changed signatures to have also a main ingredient set

EvolutionaryAlgorithm/EvolutionaryAlgorithm.py

@@ -72,6 +72,12 @@ m_mix = dill.load(open("../RecipeAnalysis/m_mix.dill", "rb"))
 m_base_act = dill.load(open("../RecipeAnalysis/m_base_act.dill", "rb"))
 m_base_mix = dill.load(open("../RecipeAnalysis/m_base_mix.dill", "rb"))
 
+
+m_grouped_mix = dill.load(open("../RecipeAnalysis/m_grouped_mix_raw.dill", "rb"))
+m_grouped_act = dill.load(open("../RecipeAnalysis/m_grouped_act_raw.dill", "rb"))
+m_grouped_base_act = dill.load(open("../RecipeAnalysis/m_grouped_base_act_raw.dill", "rb"))
+
+
 #m_act.apply_threshold(3)
 #m_mix.apply_threshold(3)
 #m_base_act.apply_threshold(5)
@@ -88,6 +94,10 @@ m_mix.compile()
 m_base_act.compile()
 m_base_mix.compile()
 
+m_grouped_mix.compile()
+m_grouped_act.compile()
+m_grouped_base_act.compile()
+
 c_act = m_act._csr
 c_mix = m_mix._csr
 c_base_act = m_base_act._csr
@@ -455,6 +465,9 @@ class RecipeTreeNode(object):
 
 # ### Mix Node
 
+# For the Node Score: just make a simple lookup whether this combination is seen or not. So the node Score is defined as:
+# 
+
 class MixNode(RecipeTreeNode):
     def __init__(self, constant=False):
         super().__init__("mix", constant, single_child=False)
@@ -536,12 +549,22 @@ class MixNode(RecipeTreeNode):
                 #s_base += sym_score(ing_a._base_ingredient, ing_b._base_ingredient, m_base_mix, c_base_mix)
                 
                 #s += sym_score(ing_a.to_json(), ing_b.to_json(), m_mix, c_mix)
-                p1 = sym_p_a_given_b(ing_a.to_json(), ing_b.to_json(), m_mix, c_mix)
-                p2 = sym_p_a_given_b(ing_b.to_json(), ing_a.to_json(), m_mix, c_mix)
                 
-                s += 0.5 * p1 + 0.5 * p2
+                # old method:
+                #p1 = sym_p_a_given_b(ing_a.to_json(), ing_b.to_json(), m_mix, c_mix)
+                #p2 = sym_p_a_given_b(ing_b.to_json(), ing_a.to_json(), m_mix, c_mix)
+                #s += 0.5 * p1 + 0.5 * p2
                 
-            except:
+                
+                ia = m_mix._label_index[ing_a.to_json()]
+                ib = m_mix._label_index[ing_b.to_json()]
+                
+                if c_mix[ia,ib] > 0 or c_mix[ib,ia] > 0:
+                    s += 1
+                
+                
+                
+            except KeyError as e:
                 pass
         
         #s_base /= len(pairwise_tuples)
@@ -652,7 +675,7 @@ class ActionNode(RecipeTreeNode):
 
 class Tree(object):
     @staticmethod
-    def build_initial_tree(ingredients: list, max_n = 4, wheel_turns = 2):
+    def build_initial_tree(ingredients: list, main_ingredients: list, max_n = 4, wheel_turns = 2):
         
         '''
         # get action sets for ingredients
@@ -833,7 +856,7 @@ class Tree(object):
         return list(constant_ingredients) + list(additional_ingredients)
 
     @staticmethod
-    def from_ingredients(ingredients: list, additional_ings=0):
+    def from_ingredients(ingredients: list, main_ingredients: list, additional_ings=0):
         root = None
         
         constant_ingredients = ingredients
@@ -997,8 +1020,8 @@ class Tree(object):
 # ## Population
 
 class Population(object):
-    def __init__(self, start_ingredients, n_population = 10, max_additional_ings=0):
+    def __init__(self, start_ingredients, main_ingredients, n_population = 10, max_additional_ings=0):
-        self.population = [Tree.from_ingredients(start_ingredients, additional_ings=max_additional_ings) for i in range(n_population)]
+        self.population = [Tree.from_ingredients(start_ingredients, main_ingredients, additional_ings=max_additional_ings) for i in range(n_population)]
         self._n = n_population
         self._mix_min = None
         self._mix_max = None
@@ -1115,3 +1138,19 @@ class Population(object):
             display(t.root().dot())
 
 
+# ## Run Evolutionary Algorithm
+
+#p = Population(["bacon", "tomato", "onion"],['noodle'], max_additional_ings=6)
+
+
+#p_ingredient_unprepared(list(p.population[0].root().childs())[0]._name) < 0.2
+
+
+#p.run(100)
+
+
+#p.plot_population(collect_scores=False)
+
+
+
+

EvolutionaryAlgorithm/InteractiveVersion.ipynb

@@ -123,7 +123,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "5796ec52773740c59e747c0e5f77410e",
+       "model_id": "affefd1263f44e97815b9c32f49d69d9",
        "version_major": 2,
        "version_minor": 0
       },
@@ -149,7 +149,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "92fd11191481475a9c40ae76201b4772",
+       "model_id": "29990add3612462abfb4dcc60f2ff7ca",
        "version_major": 2,
        "version_minor": 0
       },
@@ -163,7 +163,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "618b5a44910843bbaed8b36c3ad2bc46",
+       "model_id": "4b66883d839f49e9b764f683ba9079c9",
        "version_major": 2,
        "version_minor": 0
       },
@@ -189,7 +189,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "301ebb9ed6024493ad85c2b79402345e",
+       "model_id": "645f200c9fa441f8964683dbcd188ad0",
        "version_major": 2,
        "version_minor": 0
       },
@@ -215,7 +215,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "c90d303cd2cb43d1aae401ac6226e3a1",
+       "model_id": "d0c4b6f97b894211a5a1c2cf052e1d58",
        "version_major": 2,
        "version_minor": 0
       },
@@ -229,7 +229,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "cea1f9de60344298ac8417d755ad74df",
+       "model_id": "889ed0ea7cc24663b40beb30e8a1abcc",
        "version_major": 2,
        "version_minor": 0
       },
@@ -243,7 +243,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "3ac8e962dfeb445fa3417dbdbfd5c44c",
+       "model_id": "fa434784f8364bc692301a162824cb30",
        "version_major": 2,
        "version_minor": 0
       },
@@ -321,7 +321,10 @@
     "\n",
     "w_run_button = widgets.Button(description=\"run EA\")\n",
     "\n",
+    "p = None\n",
+    "\n",
     "def run(e=None):\n",
+    "    global p\n",
     "    w_result_out.clear_output()\n",
     "    with w_result_out:\n",
     "        p = EvolutionaryAlgorithm.Population(\n",