Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/sustainable-computing-io/kepler-model-server
Model Server for Kepler
https://github.com/sustainable-computing-io/kepler-model-server
machine-learning model-server online-machine-learning prometheus stream-processing sustainability tensorflow tensorflow-io tensorflow2
Last synced: 3 days ago
JSON representation
Model Server for Kepler
- Host: GitHub
- URL: https://github.com/sustainable-computing-io/kepler-model-server
- Owner: sustainable-computing-io
- License: apache-2.0
- Created: 2022-05-05T21:36:22.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-10-29T11:42:46.000Z (about 2 months ago)
- Last Synced: 2024-10-29T13:22:51.348Z (about 2 months ago)
- Topics: machine-learning, model-server, online-machine-learning, prometheus, stream-processing, sustainability, tensorflow, tensorflow-io, tensorflow2
- Language: Python
- Homepage:
- Size: 36.5 MB
- Stars: 25
- Watchers: 6
- Forks: 25
- Open Issues: 58
-
Metadata Files:
- Readme: README.md
- Contributing: contributing.md
- License: LICENSE
Awesome Lists containing this project
README
# Kepler Power Model
[Get started with Kepler Model Server.](https://sustainable-computing.io/kepler_model_server/get_started/)
This repository contains source code related to Kepler power model. The modules in this repository connects to [core Kepler project](https://github.com/sustainable-computing-io/kepler) and [kepler-model-db](https://github.com/sustainable-computing-io/kepler-model-db) as below.
For more details, check [the component diagram](./fig/model-server-components-simplified.png).
## Model server and estimator deployment
### Using Kepler Operator
```yaml
apiVersion: kepler.system.sustainable.computing.io/v1alpha1
kind: KeplerInternal
metadata:
name: kepler
spec:
..
modelServer:
enabled:
estimator:
node:
components:
sidecar:
initUrl:
total:
sidecar:
initUrl:
```
### Using manifests with setup script
Deploy with estimator sidecar
```bash
OPTS="ESTIMATOR" make deploy
```
Deploy with estimator sidecar and model server
```bash
OPTS="ESTIMATOR SERVER" make deploy
```
## Model Training
- [Use Tekton pipeline](./model_training/tekton/README.md)
- [Use Bash script with CPE operator](./model_training/cpe_script_instruction.md)
## Local test
### Via docker
1. Build image for testing, run
```bash
make build-test
```
2. Run the test
|Test case|Command|
|---|---|
|[Training pipeline](./tests/README.md#pipeline)|make test-pipeline|
|[Model server](./tests/README.md#estimator-model-request-to-model-server)|make test-model-server|
|[Estimator](./tests/README.md#estimator-power-request-from-collector)|make test-estimator|
|[Offline Trainer](./tests/README.md#offline-trainer)|make test-offline-trainer|
For more test information, check [here](./tests/).
### With native python environment
Compatible version: `python 3.10`
1. Install [`hatch`](https://hatch.pypa.io/latest/install/)
2. Prepare environment
```bash
hatch shell
```
3. Run the test
| Test case | Command |
|-----------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| [Training pipeline](./tests/README.md#pipeline)| python -u ./tests/pipeline_test.py |
| [Model server](./tests/README.md#estimator-model-request-to-model-server)| Terminal 1: export MODEL_PATH=$(pwd)/tests/models;python src/server/model_server.py Terminal 2: python -u tests/estimator_model_request_test.py |
| [Estimator](./tests/README.md#estimator-power-request-from-collector) | Terminal 1: python src/estimate/estimator.py Terminal 2: python -u tests/estimator_power_request_test.py |
| Estimator with Model Server | Terminal 1: export MODEL_PATH=$(pwd)/tests/models;python src/server/model_server.py Terminal 2: export MODEL_SERVER_URL=;export MODEL_SERVER_ENABLE=true;python -u src/estimate/estimator.py Terminal 3: python -u tests/estimator_power_request_test.py |
| [Offline Trainer](./tests/README.md#offline-trainer) | Terminal 1: python src/train/offline_trainer.py Terminal 2: python -u tests/offline_trainer_test.py |
For more test information, check [here](./tests/).
### Contributing
Please check the roadmap and guidelines to join us [here](./contributing.md).