0c4df216d7
* update e2e * tokenizer tests * more tests, remove unnecessary code * fix e2e setting * add tests for loading model * update workflow * fixed test |
||
|---|---|---|
| .. | ||
| ann | ||
| app | ||
| export | ||
| .dockerignore | ||
| .gitignore | ||
| Dockerfile | ||
| locustfile.py | ||
| log_conf.json | ||
| poetry.lock | ||
| pyproject.toml | ||
| README_es_ES.md | ||
| README_fr_FR.md | ||
| README.md | ||
| responses.json | ||
| start.sh | ||
Immich Machine Learning
- CLIP embeddings
- Facial recognition
Setup
This project uses Poetry, so be sure to install it first.
Running poetry install --no-root --with dev --with cpu will install everything you need in an isolated virtual environment.
CUDA and OpenVINO are supported as acceleration APIs. To use them, you can replace --with cpu with either of --with cuda or --with openvino.
To add or remove dependencies, you can use the commands poetry add $PACKAGE_NAME and poetry remove $PACKAGE_NAME, respectively.
Be sure to commit the poetry.lock and pyproject.toml files with poetry lock --no-update to reflect any changes in dependencies.
Load Testing
To measure inference throughput and latency, you can use Locust using the provided locustfile.py.
Locust works by querying the model endpoints and aggregating their statistics, meaning the app must be deployed.
You can change the models or adjust options like score thresholds through the Locust UI.
To get started, you can simply run locust --web-host 127.0.0.1 and open localhost:8089 in a browser to access the UI. See the Locust documentation for more info on running Locust.
Note that in Locust's jargon, concurrency is measured in users, and each user runs one task at a time. To achieve a particular per-endpoint concurrency, multiply that number by the number of endpoints to be queried. For example, if there are 3 endpoints and you want each of them to receive 8 requests at a time, you should set the number of users to 24.