Quickstart
This page is the practical chooser for the main ways you can use this repository.
If you want the package-only quickstart and command explanations, use:
planktonclas: https://github.com/woutdecrop/planktonclas
Choose your path
Most users should choose one of these workflows:
Package / Local CLI
Local API
Notebooks
Docker
AI4OS / OSCAR
Option 1: Package / Local CLI
Typical local workflow:
planktonclas init my_project
planktonclas validate-config --config ./my_project/config.yaml
planktonclas train --config ./my_project/config.yaml
planktonclas report --config ./my_project/config.yaml
For package-level details and command meaning:
Next page in this docs set:
Option 2: Local API
planktonclas api --config ./my_project/config.yaml
Then open:
http://127.0.0.1:5000/uihttp://127.0.0.1:5000/api#/
For package-focused API details:
Next page in this docs set:
Option 3: Notebooks
pip install "planktonclas[notebooks]"
planktonclas notebooks my_project
This copies the packaged notebooks into my_project/notebooks/.
For package-focused notebook details:
Next page in this docs set:
Option 4: Docker
git clone https://github.com/ai4os-hub/phyto-plankton-classification
cd phyto-plankton-classification
docker run -ti -p 8888:8888 -p 5000:5000 -v "$(pwd):/srv/phyto-plankton-classification" ai4oshub/phyto-plankton-classification:latest /bin/bash
Inside the container, you can use the same planktonclas workflow.
This can also lead to the same local API workflow as Option 2 if you start the API inside the container.
Next page in this docs set:
Option 5: AI4OS / OSCAR
Use this path when you want hosted execution or deployment.
Useful links:
What comes next
After your chosen setup:
use Python Usage for Option 1 context
use API Usage for Option 2 and Option 4 details
use Notebooks for Option 3 details
use Reference for project structure, outputs, and conventions