Added resultsExplorer.py

src/AI/resultsExplorer/resultsExplorer.py

@@ -0,0 +1,54 @@
+"""
+A small script to help you analyze the output logs from GA algorithm.
+It shows you best and worst population with all values.
+Specify the input file as an exec parameter.
+"""
+import sys
+from ast import literal_eval
+import matplotlib.pyplot as plt
+import numpy
+
+
+def fitnessPlot(pop):
+    fig, ax = plt.subplots()
+    ax.plot([i[4] for i in pop])
+    ax.set_title("Best fitness by generation")
+    ax.set_ylabel("Fitness")
+    ax.set_xlabel("Generation")
+    fig.legend()
+    fig.show()
+    fig.savefig("fitnessPlot")
+    print("Figure saved to file.")
+
+
+def printBestWorstGenes(pop):
+    fitnesses = [i[4] for i in pop]
+    bestIdxs = (numpy.where(fitnesses == numpy.max(fitnesses)))[0]
+
+    print("Best Fitness:", max(fitnesses))
+    for idx in bestIdxs:
+        print("Affinities: food={}, water={}, rest={}, walking={}".format(pop[idx][0], pop[idx][1],
+                                                                          pop[idx][2], pop[idx][3]))
+
+    worstIdxs = (numpy.where(fitnesses == numpy.min(fitnesses)))[0]
+
+    print("Worst Fitness:", min(fitnesses))
+    for idx in worstIdxs:
+        print("Affinities: food={}, water={}, rest={}, walking={}".format(pop[idx][0], pop[idx][1],
+                                                                          pop[idx][2], pop[idx][3]))
+
+
+if len(sys.argv) < 2:
+    print("No file specified.")
+else:
+    with open(sys.argv[1], "r") as file:
+        if file.readline() != "=HEADER=GA=\n":
+            print("Invalid header. Incorrect file or missing header?")
+        else:
+            # Load data into memory
+            pop = []
+            for line in file:
+                pop.append(literal_eval(line))
+
+            fitnessPlot(pop)
+            printBestWorstGenes(pop)