Frontend to download candidate data as CSV
What does this MR do and why?
Candidate data from an experiment (parameters, metrics and metadata) are often used by Data Scientists to create reports on their models. Allowing them to download this data as csv makes it easier for them to use it with their own tools.
Screenshots or screen recordings
How to set up and validate locally
How to set up and validate locally
-
Enable the feature flag
echo "Feature.enable(:ml_experiment_tracking)" | bundle exec rails c
-
Create an Experiment and 10 candidates
# Added variables user_id = 1 project_id = 1 exp = Ml::Experiment.create!(name: "Awesome Gitlab Experiment", user_id: user_id, project_id: project_id) 10.times.each { |i| exp.candidates.create!(user_id: user_id, start_time: 0, name: "candidate_#{i}", internal_id: i, project_id: project_id) } exp.candidates.each_with_index { |c, i| c.metrics.create!(name: "auc", value: i*0.1 , tracked_at: Time.zone.now, step: 1)} exp.candidates.each_with_index { |c, i| c.metrics.create!(name: "accuracy", value: i* 0.1+0.1 , tracked_at: Time.zone.now, step: 1)} exp.candidates.each { |c| c.params.create!(name: "algorithm", value: ["LogisticRegression", "DecisionTree"].sample )}
-
Navigate to
<your project>/-/ml/experiments/1
and click on "Download as CSV" -
A CSV file will be downloaded containing the candidate data for the experiment.
MR acceptance checklist
This checklist encourages us to confirm any changes have been analyzed to reduce risks in quality, performance, reliability, security, and maintainability.
-
I have evaluated the MR acceptance checklist for this MR.