Experiment 5.1

Running this experiment is relatively straight-forward. Basically, for every combination of algorithm (signal fitting, signal masking, and PCA) and data set, you need to run the following two steps:

1. Run the respective experiment at

   $HSR4HCI_EXPERIMENTS_DIR/main/5.1_first-results/<algorithm>/<dataset>
   

   Here is an example:

   python $HSR4HCI_SCRIPTS_DIR/experiments/single-script/01_run_pipeline.py \
     --experiment-dir $HSR4HCI_EXPERIMENTS_DIR/main/5.1_first-results/signal_fitting/beta_pictoris__lp
   

   For HSR, you can either use the scripts in single-script or multiple-scripts for this. For PCA, use the script in $HSR4HCI_SCRIPTS_DIR/experiments/run-pca.

2. Once you have run an experiment, you can create the corresponding result plot (containing the signal estimate as a PDF) by using the following command:

   python $HSR4HCI_SCRIPTS_DIR/experiments/evaluate-and-plot/evaluate_and_plot_signal_estimate.py \
     --experiment-dir $HSR4HCI_EXPERIMENTS_DIR/main/5.1_first-results/<algorithm>/<dataset>
   

   The resulting plot will be stored in:

   $HSR4HCI_EXPERIMENTS_DIR/main/5.1_first-results/<algorithm>/<dataset>/plots
   

3. Optionally, to create the kind of plots from Figure 4 in our paper, you can also run:

   python $HSR4HCI_SCRIPTS_DIR/experiments/evaluate-and-plot/plot_results_of_stage_2.py \
     --experiment-dir $HSR4HCI_EXPERIMENTS_DIR/main/5.1_first-results/<algorithm>/<dataset>
   

   The plots will be placed in the same plots directory as the plot of the signal estimate.