Skip to content
Extraits de code Groupes Projets
Valider 8f80366f rédigé par Adrien Payen's avatar Adrien Payen
Parcourir les fichiers

update readme

parent 4ceb1044
Aucune branche associée trouvée
Aucune étiquette associée trouvée
Aucune requête de fusion associée trouvée
......@@ -32,7 +32,7 @@ This **validation.py** file defines a **class validation** in which we create di
### 4. plot.py
This code allows different graphs to compare the results obtained from the strategies. Four graphs can be printed via 4 different functions depending on a layout: **plot_strategy_comparison**, **plot_state_based_turns**, **plot_state_based_comparison**, **plot_state_based_comparison_once**. The layout used to compare our strategies is this: layout = [0, 0, 3, 0, 0, 0, 0, 2, 0, 0, 0, 3, 0, 0, 1, 0]. The **plot_strategy_comparison** function allows comparing different strategies based on their average turns on a histogram. The **plot_state_based_turns** function returns a graph to compare the average turn for each state. The **plot_state_based_comparison** function was implemented in order to compare the Value Iteration algorithm (theorical cost) and different simulations (empirical cost: 100, 10.000 and 1000.000) to show the accuracy obtained as a function of the number of simulations. Finally, the **plot_state_based_comparison_once** function compares the theorical cost and the empirical cost based on a number of defined simulations.
This code allows different graphs to compare the results obtained from the strategies. Four graphs can be printed via 4 different functions depending on a layout: **plot_strategy_comparison**, **plot_state_based_turns**, **plot_state_based_comparison**, **plot_state_based_comparison_once**. The trapped layout used to compare our strategies is this: layout = [0, 0, 3, 0, 0, 0, 0, 2, 0, 0, 0, 3, 0, 0, 1, 0]. The **plot_strategy_comparison** function allows comparing different strategies based on their average turns on a histogram. The **plot_state_based_turns** function returns a graph to compare the average turn for each state. The **plot_state_based_comparison** function was implemented in order to compare the Value Iteration algorithm (theorical cost) and different simulations (empirical cost: 100, 10.000 and 1000.000) to show the accuracy obtained as a function of the number of simulations. Finally, the **plot_state_based_comparison_once** function compares the theorical cost and the empirical cost based on a number of defined simulations.
## Contact
......
0% Chargement en cours ou .
You are about to add 0 people to the discussion. Proceed with caution.
Terminez d'abord l'édition de ce message.
Veuillez vous inscrire ou vous pour commenter