loic/loosely-policies-analytics
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
Reproducibility: "(POSTER) Towards Data Dissemination Policy Prediction for Constrained Environments Using Analytics" (this repo is not cited in any papers)
60 commits 1 branch 0 tags 61 MiB
Find a file
Exact
Loïc Guégan 5b29c36441 Minor changes
2025-09-24 12:15:21 +02:00
analysis Minor changes 2025-09-24 12:15:21 +02:00
simulations Add explanation comments 2025-09-23 13:34:01 +02:00
slides_recap Update 2022-12-09 13:27:27 +01:00
README.md Correct Readme 2025-09-23 14:00:45 +02:00

README.md

loosely-policies-analytics

Folders

  • analysis/
    • Contains data analysis + offline and in-situ training code.
    • More info in analysis/README.md.
  • simulation/
    • Contains the simulator source code and script to run the experiments from the paper.
  • slides_recap/
    • Just backup slides. Nothing interesting.

Experiments

In-situ

Simulations are carried using previous CCGrid2022 data. These simulations are done in the file analysis/in-situ.R. The idea is to use a single policy each day (round-robin) until the model performance reaches 80%. For that, a day corresponds to a single simulation result from CCGrid2022 with a particular policy. Since we have 200 runs per policy from CCGrid2022, a single run result can be cherry-pick to simulate one day.

Offline

First, models are trained and predictions are generated. This is done in analysis/offline.R. Predictions are stored in analysis/inputs/input_*.csv. Then, simulations are executed with those policy predictions as inputs. This is done in simulations/results/paper.sh.