERHUB_TOKEN