https://github.com/codekow/demo-computer-vision

Simple computer vision demo for edge deployment
https://github.com/codekow/demo-computer-vision

Last synced: 8 months ago
JSON representation

Simple computer vision demo for edge deployment

• Host: GitHub
• URL: https://github.com/codekow/demo-computer-vision
• Owner: codekow
• Created: 2022-06-01T16:01:31.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2024-10-16T14:36:29.000Z (over 1 year ago)
• Last Synced: 2025-03-16T16:51:45.145Z (over 1 year ago)
• Language: Jupyter Notebook
• Homepage:
• Size: 51.3 MB
• Stars: 4
• Watchers: 3
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Computer Vision Demo

[![Check Spelling](https://github.com/codekow/demo-computer-vision/actions/workflows/spellcheck.yaml/badge.svg)](https://github.com/codekow/demo-computer-vision/actions/workflows/spellcheck.yaml)

[![File Linting](https://github.com/redhat-na-ssa/simplevis/actions/workflows/linting.yaml/badge.svg)](https://github.com/redhat-na-ssa/simplevis/actions/workflows/linting.yaml)

This repo helps to demonstrate the use of computer vision, containers, and openshift

## Quickstart

Build model server via source

s2i strategy: `source`

```sh

APP_NAME=model-serving-cv

oc new-app \

  https://github.com/redhat-na-ssa/demo-computer-vision.git#peer-review \

  --name ${APP_NAME} \

  --strategy source \

  --context-dir /components/model

oc label all --all app=demo-computer-vision

```

Build model server via `Dockerfile`

s2i strategy: `docker`

```sh

APP_NAME=model-serving-cv-dockerfile

oc new-app \

  https://github.com/redhat-na-ssa/demo-computer-vision.git#peer-review \

  --label app=demo-computer-vision \

  --name ${APP_NAME} \

  --strategy docker \

  --context-dir /components/model

```

Expose API / model server - Route

```sh

oc expose service \

  ${APP_NAME} \

  --label app=demo-computer-vision \

  --port 8080 \

  --overrides='{"spec":{"tls":{"termination":"edge"}}}'

```

Build camera capture via `Dockerfile`

s2i strategy: `docker`

```sh

APP_NAME=camera-capture-cv

oc new-app \

  https://github.com/redhat-na-ssa/demo-computer-vision.git#peer-review \

  --label app=demo-computer-vision \

  --name ${APP_NAME} \

  --strategy docker \

  --context-dir /components/camera

```

Setup Liveness Probe

```sh

oc set probe deploy/${APP_NAME} \

  --liveness \

  --get-url=http://:8080/healthz

```

## TODO

- [ ] tell story

- [ ] address dependencies