Skip to content

[Performance] Accelerate slice sampler on GPU #6092

[Performance] Accelerate slice sampler on GPU

[Performance] Accelerate slice sampler on GPU #6092

Workflow file for this run

name: Continuous Benchmark (PR)
on:
pull_request:
permissions: write-all
concurrency:
# Documentation suggests ${{ github.head_ref }}, but that's only available on pull_request/pull_request_target triggers, so using ${{ github.ref }}.
# On master, we want all builds to complete even if merging happens faster to make it easier to discover at which point something broke.
group: ${{ github.workflow }}-${{ github.ref == 'refs/heads/main' && format('ci-master-{0}', github.sha) || format('ci-{0}', github.ref) }}
cancel-in-progress: true
jobs:
benchmark_cpu:
name: CPU Pytest benchmark
runs-on: ubuntu-20.04
steps:
- name: Who triggered this?
run: |
echo "Action triggered by ${{ github.event.pull_request.html_url }}"
- name: Checkout
uses: actions/checkout@v3
with:
fetch-depth: 50 # this is to make sure we obtain the target base commit
- name: Python Setup
uses: actions/setup-python@v4
with:
python-version: '3.10'
- name: Setup benchmarks
run: |
echo "BASE_SHA=$(echo ${{ github.event.pull_request.base.sha }} | cut -c1-8)" >> $GITHUB_ENV
echo "HEAD_SHA=$(echo ${{ github.event.pull_request.head.sha }} | cut -c1-8)" >> $GITHUB_ENV
echo "BASELINE_JSON=$(mktemp)" >> $GITHUB_ENV
echo "CONTENDER_JSON=$(mktemp)" >> $GITHUB_ENV
echo "PR_COMMENT=$(mktemp)" >> $GITHUB_ENV
- name: Setup Environment and tests
run: |
python3.10 -m venv ./py310
source ./py310/bin/activate
python3 -m pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cpu -U
python3 -m pip install git+https://github.com/pytorch/tensordict
python3 setup.py develop
python3 -m pip install pytest pytest-benchmark
python3 -m pip install "gym[accept-rom-license,atari]"
python3 -m pip install "dm_control" "mujoco"
cd benchmarks/
export TORCHDYNAMO_INLINE_INBUILT_NN_MODULES=1
export TD_GET_DEFAULTS_TO_NONE=1
RUN_BENCHMARK="python3 -m pytest -vvv --rank 0 --ignore test_collectors_benchmark.py --benchmark-json "
git checkout ${{ github.event.pull_request.base.sha }}
$RUN_BENCHMARK ${{ env.BASELINE_JSON }}
git checkout ${{ github.event.pull_request.head.sha }}
$RUN_BENCHMARK ${{ env.CONTENDER_JSON }}
- name: Publish results
uses: apbard/pytest-benchmark-commenter@v3
with:
token: ${{ secrets.GITHUB_TOKEN }}
benchmark-file: ${{ env.CONTENDER_JSON }}
comparison-benchmark-file: ${{ env.BASELINE_JSON }}
benchmark-metrics: 'name,max,mean,ops'
comparison-benchmark-metric: 'ops'
comparison-higher-is-better: true
comparison-threshold: 5
benchmark-title: 'Result of CPU Benchmark Tests'
benchmark_gpu:
name: GPU Pytest benchmark
runs-on: linux.g5.4xlarge.nvidia.gpu
defaults:
run:
shell: bash -l {0}
container:
image: nvidia/cuda:12.4.1-cudnn-runtime-ubuntu22.04
options: --gpus all
steps:
- name: Set GITHUB_BRANCH environment variable
run: |
if [ "${{ github.event_name }}" == "push" ]; then
export GITHUB_BRANCH=${{ github.event.branch }}
elif [ "${{ github.event_name }}" == "pull_request" ]; then
export GITHUB_BRANCH=${{ github.event.pull_request.head.ref }}
else
echo "Unsupported event type"
exit 1
fi
echo "GITHUB_BRANCH=$GITHUB_BRANCH" >> $GITHUB_ENV
- name: Who triggered this?
run: |
echo "Action triggered by ${{ github.event.pull_request.html_url }}"
- name: Check ldd --version
run: ldd --version
- name: Checkout
uses: actions/checkout@v3
with:
fetch-depth: 50 # this is to make sure we obtain the target base commit
- name: Python Setup
uses: actions/setup-python@v4
with:
python-version: '3.10'
- name: Setup Environment
run: |
export TZ=Europe/London
export DEBIAN_FRONTEND=noninteractive # tzdata bug
apt-get update -y
apt-get install software-properties-common -y
add-apt-repository ppa:git-core/candidate -y
apt-get update -y
apt-get upgrade -y
apt-get -y install libglu1-mesa libgl1-mesa-glx libosmesa6 gcc curl g++ unzip wget libglfw3-dev libgles2-mesa-dev libglew-dev sudo git cmake libz-dev libpython3.10-dev
- name: Setup git
run: git config --global --add safe.directory /__w/rl/rl
- name: setup Path
run: |
echo /usr/local/bin >> $GITHUB_PATH
- name: Setup benchmarks
run: |
echo "BASE_SHA=$(echo ${{ github.event.pull_request.base.sha }} | cut -c1-8)" >> $GITHUB_ENV
echo "HEAD_SHA=$(echo ${{ github.event.pull_request.head.sha }} | cut -c1-8)" >> $GITHUB_ENV
echo "BASELINE_JSON=$(mktemp)" >> $GITHUB_ENV
echo "CONTENDER_JSON=$(mktemp)" >> $GITHUB_ENV
echo "PR_COMMENT=$(mktemp)" >> $GITHUB_ENV
- name: Run
run: |
python3.10 -m venv --system-site-packages ./py310
source ./py310/bin/activate
export PYTHON_INCLUDE_DIR=/usr/include/python3.10
python3.10 -m pip install --pre torch torchvision --index-url https://download.pytorch.org/whl/nightly/cu124 -U
python3.10 -m pip install cmake ninja pytest pytest-benchmark mujoco dm_control "gym[accept-rom-license,atari]"
python3.10 -m pip install git+https://github.com/pytorch/tensordict
python3.10 setup.py develop
# python3.10 -m pip install git+https://github.com/pytorch/rl@$GITHUB_BRANCH
# test import
python3 -c """import torch
assert torch.cuda.device_count()
"""
cd benchmarks/
export TORCHDYNAMO_INLINE_INBUILT_NN_MODULES=1
export TD_GET_DEFAULTS_TO_NONE=1
RUN_BENCHMARK="python3 -m pytest -vvv --rank 0 --ignore test_collectors_benchmark.py --benchmark-json "
git checkout ${{ github.event.pull_request.base.sha }}
$RUN_BENCHMARK ${{ env.BASELINE_JSON }}
git checkout ${{ github.event.pull_request.head.sha }}
$RUN_BENCHMARK ${{ env.CONTENDER_JSON }}
- name: Publish results
uses: apbard/pytest-benchmark-commenter@v3
env:
GIT_WORK_TREE: /__w/rl/rl
with:
token: ${{ secrets.GITHUB_TOKEN }}
benchmark-file: ${{ env.CONTENDER_JSON }}
comparison-benchmark-file: ${{ env.BASELINE_JSON }}
benchmark-metrics: 'name,max,mean,ops'
comparison-benchmark-metric: 'ops'
comparison-higher-is-better: true
comparison-threshold: 5
benchmark-title: 'Result of GPU Benchmark Tests'