moved data extraction to /util

test_data_extraction.py

@@ -12,7 +12,7 @@ def main():
     #instance_parameters()
     #wmis_results()
     #wmis_siman_results_alpha_num_of_assignments()
-    #wmis_qpu_results_alpha_num_of_assignments()
+    wmis_qpu_results_alpha_num_of_assignments()
     #primitive_2_siman_results_alpha_num_of_assignments()
     #primitive_5_siman_results_alpha_num_of_assignments()
     #primitive_5_qpu_results_alpha_num_of_assignments()
@@ -21,7 +21,7 @@ def main():
     #minisat_runs()
     
     #wmis_engergy()
-    wmis_2_engergy()
+    #wmis_2_engergy()
 
 def instance_parameters():
     edb = script.connect_to_experimetns_db()

util/data_extraction.py

@@ -0,0 +1,216 @@
+from . import script
+from . import queries
+import dimod
+import random 
+import numpy as np
+
+from tqdm import tqdm
+
+def extract_wmis_qpu_results():
+    edb = script.connect_to_experimetns_db()
+    edb_cursor = edb.cursor()
+    
+    idb = script.connect_to_instance_pool()
+    
+    scope = input("scope: ")
+    result_collection = input("result collection: ")
+    
+    table_name = input("table name: ")
+    
+    q = queries.WMIS_result_scope_query_raw(idb)
+    q.query(scope, result_collection)
+    
+    insert_row = ("INSERT INTO {} "
+                  "(result_id, "
+                  " run, "
+                  " instance_id, "
+                  " number_of_found_assignments, "
+                  " chain_break_fraction, "
+                  " num_occurrences, "
+                  " energy, "
+                  " satisfiable, "
+                  " anneal_time, "
+                  " energy_reach, "
+                  " sample_size) "
+                  "VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s) ").format(table_name)
+    
+    run = int(input("run: "))
+    
+    for result in tqdm(q):
+        if result["run"] == run:
+            sample_set = queries.read_raw_wmis_sample_set(result["data"])
+        
+            data = script.analyze_wmis_sample(sample_set.first)
+        
+            sat = queries.get_instance_by_id(idb["instances"], result["instance"])
+        
+            model = script.majority_vote_sample(sample_set.first.sample)
+        
+            isSatisfiable = sat.checkAssignment(model)
+        
+            anneal_time = result["data"]["info"]["timing"]["qpu_anneal_time_per_sample"]
+    
+            energy_reach = abs(sample_set.record["energy"].max() - 
+                               sample_set.record["energy"].min())
+            
+            sample_size = np.sum(sample_set.record["num_occurrences"])
+            
+            edb_cursor.execute(insert_row, (str(result["_id"]),
+                                            int(result["run"]),
+                                            str(result["instance"]),
+                                            int(data["number_of_assignments"]),
+                                            float(data["chain_break_fraction"]),
+                                            int(data["num_occurrences"]),
+                                            int(data["energy"]),
+                                            isSatisfiable,
+                                            int(anneal_time),
+                                            int(energy_reach),
+                                            int(sample_size)))
+        
+    edb.commit()
+    edb_cursor.close()
+    edb.close()
+    
+def extract_wmis_2_qpu_results():
+    edb = script.connect_to_experimetns_db()
+    edb_cursor = edb.cursor()
+    
+    idb = script.connect_to_instance_pool()
+    
+    scope = input("scope: ")
+    result_collection = input("result collection: ")
+    table_name = input("table name: ")
+    
+    q = queries.WMIS_result_scope_query_raw(idb)
+    q.query(scope, result_collection)
+    
+    insert_row = ("INSERT INTO {} "
+                  "(result_id, "
+                  " run, "
+                  " instance_id, "
+                  " chain_break_fraction, "
+                  " num_occurrences, "
+                  " energy, "
+                  " satisfiable, "
+                  " anneal_time, "
+                  " energy_reach, "
+                  " sample_size) "
+                  "VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s) ").format(table_name)
+    
+        
+    run = int(input("run: "))
+    
+    for result in tqdm(q):
+        if result["run"] == run:
+            sample_set = queries.read_raw_primitive_ising_sample_set(result["data"])
+            
+            data = script.analyze_wmis_sample(sample_set.first)
+            
+            sat = queries.get_instance_by_id(idb["instances"], result["instance"])
+            
+            post_process_results = __evaluate_wmis_2_sample(sat,
+                                                            sample_set.first.sample)
+            
+            anneal_time = result["data"]["info"]["timing"]["qpu_anneal_time_per_sample"]
+            
+            energy_reach = abs(sample_set.record["energy"].max() - 
+                               sample_set.record["energy"].min())
+            sample_size = np.sum(sample_set.record["num_occurrences"])
+            
+            edb_cursor.execute(insert_row, (str(result["_id"]),
+                                            int(result["run"]),
+                                            str(result["instance"]),
+                                            float(data["chain_break_fraction"]),
+                                            int(data["num_occurrences"]),
+                                            int(data["energy"]),
+                                            bool(post_process_results["satisfiable"]),
+                                            int(anneal_time),
+                                            int(energy_reach),
+                                            int(sample_size)))
+        
+    edb.commit()
+    edb_cursor.close()
+    edb.close()
+
+def __evaluate_wmis_2_sample(sat, sample):
+    post_process_results = {}
+    
+    assignments = {}
+    vars = set()
+    
+    for node, energy in sample.items():
+        if node[0] == "x":
+            lit = int(node[1:])
+            
+            vars.add(abs(lit))
+            
+            assignments[lit] = energy
+                
+    model = [True for i in range(len(vars))]
+    
+    for var in vars:
+        if var in assignments:
+            model[var - 1] = True if assignments[var] == 1 else False
+        elif -var in assignments:
+            model[var - 1] = True if assignments[-var] == 0 else False
+            
+    post_process_results["satisfiable"] = sat.checkAssignment(model)
+    
+    return post_process_results
+
+def extract_minisat_results():
+    edb = script.connect_to_experimetns_db()
+    edb_cursor = edb.cursor();
+    
+    idb = script.connect_to_instance_pool()
+    
+    scope = input("scope: ")
+    result_collection = input("result collection: ")
+    table_name = input("table name: ")
+    
+    runs = queries.Minisat_run_scope_query_raw(idb)
+    runs.query(scope, result_collection)
+    
+    insert_row = ("INSERT INTO {} "
+                  "(run_id, "
+                  " instance_id, "
+                  " satisfiable) "
+                  "VALUES (%s, %s, %s) ").format(table_name)
+    
+    for run in tqdm(runs):
+        data = script.analyde_minisat_run(run)
+        
+        edb_cursor.execute(insert_row, (str(run["_id"]), 
+                                        str(run["instance"]),
+                                        int(data["satisfiable"])))
+    
+    edb.commit()
+    edb_cursor.close()
+    edb.close()
+    
+def extract_instance_parameters():
+    edb = script.connect_to_experimetns_db()
+    edb_cursor = edb.cursor()
+    
+    idb = script.connect_to_instance_pool()
+    
+    scope = input("scope: ")
+    table_name = input("table name: ")
+    
+    instances = queries.Instance_scope_query(idb)
+    instances.query(scope)
+    
+    insert_row = ("INSERT INTO {} "
+                  "(instance_id, "
+                  " number_of_clauses, "
+                  " number_of_variables) "
+                  "VALUES (%s, %s, %s)").format(table_name)
+                  
+    for instance, instance_id in tqdm(instances):
+        edb_cursor.execute(insert_row, (str(instance_id),
+                                        int(instance.getNumberOfClauses()),
+                                        int(instance.getNumberOfVariables())))
+    
+    edb.commit()
+    edb_cursor.close()
+    edb.close()

util/queries.py

@@ -562,5 +562,4 @@ def extract_primitive_ising_model(sample):
         
     return model
     
-    
-    
+