Depth Estimation
File size: 7,314 Bytes
0d6eb21
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
import torch
import torch.nn as nn
import torch.optim as optim
import matplotlib.pyplot as plt

# Define grid size
grid_size = 20

# Create a grid with random initial wealth data
wealth_data = torch.rand((grid_size, grid_size))

# Define a simple neural network that will adjust the wealth data
class WealthNet(nn.Module):
    def __init__(self):
        super(WealthNet, self).__init__()
        self.fc1 = nn.Linear(grid_size * grid_size, 128)
        self.fc2 = nn.Linear(128, grid_size * grid_size)

    def forward(self, x):
        x = torch.relu(self.fc1(x))
        x = self.fc2(x)
        return x

# Instantiate the network, loss function, and optimizer
net = WealthNet()
criterion = nn.MSELoss()
optimizer = optim.Adam(net.parameters(), lr=0.01)

# Target direction to direct wealth (e.g., bottom right corner)
target_wealth = torch.zeros((grid_size, grid_size))
target_wealth[-5:, -5:] = 1  # Direct wealth towards the bottom right corner

# Convert the grid to a single vector for the neural network
input_data = wealth_data.view(-1)
target_data = target_wealth.view(-1)

# Training the network
epochs = 500
for epoch in range(epochs):
    optimizer.zero_grad()
    output = net(input_data)
    loss = criterion(output, target_data)
    loss.backward()
    optimizer.step()

# Reshape the output to the grid size
output_grid = output.detach().view(grid_size, grid_size)

# Plot the original and adjusted wealth distribution
fig, axes = plt.subplots(1, 2, figsize=(12, 6))
axes[0].imshow(wealth_data, cmap='viridis')
axes[0].set_title('Original Wealth Distribution')
axes[1].imshow(output_grid, cmap='viridis')
axes[1].set_title('Directed Wealth Distribution')
plt.show()

import torch
import torch.nn as nn
import torch.optim as optim
import matplotlib.pyplot as plt

# Define grid size
grid_size = 20

# Create a grid with random initial wealth data
wealth_data = torch.rand((grid_size, grid_size))

# Define a neural network with an additional layer for infrared conversion
class WealthNet(nn.Module):
    def __init__(self):
        super(WealthNet, self).__init__()
        self.fc1 = nn.Linear(grid_size * grid_size, 128)
        self.fc2 = nn.Linear(128, 128)
        self.fc3 = nn.Linear(128, grid_size * grid_size)
        self.infrared_layer = nn.Sigmoid()  # Simulating the conversion to infrared energy

    def forward(self, x):
        x = torch.relu(self.fc1(x))
        stored_wealth = torch.relu(self.fc2(x))  # Store wealth data here
        infrared_energy = self.infrared_layer(stored_wealth)  # Convert to infrared energy
        x = self.fc3(infrared_energy)
        return x, stored_wealth, infrared_energy

# Instantiate the network, loss function, and optimizer
net = WealthNet()
criterion = nn.MSELoss()
optimizer = optim.Adam(net.parameters(), lr=0.01)

# Target direction to direct wealth (e.g., bottom right corner)
target_wealth = torch.zeros((grid_size, grid_size))
target_wealth[-5:, -5:] = 1  # Direct wealth towards the bottom right corner

# Convert the grid to a single vector for the neural network
input_data = wealth_data.view(-1)
target_data = target_wealth.view(-1)

# Training the network
epochs = 500
for epoch in range(epochs):
    optimizer.zero_grad()
    output, stored_wealth, infrared_energy = net(input_data)
    loss = criterion(output, target_data)
    loss.backward()
    optimizer.step()

# Reshape the outputs to the grid size
output_grid = output.detach().view(grid_size, grid_size)
stored_wealth_grid = stored_wealth.detach().view(128)  # Displayed as a 1D representation
infrared_energy_grid = infrared_energy.detach().view(128)  # Displayed as a 1D representation

# Plot the original and adjusted wealth distribution
fig, axes = plt.subplots(1, 4, figsize=(20, 6))
axes[0].imshow(wealth_data, cmap='viridis')
axes[0].set_title('Original Wealth Distribution')
axes[1].imshow(output_grid, cmap='viridis')
axes[1].set_title('Directed Wealth Distribution')
axes[2].plot(stored_wealth_grid.numpy())
axes[2].set_title('Stored Wealth Data (1D)')
axes[3].plot(infrared_energy_grid.numpy())
axes[3].set_title('Infrared Energy (1D)')
plt.show()

import torch
import torch.nn as nn
import torch.optim as optim
import matplotlib.pyplot as plt

# Define grid size
grid_size = 20

# Create a grid with random initial wealth data
wealth_data = torch.rand((grid_size, grid_size))

# Define a neural network with an additional layer for data protection
class WealthNet(nn.Module):
    def __init__(self):
        super(WealthNet, self).__init__()
        self.fc1 = nn.Linear(grid_size * grid_size, 128)
        self.fc2 = nn.Linear(128, 128)
        self.fc3 = nn.Linear(128, grid_size * grid_size)
        self.infrared_layer = nn.Sigmoid()  # Simulating the conversion to infrared energy
        # Removed the incorrect instantiation of GaussianNoise here

    def forward(self, x):
        x = torch.relu(self.fc1(x))
        stored_wealth = torch.relu(self.fc2(x))  # Store wealth data here
        protected_wealth = self.protection_layer(stored_wealth)  # Protect the stored data
        infrared_energy = self.infrared_layer(protected_wealth)  # Convert to infrared energy
        x = self.fc3(infrared_energy)
        return x, stored_wealth, protected_wealth, infrared_energy

# Custom layer to add Gaussian noise (PyTorch does not have this built-in)
class GaussianNoise(nn.Module):
    def __init__(self, stddev):
        super(GaussianNoise, self).__init__()
        self.stddev = stddev

    def forward(self, x):
        if self.training:
            noise = torch.randn_like(x) * self.stddev
            return x + noise
        return x

# Instantiate the network, loss function, and optimizer
net = WealthNet()
# Add the GaussianNoise layer to the network instance
net.protection_layer = GaussianNoise(0.1)
criterion = nn.MSELoss()
optimizer = optim.Adam(net.parameters(), lr=0.01)

# Target direction to direct wealth (e.g., bottom right corner)
target_wealth = torch.zeros((grid_size, grid_size))
target_wealth[-5:, -5:] = 1  # Direct wealth towards the bottom right corner

# Convert the grid to a single vector for the neural network
input_data = wealth_data.view(-1)
target_data = target_wealth.view(-1)

# Training the network
epochs = 500
for epoch in range(epochs):
    optimizer.zero_grad()
    output, stored_wealth, protected_wealth, infrared_energy = net(input_data)
    loss = criterion(output, target_data)
    loss.backward()
    optimizer.step()

# Reshape the outputs to the grid size
output_grid = output.detach().view(grid_size, grid_size)
stored_wealth_grid = stored_wealth.detach().view(128)  # Displayed as a 1D representation
protected_wealth_grid = protected_wealth.detach().view(128)  # Displayed as a 1D representation
infrared_energy_grid = infrared_energy.detach().view(128)  # Displayed as a 1D representation

# Plot the original and adjusted wealth distribution
fig, axes = plt.subplots(1, 5, figsize=(25, 6))
axes[0].imshow(wealth_data, cmap='viridis')
axes[0].set_title('Original Wealth Distribution')
axes[1].imshow(output_grid, cmap='viridis')
axes[1].set_title('Directed Wealth Distribution')
axes[2].plot(stored_wealth_grid.numpy())
axes[2].set_title('Stored Wealth Data (1D)')
axes[3].plot(protected_wealth_grid.numpy())
axes[3].set_title('Protected Wealth Data (1D)')
axes[4].plot(infrared_energy_grid)