Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
174 commits
Select commit Hold shift + click to select a range
c851274
remove workflow from nightly benchmark (#2114)
lxning Feb 9, 2023
48c8153
updated walmart blog link (#2117)
agunapal Feb 10, 2023
30ec515
Changing to previous version of mdutils (#2127)
agunapal Feb 14, 2023
7e65972
Add configuration option to disable system metrics collection (#2104)
namannandan Feb 14, 2023
c84312b
Freeze pynvml version to avoid crash in nvgpu https://github.com/pyto…
mreso Feb 15, 2023
7a7cc69
Bump isort dependency for pre-commit to 5.11.5 (#2132)
mreso Feb 16, 2023
8b3ae1e
Update CI GPU to run on CUDA 11.7 (#2137)
agunapal Feb 16, 2023
485ebf8
Enable batch processing in scriptable tokenizer example (#2130)
mreso Feb 16, 2023
c417b4a
README for serving models using TorchServe Docker Container (#2118)
agunapal Feb 16, 2023
92657b2
Add Github Code Scanning codeql.yml (#2149)
msaroufim Feb 21, 2023
b8f52d8
upgrade CI to run on Ubuntu 20.04 (#2153)
agunapal Feb 21, 2023
26089d4
Add support for building docker images with specified python version …
agunapal Feb 23, 2023
87359c4
Update model file docs to be more accurate (#2148)
tmc Feb 24, 2023
b05c0ea
Added feature that allows for wildcard search in --extra-files argume…
gustavhartz Feb 27, 2023
86d4400
Fix Dead kubectl links (#2160)
msaroufim Feb 27, 2023
fd8f1b3
Push benchmark artifacts for auto-validation (#2157)
agunapal Mar 7, 2023
9412c0a
Make handler tests callable with pytest (#2173)
mreso Mar 13, 2023
1768902
Feat: Dali integration (#1958)
Mar 20, 2023
6daaa42
BERT nightly benchmark on Inferentia1 (#2167)
namannandan Mar 20, 2023
c164374
Make model-archiver tests runnable from any directory (#2191)
mreso Mar 22, 2023
890d09e
Upgrade to PyTorch 2.0 (#2194)
agunapal Mar 24, 2023
82acbbf
NVIDIA MPS documentation (#2184)
mreso Mar 27, 2023
41a3af3
Fix regex formatting (#2172)
mpoemsl Mar 27, 2023
d0510ba
Feature/grpc streaming (#2186)
lxning Mar 28, 2023
c70c501
Add zip-store option to model-archiver tool
mreso Mar 22, 2023
b6ad061
Moved test loading function of model-archiver integration test into c…
mreso Mar 23, 2023
8111e29
Added additional test to model-archiver unit test
mreso Mar 23, 2023
afe8a9b
Added zip-store help to README.md
mreso Mar 28, 2023
7d23cb2
Add NVIDIA MPS documentation to doc index
mreso Mar 28, 2023
6b5d8c8
Add mps doc to content + change title
mreso Mar 28, 2023
d662c26
platform framework (#2193)
lxning Mar 29, 2023
4ca86f3
Add frontend metrics to default metrics configuration (#2164)
namannandan Mar 31, 2023
c9fbc7c
support torchrun and optimize gpu assignment (#2209)
lxning Mar 31, 2023
8b1ae4b
Enable regression tests to run daily (#2211)
agunapal Apr 3, 2023
237675f
Refactor Dockerfile: Fix Python version; Add CI workflow to test Dock…
fabridamicelli Apr 3, 2023
98ca286
TorchServe decompress tgz format model (#2214)
lxning Apr 3, 2023
a4a1f65
Added test cases for SageMaker MME requirements (#2200)
agunapal Apr 4, 2023
c37da18
Make sanity test output useable again
mreso Apr 3, 2023
4ea172d
feat: add PyTorch/XLA support (#2182)
morgandu Apr 11, 2023
2aee8fb
Frontend IMetrics, LogMetric and PrometheusMetric (#2165)
lxning Apr 11, 2023
a460afb
Remove generation of graph plots from benchmark runs (#2228)
agunapal Apr 12, 2023
9edd461
Enable worker core pinning in CPU nightly benchmark (#2166)
min-jean-cho Apr 13, 2023
1af13ae
Fix and add test for `build_image.sh`: Make it invariant to arguments…
fabridamicelli Apr 14, 2023
fbdb26e
fix setup (#2237)
min-jean-cho Apr 17, 2023
2fa042b
[easy] lint fix broken windows conda link (#2240)
msaroufim Apr 19, 2023
cf7544b
Refactor PT2 code changes (#2222)
msaroufim Apr 19, 2023
dd8e792
Changed docker image to production (#2227)
agunapal Apr 20, 2023
b03ac7d
Frontend metrics configuration handling (#2190)
namannandan Apr 20, 2023
5974c71
move config-file arg from torchserve to torch-model-archiver (#2244)
samils7 Apr 20, 2023
9fa08a3
Revert "[easy] lint fix broken windows conda link (#2240)" (#2245)
agunapal Apr 20, 2023
e4dc84a
fixes error thrown while loading multiple models (#2235)
Apr 20, 2023
067fd91
Automate benchmark validation (#2144)
agunapal Apr 20, 2023
fa95a66
Make pre-commit usage clearer in error message (#2241)
msaroufim Apr 20, 2023
044bbc1
Updates to TorchServe page on pytorch.org (#2243)
agunapal Apr 20, 2023
d23d2ab
Large model inference (#2215)
HamidShojanazeri Apr 20, 2023
70c5712
Refactored model to validate into json file
mreso Apr 4, 2023
4dc1983
Refactored sanity tests
mreso Apr 4, 2023
99ac63f
fix (#2250)
msaroufim Apr 21, 2023
627d4b7
open github issue on regression failure (#2248)
agunapal Apr 21, 2023
e2b4511
enable torch.compile support for torch 2.0.0 pre release (#2256)
morgandu Apr 21, 2023
419edb6
http stream response via http 1.1 chunked encoding (#2233)
lxning Apr 21, 2023
03fbca6
update ping endpoint default behavior (#2254)
lxning Apr 21, 2023
a9e218a
Increase collection of benchmark artifacts to 30 for better averaging…
agunapal Apr 21, 2023
49a7993
deepspeed base handler and example (#2218)
lxning Apr 24, 2023
542385e
support client timeout (#2267)
lxning Apr 25, 2023
0ea5ae5
FAQ on slow model initialization (#2246)
msaroufim Apr 25, 2023
92bd04b
Skip dist inference test (#2279)
HamidShojanazeri Apr 25, 2023
4e08fd1
Add metric cache and migrate existing metrics to cache implementation…
lxning Apr 25, 2023
d6e072a
Refactoring dependencies for installing TorchServe (#2247)
agunapal Apr 25, 2023
1707a74
Fix dimension names and documentation for metrics with prior promethe…
namannandan Apr 27, 2023
5ee02e4
Revert "open github issue on regression failure (#2248)" (#2315)
msaroufim May 3, 2023
614bfc0
upgrade transformers version (#2325)
namannandan May 5, 2023
2f1f52f
Pippy deferred init (#2310)
HamidShojanazeri May 10, 2023
4fe5273
add omp in env for torchrun and update doc (#2320)
lxning May 10, 2023
fbcde1d
remove redundent Download_models.py (#2331)
lxning May 10, 2023
3676186
Updated version for v0.8.0 (#2335)
agunapal May 11, 2023
8ea1dab
HF optimum is installed by default in HF Transformers example (#2336)
agunapal May 11, 2023
35fb574
Issues/fix docker dependencies (#2340)
agunapal May 11, 2023
f01868f
Updating index.md to fix 8 broken links (#2329)
sekyondaMeta May 15, 2023
25f3700
BERT nightly benchmark on Inferentia2 (#2283)
namannandan May 16, 2023
77ca82d
fix: kserve fastapi migration issues (#2175)
May 19, 2023
4450287
Fixing FAQs doc per issue #2204 (#2351)
sekyondaMeta May 22, 2023
1fc15b3
Hardcode branch name
mpoemsl Mar 1, 2023
ba5097d
Add priority attribute which is set via X-Priority header
mpoemsl Mar 1, 2023
fa3c57a
Implement inbuilt metric RequestPriority
mpoemsl Mar 1, 2023
cffa44a
Add RequestPriority to registry
mpoemsl Mar 1, 2023
7bdb11c
Implement PriorityLinkedBlockingDeque with same interface as LinkedBl…
mpoemsl Mar 1, 2023
b64d7d9
Implement isEmpty correctly
mpoemsl Mar 1, 2023
6c59d08
Implement sampled polling, but this does not make sense yet
mpoemsl Mar 1, 2023
1cfbac5
Override unlinkFirst()
mpoemsl Mar 2, 2023
fbc19ba
Implement working polling & offering (to be tested)
mpoemsl Mar 2, 2023
d21a1fa
Adjust queue rejection error message to include priority value
mpoemsl Mar 3, 2023
13f9201
Remove serialization boilerplate
mpoemsl Mar 3, 2023
a465e5a
Implement number of priorities parameter
mpoemsl Mar 3, 2023
8a8ed25
Remove debug statements
mpoemsl Mar 3, 2023
5f13da6
Add newline
mpoemsl Mar 3, 2023
4bfbf28
Remove unused imports
mpoemsl Mar 3, 2023
ce3344a
Revert branch hardcoding
mpoemsl Mar 3, 2023
1cab897
Fix install from source
mpoemsl Mar 3, 2023
e7afbb9
Expand documentation
mpoemsl Mar 6, 2023
2ec63c7
Prefix header with X-TS- to avoid collision with other X-Priority hea…
mpoemsl Mar 6, 2023
c602f4c
Add reference to LinkedBlockingDeque where appropriate
mpoemsl Mar 6, 2023
a6c8add
:fire: Remove pygit2
pypae Mar 6, 2023
b789611
:whale: Pass git branch
pypae Mar 6, 2023
9a8ad88
Implement setting and getting request ids via X-TS-Request-ID header
mpoemsl Mar 16, 2023
e6694fc
Adjust header casing
mpoemsl Mar 17, 2023
7b7928d
Add GPUManager
mpoemsl Mar 17, 2023
fb5a370
Fix GPUManager initialization
mpoemsl Mar 17, 2023
7bdd714
Implement GPU failure tracking
mpoemsl Mar 20, 2023
7ccca1b
Implement free memory querying
mpoemsl Mar 20, 2023
811fed7
Implement random selection weighted by free memory
mpoemsl Mar 20, 2023
09bc4a1
Handle case of none or only one eligible GPU
mpoemsl Mar 20, 2023
d2696d4
Pass GPUManager parameters through
mpoemsl Mar 20, 2023
f08104c
Make priorities into categorical values 'low', 'high', 'max'
mpoemsl Mar 17, 2023
be20578
Adapt RequestPriority metric to categorical value
mpoemsl Mar 20, 2023
a83c8f6
Make priority enum
mpoemsl Mar 20, 2023
641774a
Validate high priority probability
mpoemsl Mar 20, 2023
a296201
Shorten empty check
mpoemsl Mar 20, 2023
11b6110
Remove unnecessary queueSize storage
mpoemsl Mar 20, 2023
6b39d18
Make private attributes final
mpoemsl Mar 20, 2023
1fb4910
Refactor random deque selection
mpoemsl Mar 20, 2023
624f108
Remove unused import
mpoemsl Mar 22, 2023
181f549
Remove unused import
mpoemsl Mar 22, 2023
84c9d53
Make variables final
mpoemsl Mar 22, 2023
eddb670
Improve log formatting
mpoemsl Mar 22, 2023
bb86317
Remove dtype specification
mpoemsl Mar 22, 2023
92f4247
Improve logging syntax
mpoemsl Mar 22, 2023
48a9419
Implement fixed-length gpu failure history
mpoemsl Mar 22, 2023
9ed4996
Make public methods synchronized and replace concurrent data structures
mpoemsl Mar 22, 2023
46969a3
Fix divide by zero on empty failure history
mpoemsl Mar 23, 2023
330647e
Add debug statements
mpoemsl Mar 23, 2023
84a4aa7
Fix ceiling semantics
mpoemsl Mar 23, 2023
33cd1df
Adopt x-request-id value has prefix for internal request id to ensure…
mpoemsl Mar 27, 2023
d9e9c24
Implement regularly polled and logged queue status metric
mpoemsl Mar 30, 2023
ef24404
Add max queue size to status string
mpoemsl Mar 30, 2023
a9a9066
Add queueSize as last entry
mpoemsl Mar 30, 2023
68e27e2
Resolve conflicts introduced during rebase
simonschoelly May 24, 2023
549631e
:whale: Update Dockerfile.dev with python version arg
pypae May 22, 2023
2b766c7
:wrench: Fix update-alternatives
pypae May 22, 2023
93a52b7
:wrench: Add back ssh agent
pypae May 22, 2023
fe78550
:wrench: Make progress plain again
pypae May 23, 2023
4f85477
Micro batching example (#2210)
mreso May 25, 2023
63ebd4d
Install torch before other ts dependencies
mreso May 24, 2023
680b6fd
Make use of venv
mreso May 25, 2023
5f75710
move torch dependencies into separate file
mreso May 25, 2023
ffa6847
Remove deprecated pygit2 installation
mreso May 25, 2023
4bc5e46
Update custom images to new branch naming
mpoemsl Mar 17, 2023
10b81b5
Add support for building docker images with specified python version …
agunapal Feb 23, 2023
71518bb
Make dockerfile work
simonschoelly May 29, 2023
4492505
Fix merge conflict in Dockerfile
simonschoelly May 30, 2023
a568291
Avoid running processes in integration test
mreso Apr 1, 2023
d229611
Fix error case
mreso Apr 1, 2023
c816465
Remove test cross-talking
mreso Apr 3, 2023
81c3d59
Enable Regression Test in CI (#2370)
msaroufim May 30, 2023
130a29e
Fix regression test failures associated with fixes in PR: https://git…
namannandan May 31, 2023
1f863f9
Remove evil session_mocker usage to prevent test cross talking
mreso May 31, 2023
27a6ed5
Enable ONNX test in CI (#2363)
msaroufim May 31, 2023
c685b5c
Update cpuinfo (#2372)
min-jean-cho May 31, 2023
a81fc52
fix tcp port in model_service_worker.py (#2377)
lxning Jun 1, 2023
77f8c0b
Remove cu118 from regression tests (#2380)
agunapal Jun 1, 2023
e5004b2
Update cpuinfo (#2379)
min-jean-cho Jun 2, 2023
e205e6b
Upgrade torch, torchvision, torchtext and torchaudio versions (#2374)
namannandan Jun 2, 2023
9542c45
Remove wait time when stopping and starting torchserve in tests
mreso Mar 22, 2023
9a23ef7
Remove 10 second wait time for start/stop torchserve
mreso Mar 23, 2023
fe27e07
Prepare log folder + remove spaces in command + obsolete kwarg
mreso Jun 1, 2023
28a2525
Allow --stop to be combined with --forward argument to wait for model…
mreso Jun 1, 2023
08a9093
Ts type annotations (#2384)
josephcalise Jun 6, 2023
43d84ff
fix: reflect model config in config.properties (#2382)
lxning Jun 12, 2023
580c28a
Bump transformers from 4.28.1 to 4.30.0 in /requirements (#2410)
dependabot[bot] Jun 13, 2023
7f9967e
fix model cpu config (#2408)
lxning Jun 13, 2023
89fbd88
support jobQueueSize and job ticket per model in model config YAML (#…
lxning Jun 13, 2023
6eea9e1
Merge commit '89fbd885b11e4d2f40bea779d8cee7c6a87f9c20' of github.com…
simonschoelly Jun 13, 2023
f01fab6
Use /pytorch-serve insteaf of /serve directory in dockerfile
simonschoelly Jun 14, 2023
5124b67
Merge pull request #14 from textshuttle/torchserve-23mt-v0.8.0-fix-1
pypae Jun 14, 2023
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
89 changes: 89 additions & 0 deletions .github/workflows/benchmark_nightly.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,89 @@
name: Benchmark torchserve nightly

on:
# run every day at 2:15am
schedule:
- cron: '15 02 * * *'

jobs:
nightly:
strategy:
fail-fast: false
matrix:
hardware: [cpu, gpu, inf1, inf2]
runs-on:
- self-hosted
- ${{ matrix.hardware }}
timeout-minutes: 1320
steps:
- name: Clean up previous run
run: |
echo "Cleaning up previous run"
cd $RUNNER_WORKSPACE
pwd
cd ..
pwd
rm -rf _tool
- name: Setup Python 3.8
uses: actions/setup-python@v4
with:
python-version: 3.8
architecture: x64
- name: Setup Java 17
uses: actions/setup-java@v3
with:
distribution: 'zulu'
java-version: '17'
- name: Checkout TorchServe
uses: actions/checkout@v3
- name: Install dependencies
run: |
sudo apt-get update -y
sudo apt-get install -y apache2-utils
pip install -r benchmarks/requirements-ab.txt
- name: Benchmark cpu nightly
if: ${{ matrix.hardware == 'cpu' }}
run: python benchmarks/auto_benchmark.py --input benchmarks/benchmark_config_cpu.yaml --skip false
- name: Benchmark gpu nightly
if: ${{ matrix.hardware == 'gpu' }}
run: python benchmarks/auto_benchmark.py --input benchmarks/benchmark_config_gpu.yaml --skip false
- name: Benchmark inf1 nightly
if: ${{ matrix.hardware == 'inf1' }}
env:
NEURON_RT_NUM_CORES: 4
run: python benchmarks/auto_benchmark.py --input benchmarks/benchmark_config_neuron.yaml --skip false
- name: Benchmark inf2 nightly
if: ${{ matrix.hardware == 'inf2' }}
env:
NEURON_RT_NUM_CORES: 1
run: python benchmarks/auto_benchmark.py --input benchmarks/benchmark_config_neuronx.yaml --skip false
- name: Save benchmark artifacts
uses: actions/upload-artifact@v2
with:
name: nightly ${{ matrix.hardware }} artifact
path: /tmp/ts_benchmark
- name: Download benchmark artifacts for auto validation
uses: dawidd6/action-download-artifact@v2
with:
workflow: ${{ github.event.workflow_run.workflow_id }}
workflow_conclusion: success
if_no_artifact_found: ignore
path: /tmp/ts_artifacts
name: ${{ matrix.hardware }}_benchmark_validation
- name: Validate Benchmark result
run: python benchmarks/validate_report.py --input-artifacts-dir /tmp/ts_artifacts/${{ matrix.hardware }}_benchmark_validation
- name: Update benchmark artifacts for auto validation
run: python benchmarks/utils/update_artifacts.py --output /tmp/ts_artifacts/${{ matrix.hardware }}_benchmark_validation
- name: Upload the updated benchmark artifacts for auto validation
uses: actions/upload-artifact@v2
with:
name: ${{ matrix.hardware }}_benchmark_validation
path: /tmp/ts_artifacts
- name: Open issue on failure
if: ${{ failure() && github.event_name == 'schedule' && matrix.hardware == 'cpu' }}
uses: dacbd/create-issue-action@v1
with:
token: ${{ secrets.GITHUB_TOKEN }}
title: Nightly ${{ matrix.hardware }} benchmark failed
body: Commit ${{ github.sha }} daily scheduled [CI run](https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}) failed, please check why
assignees: ''
53 changes: 0 additions & 53 deletions .github/workflows/benchmark_nightly_cpu.yml

This file was deleted.

45 changes: 0 additions & 45 deletions .github/workflows/benchmark_nightly_gpu.yml

This file was deleted.

2 changes: 1 addition & 1 deletion .github/workflows/ci_cpu.yml
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ jobs:
strategy:
fail-fast: false
matrix:
os: [ubuntu-18.04, macOS-latest]
os: [ubuntu-20.04, macOS-latest]
steps:
- name: Setup Python 3.8
uses: actions/setup-python@v4
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/ci_gpu.yml
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ jobs:
uses: actions/checkout@v3
- name: Install dependencies
run: |
python ts_scripts/install_dependencies.py --environment=dev --cuda=cu102
python ts_scripts/install_dependencies.py --environment=dev --cuda=cu117
- name: Torchserve Sanity
uses: nick-fields/retry@v2
with:
Expand Down
65 changes: 65 additions & 0 deletions .github/workflows/codeql.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,65 @@
# For most projects, this workflow file will not need changing; you simply need
# to commit it to your repository.
#
# You may wish to alter this file to override the set of languages analyzed,
# or to provide custom queries or build logic.
#
# ******** NOTE ********
# We have attempted to detect the languages in your repository. Please check
# the `language` matrix defined below to confirm you have the correct set of
# supported CodeQL languages.
#
name: "CodeQL"

on:
# run every day at 11:15am
schedule:
- cron: '15 11 * * *'

jobs:
analyze:
name: Analyze
runs-on: ubuntu-latest
permissions:
actions: read
contents: read
security-events: write

strategy:
fail-fast: false
matrix:
language: [ 'java', 'python' ]

steps:
- name: Checkout repository
uses: actions/checkout@v3

- name: Setup Python 3.8
uses: actions/setup-python@v4
with:
python-version: 3.8
architecture: x64
- name: Setup Java 17
uses: actions/setup-java@v3
with:
distribution: 'zulu'
java-version: '17'

# Initializes the CodeQL tools for scanning.
- name: Initialize CodeQL
uses: github/codeql-action/init@v2
with:
languages: ${{ matrix.language }}


- name: Build TorchServe frontend
run: |
python ts_scripts/install_dependencies.py --environment=dev
cd frontend
./gradlew build -x test
cd ..

- name: Perform CodeQL Analysis
uses: github/codeql-action/analyze@v2
with:
category: "/language:${{matrix.language}}"
42 changes: 42 additions & 0 deletions .github/workflows/docker-ci.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,42 @@
name: Docker CI

on:
push:
branches: [ "master" ]
pull_request:
branches: [ "master" ]

jobs:
test-build-and-container:
runs-on: ubuntu-latest
strategy:
fail-fast: false
matrix:
python-version: ["3.8", "3.9", "3.10"]
steps:
- uses: actions/checkout@v3

- name: Test build_image.sh script with custom tagging and gpu flag
working-directory: docker
run: ./test_build_image_tagging.sh ${{ matrix.python-version }}

- name: Build Image for container test
id: image_build
working-directory: docker
run: |
IMAGE_TAG=test-image-${{ matrix.python-version }}
./build_image.sh -py "${{ matrix.python-version }}" -t "${IMAGE_TAG}"
echo "IMAGE_TAG=${IMAGE_TAG}" >> $GITHUB_OUTPUT

- name: Container Healthcheck
working-directory: docker
run: ./test_container_health.sh ${{ steps.image_build.outputs.IMAGE_TAG }}

- name: Check Python version in container
working-directory: docker
run: ./test_container_python_version.sh ${{ steps.image_build.outputs.IMAGE_TAG }} ${{ matrix.python-version }}

- name: Test model running in container with sample image data
working-directory: docker
run: |
./test_container_model_prediction.sh ${{ steps.image_build.outputs.IMAGE_TAG }}
1 change: 1 addition & 0 deletions .github/workflows/lint.yml
Original file line number Diff line number Diff line change
Expand Up @@ -77,6 +77,7 @@ jobs:
echo "cd serve/"
echo "pre-commit install"
echo "pre-commit will lint your code for you, so git add and commit those new changes and this check should become green"
echo "If you've already pushed some files remotely then run git diff --name-only main | xargs pre-commit run --files"

spellcheck:
runs-on: ubuntu-20.04
Expand Down
12 changes: 11 additions & 1 deletion .github/workflows/regression_tests_cpu.yml
Original file line number Diff line number Diff line change
@@ -1,6 +1,16 @@
name: Run Regression Tests on CPU

on: workflow_dispatch
on:
push:
branches:
- master
pull_request:
branches:
- master

concurrency:
group: ci-cpu-${{ github.workflow }}-${{ github.ref == 'refs/heads/master' && github.run_number || github.ref }}
cancel-in-progress: true

jobs:
regression-cpu:
Expand Down
18 changes: 12 additions & 6 deletions .github/workflows/regression_tests_gpu.yml
Original file line number Diff line number Diff line change
@@ -1,15 +1,21 @@
name: Run Regression Tests on GPU

on: workflow_dispatch
on:
push:
branches:
- master
pull_request:
branches:
- master

concurrency:
group: ci-cpu-${{ github.workflow }}-${{ github.ref == 'refs/heads/master' && github.run_number || github.ref }}
cancel-in-progress: true

jobs:
regression-gpu:
# creates workflows for CUDA 11.6 & CUDA 11.7 on ubuntu
runs-on: [self-hosted, regression-test-gpu]
strategy:
fail-fast: false
matrix:
cuda: ["cu116", "cu117"]
steps:
- name: Clean up previous run
run: |
Expand Down Expand Up @@ -37,7 +43,7 @@ jobs:
uses: actions/checkout@v3
- name: Install dependencies
run: |
python ts_scripts/install_dependencies.py --environment=dev --cuda=${{ matrix.cuda }}
python ts_scripts/install_dependencies.py --environment=dev --cuda=cu117
- name: Torchserve Regression Tests
run: |
python test/regression_tests.py
2 changes: 2 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@ dist/
*.egg-info/
.idea
*htmlcov*
.cache
.coverage
.github/actions/
.github/.DS_Store
Expand All @@ -18,6 +19,7 @@ plugins/*/bin
*.backup
docs/sphinx/src/
ts_scripts/spellcheck_conf/wordlist.dic
venv/

# Postman files
test/artifacts/
Expand Down
Loading