aioz-ai/fadnet

Deep Federated Learning for Autonomous Driving (IV'22)

fadnet's Issues

Some confusions about network structure

Hi, AIOZ AI!
Thanks for your great work.

I'm a little confused about some of the snippets in the code.

1. According to Fig 3 in the paper, Feature 1, Feature 2 and Feature 3 are not the output of ResBlock1, ResBlock2 and ResBlock3 respectively. However, the code is implemented like this.

   def forward(self, inputs):
   
    x1 = self.norm(inputs)
    x1 = self.conv1(x1)
    x1 = self.max_pool1(x1)
   
    x2 = self.res_block1(x1)
    x1 = self.conv2(x1)
    x3 = torch.add(x1, x2)
   
    x3_1 = x3.view(inputs.shape[0], -1)
   
    x4 = self.res_block2(x3)
    x3 = self.conv3(x3)
    x4 = torch.add(x3, x4)
   
    x4_1 = x4.view(inputs.shape[0], -1)
   
    x5 = self.res_block3(x4)
    x4 = self.conv4(x4)
    x5 = torch.add(x4, x5)
   
    x6 = x5.view(inputs.shape[0], -1)
    x6 = self.relu(x6)
    x6 = self.dropout(x6)
   
    return 0.7*self.fc(x6) + 0.1*x3_1.mean() + 0.1*x4_1.mean() + 0.1*x6.mean()
   

2. In this line of code("0.7self.fc(x6) + 0.1x3_1.mean() + 0.1x4_1.mean() + 0.1x6.mean()"), what is the basis for choosing the weights ?

Cannot generate networks for CARLA dataset

I have cloned this repo and am trying to generate networks for the CARLA dataset. However, when I try to run this command, I get the following error:

After running
bash generate_network_driving-carla.sh

I get:

File "rkumar/fadnet/graph_utils/generate_networks.py", line 151, in <module>
    cycle_time = get_matcha_cycle_time(underlay.copy(), connectivity_graph.copy(),
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "rkumar/fadnet/graph_utils/utils/utils.py", line 271, in get_matcha_cycle_time
  topology_generator = RandomTopologyGenerator(underlay.copy(),
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "rkumar/fadnet/graph_utils/utils/matcha.py", line 32, in __init__
  self.matching_list, self.laplacian_matrices = matching_decomposition(self.network)
                                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "rkumar/fadnet/graph_utils/utils/matching_decomposition.py", line 26, in matching_decomposition
  laplacian_matrices = [nx.laplacian_matrix(matching, nodelist=graph.nodes(), weight=None).toarray()
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "rkumar/fadnet/graph_utils/utils/matching_decomposition.py", line 26, in <listcomp>
    laplacian_matrices = [nx.laplacian_matrix(matching, nodelist=graph.nodes(), weight=None).toarray()
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "rkumar/.local/lib/python3.11/site-packages/networkx/utils/decorators.py", line 816, in func
    return argmap._lazy_compile(__wrapper)(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "<class 'networkx.utils.decorators.argmap'> compilation 36", line 4, in argmap_laplacian_matrix_33
  File "rkumar/.local/lib/python3.11/site-packages/networkx/linalg/laplacianmatrix.py", line 54, in laplacian_matrix
    A = nx.to_scipy_sparse_array(G, nodelist=nodelist, weight=weight, format="csr")
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "rkumar/.local/lib/python3.11/site-packages/networkx/convert_matrix.py", line 891, in to_scipy_sparse_array
    raise nx.NetworkXError(f"Node {n} in nodelist is not in G")
networkx.exception.NetworkXError: Node Denver in nodelist is not in G

I am not sure how to debug, thanks.