Normalization als parameter und Edgelikelihood berechnung in main methode

ILP/butadien/butadiensynthesis.py

@@ -87,70 +87,6 @@ HYPERGRAPH3 = {
     2: (['p_{0,0}'], []),
 }
 
-VERTICES = ['Butadien', 'p_{0,0}', 'p_{0,1}', 'p_{0,2}', 'p_{0,3}', 'p_{0,4}', 'p_{0,5}', 'p_{0,6}', 'p_{0,7}', 'p_{0,8}', 'p_{0,9}', 'p_{0,10}', 'p_{0,11}', 'p_{0,12}', 'p_{0,13}', 'p_{0,14}', 'p_{0,15}', 'p_{0,16}', 'p_{0,17}', 'p_{0,18}']
-#Vergleich mit dem NMR von Ethylen und Hexatrien
-NMR1 = [0.32, 0.5, 0.87, 0.0, 0.11, 0.58, 0.06, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.09, 0.0, 0.0, 0.17, 0.0, 0.09]
-#Normalisiert
-NMR1 = [0.11, 0.18, 0.31, 0.0, 0.04, 0.21, 0.02, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03, 0.0, 0.0, 0.06, 0.0, 0.03]
-#Vergleich mit dem NMR von Ethylen und Octrien
-NMR2 = [0.33, 0.43, 0.6, 0.0, 0.13, 0.9, 0.07, 0.01, 0.0, 0.0, 0.01, 0.0, 0.0, 0.01, 0.1, 0.0, 0.0, 0.22, 0.0, 0.13]
-#Normalisiert
-NMR2 = [0.11, 0.15, 0.21, 0.0, 0.04, 0.31, 0.02, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03, 0.0, 0.0, 0.07, 0.0, 0.04]
-#Vergleich mit dem NMR von Ethylen und Benzol
-NMR3 = [0.0, 0.32, 0.0, 0.0, 0.0, 0.0, 0.05, 0.0, 0.0, 0.02, 0.0, 0.29, 0.95, 0.17, 0.0, 0.0, 0.0, 0.0, 0.0, 0.05]
-#Normalisiert
-NMR3 = [0.0, 0.17, 0.0, 0.0, 0.0, 0.0, 0.03, 0.0, 0.0, 0.01, 0.0, 0.16, 0.51, 0.09, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03]
-VERTICESSMILES = ['C=CC=C', 'C=C', 'C=CC=CC=C', 'C1CCC(C=C)CC=1', 'C(CCC(C=C)CC=C)=C', 'C=CC=CC=CC=C', 'C1CCC(C=CC=C)CC=1', 'C=CC1C=CCCC1', 'C1CCCCC=1', 'C=CC1CC=CCC1C=C', 'C1CCC(C=C)C(C=C)C=1', 'C(C1CC(C=C)C=CC1)=C', 'C1C=CC=CC=1', 'C(CCC1C=CC=CC1)=C', 'C1C(C=CC=C)CCCC=1', 'C=CC(C=C)CCCC=C', 'C=CCCCCC=C', 'C=CC1C=CC(C=C)CC1', 'C1CC2CCCCC2CC=1', 'C1CC2C=CC=CC2CC=1']
-NMRMAX = [max(n1,n2,n3) for n1,n2,n3 in zip(NMR1,NMR2,NMR3)]
-
-#Kombiniert Molekül it Wahrscheinlichkeit für NMR1:
-VERTICE1 = {}
-for vertice, likelihood in zip(VERTICES, NMR1):
-    VERTICE1[vertice] = likelihood
-#Kombiniert Molekül it Wahrscheinlichkeit für NMR2:
-VERTICE2 = {}
-for vertice, likelihood in zip(VERTICES, NMR2):
-    VERTICE2[vertice] = likelihood
-#Kombiniert Molekül it Wahrscheinlichkeit für NMR3:
-VERTICE3 = {}
-for vertice, likelihood in zip(VERTICES, NMR3):
-    VERTICE3[vertice] = likelihood
-
-#Kantenwahrscheinlichkeiten für NMR1:
-EDGE1 = {}
-#Kantenwahrscheinlichkeiten für NMR2:
-EDGE2 = {}
-#Kantenwahrscheinlichkeiten für NMR3:
-EDGE3 = {}
-#Maximum der verschiedenen Kantenwahrscheinlichkeiten:
-EDGEMAX = {}
-
-modes = ["Product", "Average"]
-mode = modes[0] 
-
-#Ist es besser, die Edukte mit zu berücksichtigen? Dann hin rück gleich wahrscheinlich, aber nur 1 kann gewählt werden
-#Ordnet die Kantenwahrscheinlichkeit basierend auf dem Mittel oder Produkt der Wahrscheinlichkeiten der Produkte
-for edge, (tails, heads) in HYPERGRAPH.items():
-    if heads == [] or tails == []:
-        EDGE1[edge] = 0.0
-        EDGE2[edge] = 0.0
-        EDGE3[edge] = 0.0
-    elif mode == "Product":
-        #print([VERTICE1[head] for head in heads])
-        EDGE1[edge] = round(np.prod([VERTICE1[head] for head in heads]),2)
-        EDGE2[edge] = round(np.prod([VERTICE2[head] for head in heads]),2)
-        EDGE3[edge] = round(np.prod([VERTICE3[head] for head in heads]),2)
-    elif mode == "Average":
-        EDGE1[edge] = round(sum(VERTICE1[head] for head in heads)/len(heads),2)
-        EDGE2[edge] = round(sum(VERTICE2[head] for head in heads)/len(heads),2)
-        EDGE3[edge] = round(sum(VERTICE3[head] for head in heads)/len(heads),2)
-
-    EDGEMAX[edge] = max([EDGE1[edge], EDGE2[edge], EDGE3[edge]])
-    #print(EDGEMAX[edge], HYPERGRAPH[edge])
-#print(EDGEMAX[4])
-#print(EDGEMAX[23])
-#print(EDGEMAX[42])
-
 
 FIXED_FLOWS = {
     #213: 3,
@@ -171,7 +107,6 @@ def build_model(name, hyperedges, elmax, el1, el2, el3, excluded_support=None):
 
     vertices = set(v for tails, heads in hyperedges.values() for v in tails + heads)
 
-    print(vertices)
     #Every item created has to be consumed:
     '''
     for v in vertices:
@@ -228,7 +163,7 @@ def build_model(name, hyperedges, elmax, el1, el2, el3, excluded_support=None):
     #2 Butadien create first different molecule and it has to be created first:
     startmolecule = ["Butadien"]
     
-    model.addConstr(quicksum(b[e_id] for e_id, (tails, _) in hyperedges.items() if list(set(tails)) == startmolecule) == 1)
+    #model.addConstr(quicksum(b[e_id] for e_id, (tails, _) in hyperedges.items() if list(set(tails)) == startmolecule) == 1)
     #model.addConstr(b[4] + b[7] == 1)
 
     model.addConstr(quicksum(b[e_id] for e_id, (tails, heads) in hyperedges.items() if tails == [] and heads == startmolecule)== 1)
@@ -251,8 +186,8 @@ def build_model(name, hyperedges, elmax, el1, el2, el3, excluded_support=None):
 def positive_entries(variable_dict, threshold = 0.5):
     return {e_id: var.X for e_id, var in variable_dict.items() if var.X > threshold}
 
-def edgelikelihoods(variable_dict, threshold = 0.5):
+def edgelikelihoods(variable_dict, edgelikelihood, threshold = 0.5):
-    return {e_id: EDGEMAX[e_id] for e_id, var in variable_dict.items() if var.X > threshold}
+    return {e_id: edgelikelihood[e_id] for e_id, var in variable_dict.items() if var.X > threshold}
 
 def print_solution(title, flow_solution, binary_solution, edge_likelihoods, hyperedges):
     print(f"\n{title}:")
@@ -269,6 +204,73 @@ def print_solution(title, flow_solution, binary_solution, edge_likelihoods, hype
     print(f"Number of used hyperedges: {len(binary_solution)}")
 
 def main():
+    VERTICES = ['Butadien', 'p_{0,0}', 'p_{0,1}', 'p_{0,2}', 'p_{0,3}', 'p_{0,4}', 'p_{0,5}', 'p_{0,6}', 'p_{0,7}', 'p_{0,8}', 'p_{0,9}', 'p_{0,10}', 'p_{0,11}', 'p_{0,12}', 'p_{0,13}', 'p_{0,14}', 'p_{0,15}', 'p_{0,16}', 'p_{0,17}', 'p_{0,18}']
+    #Vergleich mit dem NMR von Ethylen und Hexatrien
+    NMR1 = [0.32, 0.5, 0.87, 0.0, 0.11, 0.58, 0.06, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.09, 0.0, 0.0, 0.17, 0.0, 0.09]
+    #Vergleich mit dem NMR von Ethylen und Octrien
+    NMR2 = [0.33, 0.43, 0.6, 0.0, 0.13, 0.9, 0.07, 0.01, 0.0, 0.0, 0.01, 0.0, 0.0, 0.01, 0.1, 0.0, 0.0, 0.22, 0.0, 0.13]
+    #Vergleich mit dem NMR von Ethylen und Benzol
+    NMR3 = [0.0, 0.32, 0.0, 0.0, 0.0, 0.0, 0.05, 0.0, 0.0, 0.02, 0.0, 0.29, 0.95, 0.17, 0.0, 0.0, 0.0, 0.0, 0.0, 0.05]
+    VERTICESSMILES = ['C=CC=C', 'C=C', 'C=CC=CC=C', 'C1CCC(C=C)CC=1', 'C(CCC(C=C)CC=C)=C', 'C=CC=CC=CC=C', 'C1CCC(C=CC=C)CC=1', 'C=CC1C=CCCC1', 'C1CCCCC=1', 'C=CC1CC=CCC1C=C', 'C1CCC(C=C)C(C=C)C=1', 'C(C1CC(C=C)C=CC1)=C', 'C1C=CC=CC=1', 'C(CCC1C=CC=CC1)=C', 'C1C(C=CC=C)CCCC=1', 'C=CC(C=C)CCCC=C', 'C=CCCCCC=C', 'C=CC1C=CC(C=C)CC1', 'C1CC2CCCCC2CC=1', 'C1CC2C=CC=CC2CC=1']
+
+    #Chosable parameters
+    modes = ["Product", "Average"]
+    mode = modes[1] 
+    normalize = True
+
+    if normalize:
+        print("test")
+        NMR1 = [round(l/sum(NMR1), 2) for l in NMR1]
+        NMR2 = [round(l/sum(NMR2), 2) for l in NMR2]
+        NMR3 = [round(l/sum(NMR3), 2) for l in NMR3]
+
+        
+    #Kombiniert Molekül it Wahrscheinlichkeit für NMR1:
+    VERTICE1 = {}
+    for vertice, likelihood in zip(VERTICES, NMR1):
+        VERTICE1[vertice] = likelihood
+    #Kombiniert Molekül it Wahrscheinlichkeit für NMR2:
+    VERTICE2 = {}
+    for vertice, likelihood in zip(VERTICES, NMR2):
+        VERTICE2[vertice] = likelihood
+    #Kombiniert Molekül it Wahrscheinlichkeit für NMR3:
+    VERTICE3 = {}
+    for vertice, likelihood in zip(VERTICES, NMR3):
+        VERTICE3[vertice] = likelihood
+
+    #Kantenwahrscheinlichkeiten für NMR1:
+    EDGE1 = {}
+    #Kantenwahrscheinlichkeiten für NMR2:
+    EDGE2 = {}
+    #Kantenwahrscheinlichkeiten für NMR3:
+    EDGE3 = {}
+    #Maximum der verschiedenen Kantenwahrscheinlichkeiten:
+    EDGEMAX = {}
+ 
+    #Ist es besser, die Edukte mit zu berücksichtigen? Dann hin rück gleich wahrscheinlich, aber nur 1 kann gewählt werden
+    #Ordnet die Kantenwahrscheinlichkeit basierend auf dem Mittel oder Produkt der Wahrscheinlichkeiten der Produkte
+    for edge, (tails, heads) in HYPERGRAPH.items():
+        if heads == [] or tails == []:
+            EDGE1[edge] = 0.0
+            EDGE2[edge] = 0.0
+            EDGE3[edge] = 0.0
+        elif mode == "Product":
+            #print([VERTICE1[head] for head in heads])
+            EDGE1[edge] = round(np.prod([VERTICE1[head] for head in heads]),2)
+            EDGE2[edge] = round(np.prod([VERTICE2[head] for head in heads]),2)
+            EDGE3[edge] = round(np.prod([VERTICE3[head] for head in heads]),2)
+        elif mode == "Average":
+            EDGE1[edge] = round(sum(VERTICE1[head] for head in heads)/len(heads),2)
+            EDGE2[edge] = round(sum(VERTICE2[head] for head in heads)/len(heads),2)
+            EDGE3[edge] = round(sum(VERTICE3[head] for head in heads)/len(heads),2)
+
+        EDGEMAX[edge] = max([EDGE1[edge], EDGE2[edge], EDGE3[edge]])
+        #print(EDGEMAX[edge], HYPERGRAPH[edge])
+    #print(EDGEMAX[4])
+    #print(EDGEMAX[23])
+    #print(EDGEMAX[42])
+
+
     model, x, b = build_model("HypergraphFlow", HYPERGRAPH, EDGEMAX, EDGE1, EDGE2, EDGE3)
     model.optimize()
     if model.status != GRB.Status.OPTIMAL:
@@ -276,7 +278,7 @@ def main():
         return
     optimal_solution = positive_entries(x)
     optimal_binary_solution = positive_entries(b)
-    optimal_edgelikelihoods = edgelikelihoods(b)
+    optimal_edgelikelihoods = edgelikelihoods(b, EDGEMAX)
     print_solution("Optimal Solution", optimal_solution, optimal_binary_solution, optimal_edgelikelihoods, HYPERGRAPH)
 
     excluded_support = list(optimal_binary_solution.keys())
@@ -286,7 +288,7 @@ def main():
     if second_model.status == GRB.Status.OPTIMAL:
         second_solution = positive_entries(x2)
         second_binary_solution = positive_entries(b2)
-        second_edgelikelihoods = edgelikelihoods(b2)
+        second_edgelikelihoods = edgelikelihoods(b2, EDGEMAX)
         print_solution("Second Best Solution", second_solution, second_binary_solution, second_edgelikelihoods, HYPERGRAPH)
     else:
         print("No optimal solution found for the second best model.")