Alternative genetic algorithm

MetaAugment/CP2_Max.py

@@ -22,6 +22,24 @@ np.random.seed(0)
 random.seed(0)
 
 
+# augmentation_space = [
+#             # (function_name, do_we_need_to_specify_magnitude)
+#             ("ShearX", True),
+#             ("ShearY", True),
+#             ("TranslateX", True),
+#             ("TranslateY", True),
+#             ("Rotate", True),
+#             ("Brightness", True),
+#             ("Color", True),
+#             ("Contrast", True),
+#             ("Sharpness", True),
+#             ("Posterize", True),
+#             ("Solarize", True),
+#             ("AutoContrast", False),
+#             ("Equalize", False),
+#             ("Invert", False),
+#         ]
+
 class Learner(nn.Module):
     def __init__(self, num_transforms = 3):
         super().__init__()
@@ -38,6 +56,7 @@ class Learner(nn.Module):
         self.fc3 = nn.Linear(84, 13)
         # self.sig = nn.Sigmoid()
 
+
     def forward(self, x):
         y = self.conv1(x)
         y = self.relu1(y)
@@ -60,7 +79,6 @@ class Learner(nn.Module):
         p_ret = 0.1 * torch.argmax(y[:, 3:].mean(dim = 0))
         return (idx_ret, p_ret)
 
-        # return (torch.argmax(y[0:3]), y[torch.argmax(y[3:])])
 
 class LeNet(nn.Module):
     def __init__(self):
@@ -253,7 +271,7 @@ class Evolutionary_learner():
         self.num_parents_mating = num_parents_mating
         self.initial_population = self.torch_ga.population_weights
         self.train_loader = train_loader
-        self.backup_model = sec_model
+        self.sec_model = sec_model
 
         assert num_solutions > num_parents_mating, 'Number of solutions must be larger than the number of parents mating!'
 
@@ -269,7 +287,7 @@ class Evolutionary_learner():
             return solution, solution_fitness, solution_idx
 
     def new_model(self):
-        copy_model = copy.deepcopy(self.backup_model)
+        copy_model = copy.deepcopy(self.sec_model)
         return copy_model