piegy 1.1.2__py3-none-any.whl → 1.1.4__py3-none-any.whl

piegy/figures.py

@@ -31,7 +31,7 @@ Popu-payoff correlation:
 
 
 from .tools import figure_tools as figure_t
-from . import model as model
+from . import simulation
 
 import matplotlib.pyplot as plt
 import numpy as np
@@ -43,18 +43,18 @@ CURVE_TYPE = '-'
 
 
 
-def UV_heatmap(sim, ax_U = None, ax_V = None, U_color = 'Purples', V_color = 'Greens', start = 0.95, end = 1.0, annot = False, fmt = '.3g'):
+def UV_heatmap(mod, ax_U = None, ax_V = None, U_color = 'Purples', V_color = 'Greens', start = 0.95, end = 1.0, annot = False, fmt = '.3g'):
     '''
     Makes two heatmaps for U, V average distribution over a time interval, respectively. Works best for 2D space.
     1D works as well, but figures look bad.
 
     Inputs:
-        sim:        A model.simulation object.
+        mod:        A simulation.model object.
         ax_U, ax_V: matplotlib axes to plot on. New axes will be created if None is given.
         U_color:    Color for U's heatmap, uses matplotlib color maps.
         V_color:    Color for V's heatmap.
         start:      (0,1) float, where the interval should start from. Intended as a 'percentage'. 
-                    For example, start = 0.8 means the interval should start from the 80% point of sim.maxtime.
+                    For example, start = 0.8 means the interval should start from the 80% point of mod.maxtime.
         end:        (0,1) float, where the interval ends.
         annot:      bool, whether to add annotations (show exact numbers for every patch).
         fmt:        Number format for annotations. How many digits you want to keep. Please set annot = True first and then use fmt.
@@ -63,12 +63,12 @@ def UV_heatmap(sim, ax_U = None, ax_V = None, U_color = 'Purples', V_color = 'Gr
         ax_U, ax_V: matplotlib axes with heatmaps of U, V distribution plotted upon.
     '''
     
-    start_index = int(start * sim.max_record)
-    end_index = int(end * sim.max_record)
+    start_index = int(start * mod.max_record)
+    end_index = int(end * mod.max_record)
     
     # see ave_interval below
-    U_ave = figure_t.ave_interval(sim.U, start_index, end_index)
-    V_ave = figure_t.ave_interval(sim.V, start_index, end_index)
+    U_ave = figure_t.ave_interval(mod.U, start_index, end_index)
+    V_ave = figure_t.ave_interval(mod.V, start_index, end_index)
     
     #### plot ####
     
@@ -84,29 +84,29 @@ def UV_heatmap(sim, ax_U = None, ax_V = None, U_color = 'Purples', V_color = 'Gr
     
 
 
-def UV_bar(sim, ax_U = None, ax_V = None, U_color = 'purple', V_color = 'green', start = 0.95, end = 1.0):
+def UV_bar(mod, ax_U = None, ax_V = None, U_color = 'purple', V_color = 'green', start = 0.95, end = 1.0):
     '''
     Makes two barplots for U, V average distribution over a time interval. Works best for 1D space.
     2D works as well, but figures look bad.
 
     Inputs:
-        sim:        A model.simulation object.
+        mod:        A simulation.model object.
         ax_U, ax_V: matplotlib axes to plot on. New axes will be created if None is given.
         U_color:    Color of U's barplot. Uses Matplotlib colors.
                     See available colors at: https://matplotlib.org/stable/gallery/color/named_colors.html
         V_color:    Color of V's barplot. Uses Matplotlib colors.
-        start:      (0,1) float. How much proportion of sim.maxtime you want the interval to start from.
+        start:      (0,1) float. How much proportion of mod.maxtime you want the interval to start from.
         end:        (0,1) float. Where you want the interval to end.
 
     Returns:
         ax_U, ax_V: matplotlib axes with bar plots for U and V plotted upon.
     '''
     
-    start_index = int(start * sim.max_record)
-    end_index = int(end * sim.max_record)
+    start_index = int(start * mod.max_record)
+    end_index = int(end * mod.max_record)
     
-    U_ave = figure_t.ave_interval_1D(sim.U, start_index, end_index)
-    V_ave = figure_t.ave_interval_1D(sim.V, start_index, end_index)
+    U_ave = figure_t.ave_interval_1D(mod.U, start_index, end_index)
+    V_ave = figure_t.ave_interval_1D(mod.V, start_index, end_index)
 
     #### plot ####
 
@@ -123,7 +123,7 @@ def UV_bar(sim, ax_U = None, ax_V = None, U_color = 'purple', V_color = 'green',
 
 
 
-def UV_dyna(sim, ax = None, interval = 20, grid = True):
+def UV_dyna(mod, ax = None, interval = 20, grid = True):
     '''
     Plots how total U, V change overtime.
     The curves are not directly based on every single data point. 
@@ -132,10 +132,10 @@ def UV_dyna(sim, ax = None, interval = 20, grid = True):
         So if there are 2000 data points in total, then there will be 2000 / 20 = 100 points on the curves.
 
     Inputs:
-        sim:        A model.simulation object.
+        mod:        A simulation.model object.
         ax:         matplotlib ax to plot on. New ax will be created if None is given.
         interval:   How many data points to take average over. Larger value makes curves smoother, but also loses local fluctuations.
-                    NOTE: this interval doesn't overlap with sim.compress_itv. 
+                    NOTE: this interval doesn't overlap with mod.compress_itv. 
                     e.g. you already took average over every 20 data points, then using interval <= 20 here has no smoothing effect.
         grid:       Whether to add grid lines to plot.
     
@@ -148,19 +148,19 @@ def UV_dyna(sim, ax = None, interval = 20, grid = True):
     V_curve = []
     total_curve = []
 
-    interval = figure_t.scale_interval(interval, sim.compress_itv)
-    interval_num = int(sim.max_record / interval)
+    interval = figure_t.scale_interval(interval, mod.compress_itv)
+    interval_num = int(mod.max_record / interval)
     
     for i in range(interval_num):
-        U_ave = figure_t.ave_interval(sim.U, i * interval, (i + 1) * interval)
-        V_ave = figure_t.ave_interval(sim.V, i * interval, (i + 1) * interval)
+        U_ave = figure_t.ave_interval(mod.U, i * interval, (i + 1) * interval)
+        V_ave = figure_t.ave_interval(mod.V, i * interval, (i + 1) * interval)
         
         U_curve.append(np.sum(U_ave))
         V_curve.append(np.sum(V_ave))
         total_curve.append(U_curve[-1] + V_curve[-1])
         
     #### plot ####   
-    xaxis = np.linspace(0, sim.maxtime, len(U_curve))
+    xaxis = np.linspace(0, mod.maxtime, len(U_curve))
 
     if ax == None:
         _, ax = plt.subplots()
@@ -178,7 +178,7 @@ def UV_dyna(sim, ax = None, interval = 20, grid = True):
 
 
 
-def UV_hist(sim, ax_U = None, ax_V = None, U_color = 'purple', V_color = 'green', start = 0.95, end = 1.0):
+def UV_hist(mod, ax_U = None, ax_V = None, U_color = 'purple', V_color = 'green', start = 0.95, end = 1.0):
     '''
     Makes density histograms for U, V's average distribution over an interval.
     Sometimes it may not be shown in density plots due to matplotlib features.
@@ -187,11 +187,11 @@ def UV_hist(sim, ax_U = None, ax_V = None, U_color = 'purple', V_color = 'green'
         ax_U, ax_V: matplotlib axes with heatmaps of U, V population density plotted upon.
     '''
 
-    start_index = int(start * sim.max_record)
-    end_index = int(end * sim.max_record)
+    start_index = int(start * mod.max_record)
+    end_index = int(end * mod.max_record)
     
-    U_ave = figure_t.ave_interval_1D(sim.U, start_index, end_index)
-    V_ave = figure_t.ave_interval_1D(sim.V, start_index, end_index)
+    U_ave = figure_t.ave_interval_1D(mod.U, start_index, end_index)
+    V_ave = figure_t.ave_interval_1D(mod.V, start_index, end_index)
     
     #### plot ####
     
@@ -214,7 +214,7 @@ def UV_hist(sim, ax_U = None, ax_V = None, U_color = 'purple', V_color = 'green'
 
 
 
-def UV_std(sim, ax = None, interval = 20, grid = True):
+def UV_std(mod, ax = None, interval = 20, grid = True):
     '''
     Plots how standard deviation of U, V change over time.
     Takes average over many small interval to smooth out local fluctuations.
@@ -223,21 +223,21 @@ def UV_std(sim, ax = None, interval = 20, grid = True):
         ax:    matplotlib ax, contains U's and V's std curves.
     '''
 
-    interval = figure_t.scale_interval(interval, sim.compress_itv)
-    interval_num = int(sim.max_record / interval)
+    interval = figure_t.scale_interval(interval, mod.compress_itv)
+    interval_num = int(mod.max_record / interval)
     
     U_std = []
     V_std = []
     
     for i in range(interval_num):
-        U_ave = figure_t.ave_interval(sim.U, i * interval, (i + 1) * interval)
-        V_ave = figure_t.ave_interval(sim.V, i * interval, (i + 1) * interval)
+        U_ave = figure_t.ave_interval(mod.U, i * interval, (i + 1) * interval)
+        V_ave = figure_t.ave_interval(mod.V, i * interval, (i + 1) * interval)
         
         U_std.append(np.std(U_ave))
         V_std.append(np.std(V_ave))
     
     #### plot ####
-    xaxis = np.linspace(0, sim.maxtime, len(U_std))
+    xaxis = np.linspace(0, mod.maxtime, len(U_std))
 
     if ax == None:
         _, ax = plt.subplots()
@@ -253,33 +253,33 @@ def UV_std(sim, ax = None, interval = 20, grid = True):
 
 
 
-def UV_expected_val(sim):
+def UV_expected_val(mod):
     '''
     Calculate expected U & V distribution based on matrices, assume no migration.
     To differentiate from UV_expected in figures.py: this one return arrays (values).
     '''
     
-    U_expected = np.zeros((sim.N, sim.M))
-    V_expected = np.zeros((sim.N, sim.M))
+    U_expected = np.zeros((mod.N, mod.M))
+    V_expected = np.zeros((mod.N, mod.M))
     
-    for i in range(sim.N):
-        for j in range(sim.M):
+    for i in range(mod.N):
+        for j in range(mod.M):
             # say matrix = [a, b, c, d]
             # U_proportion = (d - b) / (a - b - c + d)
-            U_prop = (sim.X[i][j][3] - sim.X[i][j][1]) / (sim.X[i][j][0] - sim.X[i][j][1] - sim.X[i][j][2] + sim.X[i][j][3])
+            U_prop = (mod.X[i][j][3] - mod.X[i][j][1]) / (mod.X[i][j][0] - mod.X[i][j][1] - mod.X[i][j][2] + mod.X[i][j][3])
             # equilibrium payoff, U_payoff = V_payoff
-            eq_payoff = U_prop * sim.X[i][j][0] + (1 - U_prop) * sim.X[i][j][1]
+            eq_payoff = U_prop * mod.X[i][j][0] + (1 - U_prop) * mod.X[i][j][1]
             
             # payoff / kappa * proportion
-            U_expected[i][j] = eq_payoff / sim.P[i][j][4] * U_prop
-            V_expected[i][j] = eq_payoff / sim.P[i][j][5] * (1 - U_prop)
+            U_expected[i][j] = eq_payoff / mod.P[i][j][4] * U_prop
+            V_expected[i][j] = eq_payoff / mod.P[i][j][5] * (1 - U_prop)
                 
     return U_expected, V_expected
 
 
 
 
-def UV_expected(sim, ax_U = None, ax_V = None, U_color = 'Purples', V_color = 'Greens', annot = False, fmt = '.3g'):
+def UV_expected(mod, ax_U = None, ax_V = None, U_color = 'Purples', V_color = 'Greens', annot = False, fmt = '.3g'):
     '''
     Calculate expected population distribution based on matrices, assuming no migration.
     For the formulas, see stochastic_mode.expected_UV
@@ -292,14 +292,14 @@ def UV_expected(sim, ax_U = None, ax_V = None, U_color = 'Purples', V_color = 'G
                 If 1D (N or M == 1), then ax_U and ax_V are barplots.
     '''
     
-    U_expected, V_expected = UV_expected_val(sim)
+    U_expected, V_expected = UV_expected_val(mod)
     
     U_text = figure_t.gen_text(np.mean(U_expected), np.std(U_expected))
     V_text = figure_t.gen_text(np.mean(V_expected), np.std(V_expected))
     
     #### plot ####
     
-    if (sim.N != 1) and (sim.M != 1):
+    if (mod.N != 1) and (mod.M != 1):
         # 2D
         ax_U = figure_t.heatmap(U_expected, ax_U, U_color, annot, fmt, title = 'Expected U', text = U_text)
         ax_V = figure_t.heatmap(V_expected, ax_V, V_color, annot, fmt, title = 'Expected V', text = V_text)
@@ -314,7 +314,7 @@ def UV_expected(sim, ax_U = None, ax_V = None, U_color = 'Purples', V_color = 'G
 
 
 
-def pi_heatmap(sim, ax_U = None, ax_V = None, U_color = 'BuPu', V_color = 'YlGn', start = 0.95, end = 1.0, annot = False, fmt = '.3g'):
+def pi_heatmap(mod, ax_U = None, ax_V = None, U_color = 'BuPu', V_color = 'YlGn', start = 0.95, end = 1.0, annot = False, fmt = '.3g'):
     '''
     Make heatmaps for payoff in a specified interval.
     Works best for 2D. 1D works as well, but figures look bad.
@@ -326,11 +326,11 @@ def pi_heatmap(sim, ax_U = None, ax_V = None, U_color = 'BuPu', V_color = 'YlGn'
         ax_U, ax_V: Seaborn heatmaps, for U's & V's payoff distribution, respectively.
     '''
     
-    start_index = int(sim.max_record * start)
-    end_index = int(sim.max_record * end)
+    start_index = int(mod.max_record * start)
+    end_index = int(mod.max_record * end)
     
-    U_pi_ave = figure_t.ave_interval(sim.U_pi, start_index, end_index)
-    V_pi_ave = figure_t.ave_interval(sim.V_pi, start_index, end_index)
+    U_pi_ave = figure_t.ave_interval(mod.U_pi, start_index, end_index)
+    V_pi_ave = figure_t.ave_interval(mod.V_pi, start_index, end_index)
     
     U_title = figure_t.gen_title('Payoff ' + r'$p_U$', start, end)
     U_text = figure_t.gen_text(np.mean(U_pi_ave), np.std(U_pi_ave))
@@ -345,7 +345,7 @@ def pi_heatmap(sim, ax_U = None, ax_V = None, U_color = 'BuPu', V_color = 'YlGn'
 
 
 
-def pi_bar(sim, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowgreen', start = 0.95, end = 1.0):
+def pi_bar(mod, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowgreen', start = 0.95, end = 1.0):
     '''
     Make barplot for payoff in a specified interval.
     Works best for 1D. 2D works as well, but figures look bad.
@@ -354,11 +354,11 @@ def pi_bar(sim, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowg
         ax_U, ax_V: matplotlib axes with barplots of U and V payoff distribution plotted upon.
     '''
     
-    start_index = int(sim.max_record * start)
-    end_index = int(sim.max_record * end)
+    start_index = int(mod.max_record * start)
+    end_index = int(mod.max_record * end)
     
-    U_pi_ave = figure_t.ave_interval_1D(sim.U_pi, start_index, end_index)
-    V_pi_ave = figure_t.ave_interval_1D(sim.V_pi, start_index, end_index)
+    U_pi_ave = figure_t.ave_interval_1D(mod.U_pi, start_index, end_index)
+    V_pi_ave = figure_t.ave_interval_1D(mod.V_pi, start_index, end_index)
     
     U_title = figure_t.gen_title(r'$p_U$', start, end)
     U_text = figure_t.gen_text(np.mean(U_pi_ave), np.std(U_pi_ave))
@@ -373,7 +373,7 @@ def pi_bar(sim, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowg
 
 
 
-def pi_dyna(sim, ax = None, interval = 20, grid = True):
+def pi_dyna(mod, ax = None, interval = 20, grid = True):
     '''
     Plot how payoffs change over time.
 
@@ -385,19 +385,19 @@ def pi_dyna(sim, ax = None, interval = 20, grid = True):
     V_curve = []
     total_curve = []
 
-    interval = figure_t.scale_interval(interval, sim.compress_itv)
-    interval_num = int(sim.max_record / interval)
+    interval = figure_t.scale_interval(interval, mod.compress_itv)
+    interval_num = int(mod.max_record / interval)
     
     for i in range(interval_num):
-        U_ave = figure_t.ave_interval(sim.U_pi, i * interval, (i + 1) * interval)
-        V_ave = figure_t.ave_interval(sim.V_pi, i * interval, (i + 1) * interval)
+        U_ave = figure_t.ave_interval(mod.U_pi, i * interval, (i + 1) * interval)
+        V_ave = figure_t.ave_interval(mod.V_pi, i * interval, (i + 1) * interval)
     
         U_curve.append(np.sum(U_ave))
         V_curve.append(np.sum(V_ave))
         total_curve.append(U_curve[-1] + V_curve[-1])
         
     #### plot ####    
-    xaxis = np.linspace(0, sim.maxtime, len(U_curve))
+    xaxis = np.linspace(0, mod.maxtime, len(U_curve))
     
     if ax == None:
         _, ax = plt.subplots()
@@ -415,7 +415,7 @@ def pi_dyna(sim, ax = None, interval = 20, grid = True):
 
 
 
-def pi_hist(sim, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowgreen', start = 0.95, end = 1.0):
+def pi_hist(mod, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowgreen', start = 0.95, end = 1.0):
     '''
     Makes deensity histograms of U's and V's payoffs in a sepcified interval.
     Sometimes it may not be shown in density plots due to matplotlib features.
@@ -424,11 +424,11 @@ def pi_hist(sim, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellow
         ax_U, ax_V:     histogram of U's and V's payoff.
     '''
 
-    start_index = int(start * sim.max_record)
-    end_index = int(end * sim.max_record)
+    start_index = int(start * mod.max_record)
+    end_index = int(end * mod.max_record)
 
-    U_pi_ave = figure_t.ave_interval_1D(sim.U_pi, start_index, end_index)
-    V_pi_ave = figure_t.ave_interval_1D(sim.V_pi, start_index, end_index)
+    U_pi_ave = figure_t.ave_interval_1D(mod.U_pi, start_index, end_index)
+    V_pi_ave = figure_t.ave_interval_1D(mod.V_pi, start_index, end_index)
     
     #### plot ####
     
@@ -451,7 +451,7 @@ def pi_hist(sim, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellow
 
 
 
-def pi_std(sim, ax = None, interval = 20, grid = True):
+def pi_std(mod, ax = None, interval = 20, grid = True):
     '''
     Plots how standard deviation of payoff change over time.
 
@@ -460,21 +460,21 @@ def pi_std(sim, ax = None, interval = 20, grid = True):
     '''
     
     
-    interval = figure_t.scale_interval(interval, sim.compress_itv)
-    interval_num = int(sim.max_record / interval)
+    interval = figure_t.scale_interval(interval, mod.compress_itv)
+    interval_num = int(mod.max_record / interval)
     
     U_pi_std = []
     V_pi_std = []
     
     for i in range(interval_num):
-        U_pi_ave = figure_t.ave_interval(sim.U_pi, i * interval, (i + 1) * interval)
-        V_pi_ave = figure_t.ave_interval(sim.V_pi, i * interval, (i + 1) * interval)
+        U_pi_ave = figure_t.ave_interval(mod.U_pi, i * interval, (i + 1) * interval)
+        V_pi_ave = figure_t.ave_interval(mod.V_pi, i * interval, (i + 1) * interval)
         
         U_pi_std.append(np.std(U_pi_ave))
         V_pi_std.append(np.std(V_pi_ave))
     
     #### plot ####
-    xaxis = np.linspace(0, sim.maxtime, len(U_pi_std))
+    xaxis = np.linspace(0, mod.maxtime, len(U_pi_std))
     
     if ax == None:
         _, ax = plt.subplots()
@@ -491,7 +491,7 @@ def pi_std(sim, ax = None, interval = 20, grid = True):
 
 
 
-def UV_pi(sim, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowgreen', alpha = 0.5, start = 0.95, end = 1.0):
+def UV_pi(mod, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowgreen', alpha = 0.5, start = 0.95, end = 1.0):
     '''
     Make two scatter plots: x-axes are population and y-axes are payoff in a specified time interval.
     Reveals relationship between population and payoff.
@@ -500,14 +500,14 @@ def UV_pi(sim, ax_U = None, ax_V = None, U_color = 'violet', V_color = 'yellowgr
         ax_U, ax_V: matplotlib axes with U and V population-payoff scatter plots.
     '''
     
-    start_index = int(start * sim.max_record)
-    end_index = int(end * sim.max_record)
+    start_index = int(start * mod.max_record)
+    end_index = int(end * mod.max_record)
     
-    U_ave = figure_t.ave_interval_1D(sim.U, start_index, end_index)
-    V_ave = figure_t.ave_interval_1D(sim.V, start_index, end_index)
+    U_ave = figure_t.ave_interval_1D(mod.U, start_index, end_index)
+    V_ave = figure_t.ave_interval_1D(mod.V, start_index, end_index)
 
-    U_pi_ave = figure_t.ave_interval(sim.U_pi, start_index, end_index)
-    V_pi_ave = figure_t.ave_interval(sim.V_pi, start_index, end_index)
+    U_pi_ave = figure_t.ave_interval(mod.U_pi, start_index, end_index)
+    V_pi_ave = figure_t.ave_interval(mod.V_pi, start_index, end_index)
     
     
     ax_U = figure_t.scatter(U_ave, U_pi_ave, ax_U, U_color, alpha, xlabel = 'U', ylabel = 'Payoff ' + r'$p_U$', title = 'U - ' + r'$p_U$')