Merge pull request #38 from lacava/reproducible

replaces np.random calls with self.random_state (issue #37 )
lacava · Aug 3, 2018 · d736e92 · d736e92
2 parents 2bdf074 + 62f6f91
commit d736e92
Show file tree

Hide file tree

Showing 5 changed files with 12 additions and 11 deletions.
diff --git a/docs/few_example.py b/docs/few_example.py
@@ -7,7 +7,8 @@
 input_data = pd.read_csv(dataset,sep=None,engine='python')
 
 #generate train/test split
-train_i, test_i = train_test_split(input_data.index, train_size=0.75, test_size=0.25)
+train_i, test_i = train_test_split(input_data.index, train_size=0.75, test_size=0.25,
+                                    random_state=10)
 
 # training data
 X_train = input_data.loc[train_i].drop('label', axis=1).values
@@ -17,7 +18,7 @@
 X_test = input_data.loc[test_i].drop('label', axis=1).values
 Y_test = input_data.loc[test_i, 'label'].values
 
-few = FEW(verbosity=1)
+few = FEW(random_state=10,verbosity=1)
 few.fit(X_train,Y_train)
 
 print('\nTraining accuracy: {}'.format(few.score(X_train, Y_train)))

diff --git a/few/lib/epsilon_lexicase.h b/few/lib/epsilon_lexicase.h
@@ -161,7 +161,7 @@ void epsilon_lexicase(const ExtMat & F, int n, int d,
 
     // pick a winner from can_locs
     locs(i) = *select_randomly(can_locs.begin(),can_locs.end(),gen);
-    // remove the winner from ind_locs
+    // survival: remove the winner from ind_locs
     for (auto l = ind_locs.begin(); l!=ind_locs.end();){
       if (*l == locs(i))
         l = ind_locs.erase(l);

diff --git a/few/population.py b/few/population.py
@@ -214,13 +214,13 @@ def make_program(self,stack,func_set,term_set,max_d,ntype):
                 raise ValueError('no ts. ntype:'+ntype+'. term_set out_types:'+
                                  ','.join([t.out_type for t in term_set]))
 
-            stack.append(ts[np.random.choice(len(ts))])
+            stack.append(ts[self.random_state.choice(len(ts))])
         else:
             fs = [f for f in func_set if (f.out_type==ntype
                                           and (f.in_type=='f' or max_d>1))]
             if len(fs)==0:
                 print('ntype:',ntype,'\nfunc_set:',[f.name for f in func_set])
-            stack.append(fs[np.random.choice(len(fs))])
+            stack.append(fs[self.random_state.choice(len(fs))])
             tmp = copy.copy(stack[-1])
 
             for i in np.arange(tmp.arity['f']):
@@ -273,7 +273,7 @@ def init_pop(self):
                 for i,p in enumerate(pop.individuals):
                     if i < self.n_features:
                         p.stack = [node('x',
-                                        loc=np.random.randint(self.n_features))]
+                                        loc=self.random_state.randint(self.n_features))]
                     else:
                         # make program if pop is bigger than n_features
                         self.make_program(p.stack,self.func_set,self.term_set,

diff --git a/few/variation.py b/few/variation.py
@@ -32,7 +32,7 @@ def variation(self,parents):
                     # softmax transformation of the weights
                     weights = np.exp(weights)/np.sum(np.exp(weights))
                     offspring = copy.deepcopy(
-                        list(np.random.choice(self.valid(parents),
+                        list(self.random_state.choice(self.valid(parents),
                                               self.population_size, p=weights)))
                 else:
                     offspring = copy.deepcopy(list(
@@ -45,7 +45,7 @@ def variation(self,parents):
                     # softmax transformation of the weights
                     weights = np.exp(weights)/np.sum(np.exp(weights))
                     offspring = copy.deepcopy(list(
-                        np.random.choice(self.valid(parents),
+                        self.random_state.choice(self.valid(parents),
                                          self.population_size, p=weights)))
                 else:
                     offspring = copy.deepcopy(list(

diff --git a/setup.py b/setup.py
@@ -10,10 +10,10 @@
 import importlib
 try:
     importlib.import_module('eigency')
-except ImportError:
+except (ImportError, AttributeError):
     try:
-        import pip
-        pip.main(['install', 'eigency'])
+        from pip._internal import main 
+        main(['install', 'eigency'])
     except ImportError:
         raise ImportError('The eigency library must be installed before FEW. '
                           'Automatic install with pip failed.')