11

model.py

@@ -0,0 +1,46 @@
+import torch
+import torch.nn as nn
+
+N_VEL = 2
+N_SUB = 2
+state_dim = N_VEL * N_SUB
+num_actions = 2 ** (N_SUB * N_VEL)
+M = 16
+
+
+class Net(nn.Module):
+    def __init__(self, n_input_channels, n_actions):
+        super(Net, self).__init__()
+        self.conv1 = nn.Conv2d(in_channels=n_input_channels, out_channels=16, kernel_size=5, stride=1, padding=0)
+        self.conv2 = nn.Conv2d(in_channels=16, out_channels=32, kernel_size=3, stride=1, padding=0)
+        self.fc1 = nn.Linear(in_features=4 * 4 * 32, out_features=512)
+        self.fc2 = nn.Linear(in_features=512, out_features=512)
+        self.fc3 = nn.Linear(in_features=512, out_features=n_actions)
+        # nn.init.normal_(self.fc1.weight, 0., 0.1)
+        # nn.init.constant_(self.fc1.bias, 0.1)
+        # self.fc2 = nn.Linear(500, 300)
+        # nn.init.normal_(self.fc2.weight, 0., 0.1)
+        # nn.init.constant_(self.fc2.bias, 0.1)
+        # self.fc3 = nn.Linear(300, 250)
+        # nn.init.normal_(self.fc3.weight, 0., 0.1)
+        # nn.init.constant_(self.fc3.bias, 0.1)
+        # self.fc4 = nn.Linear(250, n_actions)
+        # nn.init.normal_(self.fc4.weight, 0., 0.1)
+        # nn.init.constant_(self.fc4.bias, 0.1)
+
+    def forward(self, x):
+        x = torch.sigmoid(self.conv1(x))
+        x = torch.sigmoid(self.conv2(x))
+        x = x.view(x.size(0), -1)
+        x = torch.sigmoid(self.fc1(x))
+        x = torch.sigmoid(self.fc2(x))
+        output = self.fc3(x)
+        return output
+
+
+if __name__ == "__main__":
+    input_channels = 10 * 10 * 16
+    num_actions = 16
+    q_network = Net(input_channels, num_actions)
+
+    print(q_network)