Skip to main content

Setting up a Zooniverse project
(
notebook #2)

In order to classify movies and images you can set up a Zooniverse project. The tutorial on the right hand side will guide you through the process.

Upload clips to Zooniverse
(
notebook #3)

If you want to analyse footage, use the below link to run Jupyter notebook #3 "Upload clips to Zooniverse" and transfer your video clips from Cloudina to Zooniverse. On the right hand side you can find a quick guide.

https://github.com/ocean-data-factory-sweden/kso

Upload frames to Zooniverse
(
notebook #4)

If you want to analyse images, use the below link to run Jupyter notebook #4 "Upload frames to Zooniverse" and transfer your image frames from Cloudina to Zooniverse. On the right hand side you can find a quick guide.

https://github.com/ocean-data-factory-sweden/kso

Analyse Zooniverse classifications & publish to GBIF 
(notebook #8)

If you want to analyse the classifications made by yourself and/or citizen scientists, then use the below link to run Jupyter notebook #8 "Analyse Zooniverse classifications". You can use this notebook to explore and pull up-to-date classifications from Zooniverse and format these for model training. The notebook also includes the option to export observations to the Global Biodiversity Information Facility (GBIF). On the right hand side you can find a quick guide.

https://github.com/ocean-data-factory-sweden/kso

Train ML-models 
(notebook #5)

To prepare the training and test data, set model parameters and train models use the below link to run Jupyter notebook #5 "Train ML models". On the right hand side you can find a quick guide.

https://github.com/ocean-data-factory-sweden/kso

Evaluate ML-models 
(notebook #6)

To use ecologically relevant metrics to test the performance of the trained model, use the below link to run Jupyter notebook #6 "Evaluate ML models". On the right hand side you can find a quick guide.

https://github.com/ocean-data-factory-sweden/kso

Run ML-models on footage/images & publish observations to GBIF 
(notebook #9)

To automatically classify new footage using and trained ML-modeluse the below link to run Jupyter notebook #9 "Run ML models on footage". This notebook also includes the option to export the ML-based observations to the Global Biodiversity Information Facility (GBIF). On the right hand side you can find a quick guide.

https://github.com/ocean-data-factory-sweden/kso

Publish ML-models 
(notebook #7)

To publish a trained model to a public repository, use the below link to run Jupyter notebook #7 "Publish ML models". On the right hand side you can find a quick guide.

https://github.com/ocean-data-factory-sweden/kso