Roadmap

[tpham93]

• Creation of AL tasks
• Local experiments
• Predictions ARFF file
• Serialization
• Version information of skactiveml
• Finalization of AL task
• Handling of extensions (explicit passing of the extension to use)
• Testing of batch_size
• Testing of annotation_cost
• Minimal demo
• Upload of flows (collision between sklearn and skactiveml extension)
• Upload of runs
  • Upload of predictions file
  • Upload of utility file
    • header: repeat, fold, cycle, batch_idx, train_idx, utility, was_queried
• Evaluation based on annotation costs
• Frontend Task, Flow, Run
• Frontend evaluation visualization