added some files

MetaAugment/autoaugment_learners/README.md

+write `import MetaAugment.autoaugment_learners as aa`
+and `aa_learner = aa.randomsearch_learner()`
+to use
\ No newline at end of file

MetaAugment/autoaugment_learners/__init__.py

+from .randomsearch_learner import *
\ No newline at end of file

MetaAugment/autoaugment_learners/aa_learner.py

+# DUMMY PSEUDOCODE!
+#
+# this might become the superclass for all other autoaugment_learners
+# This is sort of how our AA_Learner class should look like:
+
+class aa_learner:
+    def __init__(self, controller):
+        self.controller = controller
+
+    def learn(self, train_dataset, test_dataset, child_network, res, toy_flag):
+        '''
+        Does what is seen in Figure 1 in the AutoAugment paper.
+
+        'res' stands for resolution of the discretisation of the search space. It could be
+        a tuple, with first entry regarding probability, second regarding magnitude
+        '''
+        good_policy_found = False
+
+        while not good_policy_found:
+            policy = self.controller.pop_policy()
+
+            train_loader, test_loader = create_toy(train_dataset, test_dataset,
+                                                    batch_size=32, n_samples=0.005)
+
+            reward = train_child_network(child_network, train_loader, test_loader, sgd, cost, epoch)
+
+            self.controller.update(reward, policy)
+        
+        return good_policy
\ No newline at end of file

MetaAugment/autoaugment_learners/randomsearch_learner.py

+import MetaAugment.autoaugment_learners as aa
+
+
+class RandomSearch_Learner(aa):
+    def __init__(self):
+        super().__init__()
+        
+
+
+
+
+def randomsearch_learner():
+    model = RandomSearch_Learner()
+    return model
\ No newline at end of file

MetaAugment/child_networks/README.md

+write `import MetaAugment.child_networks as child_networks`
+and `child_network = child_networks.lenet()`
+to use
\ No newline at end of file

MetaAugment/child_networks/__init__.py

+from .lenet import *
+from .bad_lenet import *
\ No newline at end of file

MetaAugment/child_networks/bad_lenet.py

+import torch.nn as nn
+
+
+class Bad_LeNet(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.conv1 = nn.Conv2d(1, 6, 5)
+        self.relu1 = nn.ReLU()
+        self.pool1 = nn.MaxPool2d(2)
+        self.conv2 = nn.Conv2d(6, 16, 5)
+        self.relu2 = nn.ReLU()
+        self.pool2 = nn.MaxPool2d(2)
+        self.fc1 = nn.Linear(256, 120)
+        self.relu3 = nn.ReLU()
+        self.fc2 = nn.Linear(120, 84)
+        self.relu4 = nn.ReLU()
+        self.fc3 = nn.Linear(84, 10)
+        self.relu5 = nn.ReLU()
+
+    def forward(self, x):
+        y = self.conv1(x)
+        y = self.relu1(y)
+        y = self.pool1(y)
+        y = self.conv2(y)
+        y = self.relu2(y)
+        y = self.pool2(y)
+        y = y.view(y.shape[0], -1)
+        y = self.fc1(y)
+        y = self.relu3(y)
+        y = self.fc2(y)
+        y = self.relu4(y)
+        y = self.fc3(y)
+        y = self.relu5(y)
+        return y
+
+
+def bad_lenet():
+    model = Bad_LeNet()
+    return model
\ No newline at end of file

MetaAugment/child_networks/lenet.py

+import torch.nn as nn
+
+
+class LeNet(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.conv1 = nn.Conv2d(1, 6, 5)
+        self.relu1 = nn.ReLU()
+        self.pool1 = nn.MaxPool2d(2)
+        self.conv2 = nn.Conv2d(6, 16, 5)
+        self.relu2 = nn.ReLU()
+        self.pool2 = nn.MaxPool2d(2)
+        self.fc1 = nn.Linear(256, 120)
+        self.relu3 = nn.ReLU()
+        self.fc2 = nn.Linear(120, 84)
+        self.relu4 = nn.ReLU()
+        self.fc3 = nn.Linear(84, 10)
+        self.relu5 = nn.ReLU()
+
+    def forward(self, x):
+        y = self.conv1(x)
+        y = self.relu1(y)
+        y = self.pool1(y)
+        y = self.conv2(y)
+        y = self.relu2(y)
+        y = self.pool2(y)
+        y = y.view(y.shape[0], -1)
+        y = self.fc1(y)
+        y = self.relu3(y)
+        y = self.fc2(y)
+        y = self.relu4(y)
+        y = self.fc3(y)
+        y = self.relu5(y)
+        return y
+
+
+def lenet():
+    model = LeNet()
+    return model
\ No newline at end of file