diff options
| -rw-r--r-- | README.md | 27 | ||||
| -rw-r--r-- | simulations/README.md | 12 |
2 files changed, 18 insertions, 21 deletions
@@ -1,20 +1,11 @@ # loosely-policies-analytics -## Analysis folder -- offline.R: contains two major functions: - - build_models: To generate K-fold cross-validation results (note that hyper-parameters for decisions tree is fixed (no validation set)) - - generate_inputs: generate the inputs for the simulations experiments + the decision tree plots -- in-situ.R: Implement the in-situ learning approach (Figure 4a 4b and 4c) - - For figure 4a and 4b we train the model with increasing amount of data from previous results as if we were using one policy per day (see section IV.A) - - For figure 4c, delta is generated by comparing using each policies in round-robin (one per days to perform the training) - to each previous paper results with single policy only (see paper section IV.A) - -Todo: remove minbucket=1 (does not impact the results) - -## Simulation folder - -- src/: contains the simulator code (based on SimGrid) -- libs/: contains a setup script that will fetch and configure the correct SimGrid version - - see simulation/README.md for more info -- results/: Contains all needed script to run the experiments - - In particular paper.sh generates the results present in the paper +## 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 interresting. diff --git a/simulations/README.md b/simulations/README.md index 32ad38c..1f7baad 100644 --- a/simulations/README.md +++ b/simulations/README.md @@ -1,18 +1,24 @@ # Impact of loosely coupled data dissemination policies for resource challenged environements -### Setup +## Folders +- src/: contains the simulator code (based on SimGrid) +- libs/: contains a setup script that will fetch and configure the correct SimGrid version + - see simulation/README.md for more info +- results/: Contains all needed script to run the experiments + - In particular paper.sh generates the results present in the paper +## Setup - First you need [Boost](https://www.boost.org/) - Then setup SimGrid with `cd ./libs && ./setup.sh && cd -` - Then run `make` in this folder to compile the simulator code - If the project compiles successfully, you are ready to execute the simulations! -### Paper's Simulations +## Paper's Simulations - To run all the simulations presented in the paper simply run `./results/paper.sh` this will create the file `results/results.csv` containing all the results. This will also create an `inputs.json` that you can modify at your convenience - Note: `inputs.json` file is generated by `src/scenarios.cc` (see paper.sh to see how it is used) - The idea of scenarios.cc is to generate a random node schedule (turn_on and turn_off cycles) based on the various suppplied arguments - Then, `simulator.cc` can use this `inputs.json` (via `Inputs.cc`) to simulate those cycles -### Simulations +## Simulations - The file `inputs.json` controls the inputs to the simulator. You should only care about this file if you want custom simulations - To run the simulation just run `make run` This will also generate a file called `platform.xml` corresponding to the simulated platform - To generate a *csv* output from the simulation results you can use: |