Plankton ML Documentation

Machine learning pipelines, embeddings search and backends for plankton image classification. Experimental in nature, designed to inform future work.

Features

• FastAPI service for image embeddings

• Label Studio ML Backend for pre-annotations

• Luigi pipeline for image processing

• DVC pipeline for model training

Contents

Contents:

• Installation
  • Using pip
  • Using conda
• Quick Start
  • Running the API
  • Label Studio Backend
• Label Studio
  • ML Backend notes
  • Return format
  • Connection to Label Studio
  • Label Studio analytics
• src
  • cyto_ml package
• cyto_ml.models package
  • Submodules
  • cyto_ml.models.api module
  • cyto_ml.models.utils module
  • Module contents
• cyto_ml.data package
  • Submodules
  • cyto_ml.data.db_config module
  • cyto_ml.data.flowcam module
  • cyto_ml.data.image module
  • cyto_ml.data.labels module
  • cyto_ml.data.s3 module
  • cyto_ml.data.vectorstore module
  • Module contents
• cyto_ml.pipeline package
  • Submodules
  • cyto_ml.pipeline.pipeline_decollage module
  • Module contents
• cyto_ml.visualisation package
  • Submodules
  • cyto_ml.visualisation.app module
  • cyto_ml.visualisation.config module
  • Module contents

Indices and tables

• Index

• Module Index

• Search Page