plots module

Auxiliary module with utility functions for plots related MCTS-NC project and its experiments.

Link to project repository

https://github.com/pklesk/mcts_numba_cuda

plots.scores_array_plot(data, details, label_x, label_y, ticks_x, ticks_y, title)

Displays an array-like plot - a color map with averages of: scores, steps and depths - based on data from several experiments.

plots.scores_array_plot_generator(experiments_hs_array, label_x, label_y, ticks_x, ticks_y, title)

Reads data from several experiments computes averages of scores, steps and depths and generates an array-like plot by calling scores_array_plot function.

plots.scores_array_plot_ocp_thrifty_vs_vanilla_c4()

Generates an array-like plot - a color map with averages of: scores, steps and depths - based on data from experiments: ocp_thrifty vs vanilla (Connect 4).

plots.scores_array_plot_ocp_prodigal_vs_vanilla_c4()

Generates an array-like plot - a color map with averages of: scores, steps and depths - based on data from experiments: ocp_prodigal vs vanilla (Connect 4).

plots.scores_array_plot_acp_thrifty_vs_vanilla_c4()

Generates an array-like plot - a color map with averages of: scores, steps and depths - based on data from experiments: acp_thrifty vs vanilla (Connect 4).

plots.scores_array_plot_acp_prodigal_vs_vanilla()

Generates an array-like plot - a color map with averages of: scores, steps and depths - based on data from experiments: acp_prodigal vs vanilla (Connect 4).

plots.best_action_plot(moves_rounds_black, qs_black, ucbs_black, moves_rounds_white, qs_white, ucbs_white, label_qs_black, label_ucbs_black, label_qs_white, label_ucbs_white, label_x, label_y, title_1, title_2, ucbs_factor=1.0, ucbs_black_color=None, ucbs_white_color=None)

Displays plot of estimates on best actions’ values (and their UCBs) along a given game.

plots.best_action_plot_generator(experiments_hs, game_index, label_qs_black, label_ucbs_black, label_qs_white, label_ucbs_white, label_x, label_y, title_1, title_2, ucbs_factor=1.0, ucbs_black_color=None, ucbs_white_color=None)

Reads data from an experiment and generates a ‘best action plot’ by calling best_action_plot function.

plots.depths_plot(moves_rounds_black, mean_depths_black, max_depths_black, moves_rounds_white, mean_depths_white, max_depths_white, label_mean_depths_black, label_max_depths_black, label_mean_depths_white, label_max_depths_white, label_x, label_y, title_1, title_2)

Displays plot of averages of reached depths (mean and maximum) along a given game.

plots.depths_plot_generator(experiments_hs, game_index, label_mean_depths_black, label_max_depths_black, label_mean_depths_white, label_max_depths_white, label_x, label_y, title_1, title_2)

Reads data from an experiment and generates a ‘depths plot’ by calling best_action_plot function.

plots.averages_printout_generator(experiments_hs_array, ai_instance_name)

Prints out averages of: playouts / steps and mean / maximum depths for a given series of experiments.

plots.averages_printout_5s_vanilla_c4()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 5s vanilla (Connect 4).

plots.averages_printout_5s_ocp_thrifty_c4()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 5s ocp_thrifty (Connect 4).

plots.averages_printout_5s_ocp_prodigal_c4()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 5s ocp_prodigal (Connect 4).

plots.averages_printout_5s_acp_thrifty_c4()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 5s acp_thrifty (Connect 4).

plots.averages_printout_5s_acp_prodigal_c4()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 5s acp_prodigal (Connect 4).

plots.averages_printout_30s_vanilla_gomoku()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 30s vanilla (Gomoku).

plots.averages_printout_30s_ocp_thrifty_gomoku()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 30s ocp_thrifty (Gomoku).

plots.averages_printout_30s_ocp_prodigal_gomoku()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 30s ocp_prodigal (Gomoku).

plots.averages_printout_30s_acp_thrifty_gomoku()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 30s acp_thrifty (Gomoku).

plots.averages_printout_30s_acp_prodigal_gomoku()

Prints out averages of: playouts / steps and mean / maximum depths for experiments involving: 30s acp_prodigal (Gomoku).