repo
stringlengths
2
99
file
stringlengths
13
225
code
stringlengths
0
18.3M
file_length
int64
0
18.3M
avg_line_length
float64
0
1.36M
max_line_length
int64
0
4.26M
extension_type
stringclasses
1 value
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/nets/Planetoid_node_classification/load_net.py
""" Utility file to select GraphNN model as selected by the user """ from nets.Planetoid_node_classification.gated_gcn_net import GatedGCNNet, GatedGCNNet_pyg, ResGatedGCNNet_pyg from nets.Planetoid_node_classification.gcn_net import GCNNet, GCNNet_pyg from nets.Planetoid_node_classification.gat_net import GATNet, GATNet_pyg from nets.Planetoid_node_classification.graphsage_net import GraphSageNet, GraphSageNet_pyg from nets.Planetoid_node_classification.mlp_net import MLPNet, MLPNet_pyg from nets.Planetoid_node_classification.gin_net import GINNet, GINNet_pyg from nets.Planetoid_node_classification.mo_net import MoNet as MoNet_, MoNetNet_pyg from gcn_lib.sparse import MultiSeq, PlainDynBlock, ResDynBlock, DenseDynBlock, DilatedKnnGraph from gcn_lib.sparse import MLP as MLPpyg from gcn_lib.sparse import GraphConv as GraphConvNet def GatedGCN(net_params): return GatedGCNNet(net_params) def GCN(net_params): return GCNNet(net_params) def GAT(net_params): return GATNet(net_params) def GraphSage(net_params): return GraphSageNet(net_params) def MLP(net_params): return MLPNet(net_params) def GIN(net_params): return GINNet(net_params) def MoNet(net_params): return MoNet_(net_params) def GIN_pyg(net_params): model = GINNet_pyg(net_params) if net_params['neighbor_aggr_GIN'] == 'mean': model.aggr = str('mean') elif net_params['neighbor_aggr_GIN'] == 'max': model.aggr = str('max') return model def MLP_pyg(net_params): return MLPNet_pyg(net_params) def GCN_pyg(net_params): return GCNNet_pyg(net_params) def GatedGCN_pyg(net_params): return GatedGCNNet_pyg(net_params) def ResGatedGCN_pyg(net_params): return ResGatedGCNNet_pyg(net_params) def GAT_pyg(net_params): return GATNet_pyg(net_params) # self.head = GraphConv(opt.in_channels, channels, conv, act, norm, bias, heads) def GraphSage_pyg(net_params): return GraphSageNet_pyg(net_params) def MoNet_pyg(net_params): return MoNetNet_pyg(net_params) def gnn_model(MODEL_NAME, net_params): models = { 'GatedGCN': GatedGCN, 'GCN': GCN, 'GAT': GAT, 'GraphSage': GraphSage, 'MLP': MLP, 'GIN': GIN, 'MoNet': MoNet, 'MLP_pyg': MLP_pyg, 'GIN_pyg': GIN_pyg, 'GCN_pyg': GCN_pyg, 'GatedGCN_pyg': GatedGCN_pyg, 'GAT_pyg': GAT_pyg, 'GraphSage_pyg': GraphSage_pyg, 'MoNet_pyg': MoNet_pyg, 'ResGatedGCN_pyg': ResGatedGCN_pyg } return models[MODEL_NAME](net_params)
2,570
27.566667
109
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/nets/Planetoid_node_classification/graphsage_net.py
import torch import torch.nn as nn import torch.nn.functional as F import dgl """ GraphSAGE: William L. Hamilton, Rex Ying, Jure Leskovec, Inductive Representation Learning on Large Graphs (NeurIPS 2017) https://cs.stanford.edu/people/jure/pubs/graphsage-nips17.pdf """ from layers.graphsage_layer import GraphSageLayer from layers.mlp_readout_layer import MLPReadout from torch_geometric.nn import SAGEConv class GraphSageNet(nn.Module): """ Grahpsage network with multiple GraphSageLayer layers """ def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) hidden_dim = net_params['hidden_dim'] out_dim = net_params['out_dim'] n_classes = net_params['n_classes'] in_feat_dropout = net_params['in_feat_dropout'] dropout = net_params['dropout'] aggregator_type = net_params['sage_aggregator'] n_layers = net_params['L'] batch_norm = net_params['batch_norm'] residual = net_params['residual'] self.readout = net_params['readout'] self.n_classes = n_classes self.device = net_params['device'] self.embedding_h = nn.Embedding(in_dim_node, hidden_dim) # node feat is an integer self.in_feat_dropout = nn.Dropout(in_feat_dropout) self.layers = nn.ModuleList([GraphSageLayer(hidden_dim, hidden_dim, F.relu, dropout, aggregator_type, batch_norm, residual) for _ in range(n_layers-1)]) self.layers.append(GraphSageLayer(hidden_dim, out_dim, F.relu, dropout, aggregator_type, batch_norm, residual)) self.MLP_layer = MLPReadout(out_dim, n_classes) def forward(self, g, h, e): # input embedding h = self.embedding_h(h) h = self.in_feat_dropout(h) # graphsage for conv in self.layers: h = conv(g, h) # output h_out = self.MLP_layer(h) return h_out def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss """ GraphSAGE: William L. Hamilton, Rex Ying, Jure Leskovec, Inductive Representation Learning on Large Graphs (NeurIPS 2017) https://cs.stanford.edu/people/jure/pubs/graphsage-nips17.pdf # the implementation of dgl and pyg are different """ class GraphSageNet_pyg(nn.Module): """ Grahpsage network with multiple GraphSageLayer layers """ def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) hidden_dim = net_params['hidden_dim'] out_dim = net_params['out_dim'] n_classes = net_params['n_classes'] in_feat_dropout = net_params['in_feat_dropout'] dropout = net_params['dropout'] aggregator_type = net_params['sage_aggregator'] self.n_layers = net_params['L'] self.batch_norm = net_params['batch_norm'] self.residual = net_params['residual'] self.readout = net_params['readout'] self.n_classes = n_classes self.device = net_params['device'] self.embedding_h = nn.Linear(in_dim_node, hidden_dim) # node feat is an integer self.in_feat_dropout = nn.Dropout(in_feat_dropout) self.dropout = nn.Dropout(p=dropout) if self.batch_norm: self.batchnorm_h = nn.ModuleList([nn.BatchNorm1d(hidden_dim) for _ in range(self.n_layers - 1)]) self.batchnorm_h.append(nn.BatchNorm1d(out_dim)) self.layers = nn.ModuleList([SAGEConv(hidden_dim, hidden_dim) for _ in range(self.n_layers - 1)]) self.layers.append( SAGEConv(hidden_dim, out_dim)) # self.layers = nn.ModuleList([GraphSageLayer(hidden_dim, hidden_dim, F.relu, # dropout, aggregator_type, batch_norm, residual) for _ in # range(n_layers - 1)]) # self.layers.append( # GraphSageLayer(hidden_dim, out_dim, F.relu, dropout, aggregator_type, batch_norm, residual)) self.MLP_layer = MLPReadout(out_dim, n_classes) if aggregator_type == 'maxpool': self.aggr = 'max' elif aggregator_type == 'mean': self.aggr = 'mean' # to do def forward(self, h, edge_index, e): # input embedding h = self.embedding_h(h) h = self.in_feat_dropout(h) # graphsage for i in range(self.n_layers): h_in = h h = self.dropout(h) h = self.layers[i](h, edge_index) if self.batch_norm: h = self.batchnorm_h[i](h) #h = F.relu(h) if self.residual: h = h_in + h # residual connection # output h_out = self.MLP_layer(h) return h_out def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss
5,141
35.211268
122
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/nets/Planetoid_node_classification/gin_net.py
import torch import torch.nn as nn import torch.nn.functional as F import dgl from dgl.nn.pytorch.glob import SumPooling, AvgPooling, MaxPooling """ GIN: Graph Isomorphism Networks HOW POWERFUL ARE GRAPH NEURAL NETWORKS? (Keyulu Xu, Weihua Hu, Jure Leskovec and Stefanie Jegelka, ICLR 2019) https://arxiv.org/pdf/1810.00826.pdf """ from layers.gin_layer import GINLayer, ApplyNodeFunc, MLP from gcn_lib.sparse import MultiSeq, PlainDynBlock, ResDynBlock, DenseDynBlock, DilatedKnnGraph from gcn_lib.sparse import MLP as MLPpyg from gcn_lib.sparse import GraphConv as GraphConvNet # import torch_geometric as tg from torch_geometric.nn import GINConv class GINNet(nn.Module): def __init__(self, net_params): super().__init__() in_dim = net_params['in_dim'] hidden_dim = net_params['hidden_dim'] n_classes = net_params['n_classes'] dropout = net_params['dropout'] self.n_layers = net_params['L'] n_mlp_layers = net_params['n_mlp_GIN'] # GIN learn_eps = net_params['learn_eps_GIN'] # GIN neighbor_aggr_type = net_params['neighbor_aggr_GIN'] # GIN readout = net_params['readout'] # this is graph_pooling_type batch_norm = net_params['batch_norm'] residual = net_params['residual'] self.n_classes = n_classes self.device = net_params['device'] # List of MLPs self.ginlayers = torch.nn.ModuleList() self.embedding_h = nn.Embedding(in_dim, hidden_dim) for layer in range(self.n_layers): mlp = MLP(n_mlp_layers, hidden_dim, hidden_dim, hidden_dim) self.ginlayers.append(GINLayer(ApplyNodeFunc(mlp), neighbor_aggr_type, dropout, batch_norm, residual, 0, learn_eps)) # Linear function for output of each layer # which maps the output of different layers into a prediction score self.linears_prediction = torch.nn.ModuleList() for layer in range(self.n_layers+1): self.linears_prediction.append(nn.Linear(hidden_dim, n_classes)) def forward(self, g, h, e): h = self.embedding_h(h) # list of hidden representation at each layer (including input) hidden_rep = [h] for i in range(self.n_layers): h = self.ginlayers[i](g, h) hidden_rep.append(h) score_over_layer = 0 for i, h in enumerate(hidden_rep): score_over_layer += self.linears_prediction[i](h) return score_over_layer def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss class GINNet_pyg(nn.Module): def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] hidden_dim = net_params['hidden_dim'] n_classes = net_params['n_classes'] dropout = net_params['dropout'] self.n_layers = net_params['L'] n_mlp_layers = net_params['n_mlp_GIN'] # GIN learn_eps = net_params['learn_eps_GIN'] # GIN neighbor_aggr_type = net_params['neighbor_aggr_GIN'] # GIN readout = net_params['readout'] # this is graph_pooling_type batch_norm = net_params['batch_norm'] residual = net_params['residual'] self.n_classes = n_classes self.device = net_params['device'] # List of MLPs self.ginlayers = torch.nn.ModuleList() self.normlayers = torch.nn.ModuleList() self.embedding_h = nn.Linear(in_dim_node, hidden_dim) # node feat is an integer self.dropout = dropout self.batch_norm = batch_norm self.residual = residual for layer in range(self.n_layers): mlp = MLP(n_mlp_layers, hidden_dim, hidden_dim, hidden_dim) self.ginlayers.append(GINConv(ApplyNodeFunc(mlp), 0, learn_eps)) # note that neighbor_aggr_type can not work because the father # self.ginlayers.append(GINLayer(ApplyNodeFunc(mlp), neighbor_aggr_type, # dropout, batch_norm, residual, 0, learn_eps)) if batch_norm: self.normlayers.append(nn.BatchNorm1d(hidden_dim)) # Linear function for output of each layer # which maps the output of different layers into a prediction score self.linears_prediction = torch.nn.ModuleList() for layer in range(self.n_layers + 1): self.linears_prediction.append(nn.Linear(hidden_dim, n_classes)) def forward(self, h, edge_index, e): h = self.embedding_h(h) # list of hidden representation at each layer (including input) hidden_rep = [h] for i in range(self.n_layers): h_in = h h = self.ginlayers[i](h, edge_index) if self.batch_norm: h = self.normlayers[i](h) # batch normalization h = F.relu(h) # non-linear activation if self.residual: h = h_in + h # residual connection h = F.dropout(h, self.dropout, training=self.training) hidden_rep.append(h) score_over_layer = 0 for i, h in enumerate(hidden_rep): score_over_layer += self.linears_prediction[i](h) return score_over_layer def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss
5,612
35.448052
113
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/nets/Planetoid_node_classification/gcn_net.py
import torch import torch.nn as nn import torch.nn.functional as F from torch_geometric.nn import GCNConv import dgl import numpy as np """ GCN: Graph Convolutional Networks Thomas N. Kipf, Max Welling, Semi-Supervised Classification with Graph Convolutional Networks (ICLR 2017) http://arxiv.org/abs/1609.02907 """ from layers.gcn_layer import GCNLayer from layers.mlp_readout_layer import MLPReadout class GCNNet(nn.Module): def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) hidden_dim = net_params['hidden_dim'] out_dim = net_params['out_dim'] n_classes = net_params['n_classes'] in_feat_dropout = net_params['in_feat_dropout'] dropout = net_params['dropout'] n_layers = net_params['L'] self.readout = net_params['readout'] self.batch_norm = net_params['batch_norm'] self.residual = net_params['residual'] self.n_classes = n_classes self.device = net_params['device'] self.embedding_h = nn.Embedding(in_dim_node, hidden_dim) # node feat is an integer self.in_feat_dropout = nn.Dropout(in_feat_dropout) # note that the GCNLayer is a little different from the builtin function, # it averaging the received message by reduce, not c_{ij} the papers apply self.layers = nn.ModuleList([GCNLayer(hidden_dim, hidden_dim, F.relu, dropout, self.batch_norm, self.residual) for _ in range(n_layers-1)]) self.layers.append(GCNLayer(hidden_dim, out_dim, F.relu, dropout, self.batch_norm, self.residual)) self.MLP_layer = MLPReadout(out_dim, n_classes) def forward(self, g, h, e): # input embedding h = self.embedding_h(h) h = self.in_feat_dropout(h) # GCN for conv in self.layers: h = conv(g, h) # output h_out = self.MLP_layer(h) return h_out def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss class GCNNet_pyg(nn.Module): def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) hidden_dim = net_params['hidden_dim'] out_dim = net_params['out_dim'] n_classes = net_params['n_classes'] in_feat_dropout = net_params['in_feat_dropout'] dropout = net_params['dropout'] self.n_layers = net_params['L'] self.readout = net_params['readout'] self.batch_norm = net_params['batch_norm'] self.residual = net_params['residual'] self.n_classes = n_classes self.device = net_params['device'] self.embedding_h = nn.Linear(in_dim_node, hidden_dim) # node feat is an integer self.in_feat_dropout = nn.Dropout(in_feat_dropout) self.dropout = dropout self.layers = nn.ModuleList([GCNConv(hidden_dim, hidden_dim, improved = False) for _ in range(self.n_layers)]) if self.batch_norm: self.normlayers = nn.ModuleList([nn.BatchNorm1d(hidden_dim) for _ in range(self.n_layers)]) # self.layers = nn.ModuleList([GCNLayer(hidden_dim, hidden_dim, F.relu, dropout, # self.batch_norm, self.residual) for _ in range(n_layers - 1)]) # self.layers.append(GCNLayer(hidden_dim, out_dim, F.relu, dropout, self.batch_norm, self.residual)) self.MLP_layer = MLPReadout(out_dim, n_classes) # self.MLP_layer = nn.Linear(hidden_dim, n_classes, bias=True) def forward(self, h, edge_index, e): # input embedding h = self.embedding_h(h) h = self.in_feat_dropout(h) # GCN for i in range(self.n_layers): h_in = h h = self.layers[i](h, edge_index) if self.batch_norm: h = self.normlayers[i](h) # batch normalization h = F.relu(h) # non-linear activation if self.residual: h = h_in + h # residual connection h = F.dropout(h, self.dropout, training=self.training) # i = 0 # for conv in self.layers: # h_in = h # h = conv(h, e) # if self.batch_norm: # h = self.normlayers[i](h) # batch normalization # i += 1 # h = F.relu(h) # if self.residual: # h = h_in + h # residual connection # h = F.dropout(h, self.dropout, training=self.training) # output h_out = self.MLP_layer(h) return h_out def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss def loss_proteins(self, pred, label): criterion = nn.BCEWithLogitsLoss() loss = criterion(pred, label.to(torch.float)) return loss
5,125
33.635135
110
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/nets/Planetoid_node_classification/gated_gcn_net.py
import torch import torch.nn as nn import torch.nn.functional as F import dgl import numpy as np """ ResGatedGCN: Residual Gated Graph ConvNets An Experimental Study of Neural Networks for Variable Graphs (Xavier Bresson and Thomas Laurent, ICLR 2018) https://arxiv.org/pdf/1711.07553v2.pdf """ from layers.gated_gcn_layer import GatedGCNLayer, ResGatedGCNLayer from layers.mlp_readout_layer import MLPReadout from torch_geometric.nn import GatedGraphConv class GatedGCNNet(nn.Module): def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) in_dim_edge = 1 # edge_dim (feat is a float) hidden_dim = net_params['hidden_dim'] n_classes = net_params['n_classes'] dropout = net_params['dropout'] n_layers = net_params['L'] self.readout = net_params['readout'] self.batch_norm = net_params['batch_norm'] self.residual = net_params['residual'] self.n_classes = n_classes self.device = net_params['device'] self.pos_enc = net_params['pos_enc'] if self.pos_enc: pos_enc_dim = net_params['pos_enc_dim'] self.embedding_pos_enc = nn.Linear(pos_enc_dim, hidden_dim) self.embedding_h = nn.Embedding(in_dim_node, hidden_dim) # node feat is an integer self.embedding_e = nn.Linear(in_dim_edge, hidden_dim) # edge feat is a float self.layers = nn.ModuleList([ GatedGCNLayer(hidden_dim, hidden_dim, dropout, self.batch_norm, self.residual) for _ in range(n_layers) ]) self.MLP_layer = MLPReadout(hidden_dim, n_classes) def forward(self, g, h, e, h_pos_enc=None): # input embedding h = self.embedding_h(h) if self.pos_enc: h_pos_enc = self.embedding_pos_enc(h_pos_enc.float()) h = h + h_pos_enc e = self.embedding_e(e) # res gated convnets for conv in self.layers: h, e = conv(g, h, e) # output h_out = self.MLP_layer(h) return h_out def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss """ ResGatedGCN: Residual Gated Graph ConvNets An Experimental Study of Neural Networks for Variable Graphs (Xavier Bresson and Thomas Laurent, ICLR 2018) https://arxiv.org/pdf/1711.07553v2.pdf """ class ResGatedGCNNet_pyg(nn.Module): def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) in_dim_edge = 1 # edge_dim (feat is a float) num_bond_type = 3 hidden_dim = net_params['hidden_dim'] n_classes = net_params['n_classes'] self.dropout = net_params['dropout'] self.n_layers = net_params['L'] self.readout = net_params['readout'] self.batch_norm = net_params['batch_norm'] self.edge_feat = net_params['edge_feat'] self.residual = net_params['residual'] self.n_classes = n_classes self.device = net_params['device'] self.pos_enc = net_params['pos_enc'] if self.pos_enc: pos_enc_dim = net_params['pos_enc_dim'] self.embedding_pos_enc = nn.Linear(pos_enc_dim, hidden_dim) self.embedding_h = nn.Linear(in_dim_node, hidden_dim) # node feat is an integer if self.edge_feat: self.embedding_e = nn.Embedding(num_bond_type, hidden_dim) else: self.embedding_e = nn.Linear(1, hidden_dim) self.embedding_e = nn.Linear(in_dim_edge, hidden_dim) # edge feat is a float self.layers = nn.ModuleList([ResGatedGCNLayer(hidden_dim, hidden_dim, self.dropout, self.batch_norm, self.residual) for _ in range(self.n_layers)]) # self.layers = nn.ModuleList([GatedGCNLayer(hidden_dim, hidden_dim, dropout, # self.batch_norm, self.residual) for _ in range(n_layers)]) if self.batch_norm: self.normlayers_h = nn.ModuleList([nn.BatchNorm1d(hidden_dim) for _ in range(self.n_layers)]) self.normlayers_e = nn.ModuleList([nn.BatchNorm1d(hidden_dim) for _ in range(self.n_layers)]) self.MLP_layer = MLPReadout(hidden_dim, n_classes) def forward(self, h, edge_index, e, h_pos_enc=None): # input embedding h = self.embedding_h(h) if self.pos_enc: h_pos_enc = self.embedding_pos_enc(h_pos_enc.float()) h = h + h_pos_enc e = self.embedding_e(e) # res gated convnets for i in range(self.n_layers): h_in = h e_in = e h, e = self.layers[i](h, edge_index, e) if self.batch_norm: h = self.normlayers_h[i](h) e = self.normlayers_e[i](e) # batch normalization if self.residual: h = h_in + h # residual connection e = e_in + e h = F.dropout(h, self.dropout, training=self.training) e = F.dropout(e, self.dropout, training=self.training) # output h_out = self.MLP_layer(h) return h_out def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss def loss_proteins(self, pred, label): criterion = nn.BCEWithLogitsLoss() loss = criterion(pred, label.to(torch.float)) return loss """ Gated Graph Sequence Neural Networks An Experimental Study of Neural Networks for Variable Graphs Li Y, Tarlow D, Brockschmidt M, et al. Gated graph sequence neural networks[J]. arXiv preprint arXiv:1511.05493, 2015. https://arxiv.org/abs/1511.05493 Note that the pyg and dgl of the gatedGCN are different models. """ class GatedGCNNet_pyg(nn.Module): def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) in_dim_edge = 1 # edge_dim (feat is a float) hidden_dim = net_params['hidden_dim'] n_classes = net_params['n_classes'] self.dropout = net_params['dropout'] n_layers = net_params['L'] self.readout = net_params['readout'] self.batch_norm = net_params['batch_norm'] self.residual = net_params['residual'] self.n_classes = n_classes self.device = net_params['device'] self.pos_enc = net_params['pos_enc'] if self.pos_enc: pos_enc_dim = net_params['pos_enc_dim'] self.embedding_pos_enc = nn.Linear(pos_enc_dim, hidden_dim) self.embedding_h = nn.Linear(in_dim_node, hidden_dim) # node feat is an integer # self.embedding_e = nn.Linear(in_dim_edge, hidden_dim) # edge feat is a float self.layers = nn.ModuleList([GatedGraphConv(hidden_dim, n_layers, aggr = 'add')]) # self.layers = nn.ModuleList([GatedGCNLayer(hidden_dim, hidden_dim, dropout, # self.batch_norm, self.residual) for _ in range(n_layers)]) if self.batch_norm: self.normlayers = nn.ModuleList([nn.BatchNorm1d(hidden_dim)]) self.MLP_layer = MLPReadout(hidden_dim, n_classes) def forward(self, h, edge_index, e, h_pos_enc=None): # input embedding h = self.embedding_h(h) if self.pos_enc: h_pos_enc = self.embedding_pos_enc(h_pos_enc.float()) h = h + h_pos_enc # e = self.embedding_e(e) # res gated convnets for conv in self.layers: h_in = h h = conv(h, edge_index, e) if self.batch_norm: h = self.normlayers[0](h) if self.residual: h = h_in + h # residual connection h = F.dropout(h, self.dropout, training=self.training) # output h_out = self.MLP_layer(h) return h_out def loss(self, pred, label): # weighted cross-entropy for unbalanced classes criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) # dataset.dataset[0].y.view(-1).size() return loss
8,352
37.493088
122
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/nets/Planetoid_node_classification/mlp_net.py
import torch import torch.nn as nn import torch.nn.functional as F import dgl from layers.mlp_readout_layer import MLPReadout class MLPNet(nn.Module): def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) hidden_dim = net_params['hidden_dim'] n_classes = net_params['n_classes'] in_feat_dropout = net_params['in_feat_dropout'] dropout = net_params['dropout'] n_layers = net_params['L'] self.gated = net_params['gated'] self.n_classes = n_classes self.device = net_params['device'] self.embedding_h = nn.Embedding(in_dim_node, hidden_dim) # node feat is an integer self.in_feat_dropout = nn.Dropout(in_feat_dropout) feat_mlp_modules = [ nn.Linear(hidden_dim, hidden_dim, bias=True), nn.ReLU(), nn.Dropout(dropout), ] for _ in range(n_layers-1): feat_mlp_modules.append(nn.Linear(hidden_dim, hidden_dim, bias=True)) feat_mlp_modules.append(nn.ReLU()) feat_mlp_modules.append(nn.Dropout(dropout)) self.feat_mlp = nn.Sequential(*feat_mlp_modules) if self.gated: self.gates = nn.Linear(hidden_dim, hidden_dim, bias=True) self.readout_mlp = MLPReadout(hidden_dim, n_classes) def forward(self, g, h, e): # input embedding h = self.embedding_h(h) h = self.in_feat_dropout(h) # MLP h = self.feat_mlp(h) if self.gated: h = torch.sigmoid(self.gates(h)) * h # output h_out = self.readout_mlp(h) return h_out def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss class MLPNet_pyg(nn.Module): def __init__(self, net_params): super().__init__() in_dim_node = net_params['in_dim'] # node_dim (feat is an integer) hidden_dim = net_params['hidden_dim'] n_classes = net_params['n_classes'] in_feat_dropout = net_params['in_feat_dropout'] dropout = net_params['dropout'] n_layers = net_params['L'] self.gated = net_params['gated'] self.n_classes = n_classes self.device = net_params['device'] self.embedding_h = nn.Linear(in_dim_node, hidden_dim) # node feat is an integer self.in_feat_dropout = nn.Dropout(in_feat_dropout) feat_mlp_modules = [ nn.Linear(hidden_dim, hidden_dim, bias=True), nn.ReLU(), nn.Dropout(dropout), ] for _ in range(n_layers - 1): feat_mlp_modules.append(nn.Linear(hidden_dim, hidden_dim, bias=True)) feat_mlp_modules.append(nn.ReLU()) feat_mlp_modules.append(nn.Dropout(dropout)) self.feat_mlp = nn.Sequential(*feat_mlp_modules) if self.gated: self.gates = nn.Linear(hidden_dim, hidden_dim, bias=True) self.readout_mlp = MLPReadout(hidden_dim, n_classes) def forward(self, h, edge_index, e): # input embedding h = self.embedding_h(h) h = self.in_feat_dropout(h) # MLP h = self.feat_mlp(h) if self.gated: h = torch.sigmoid(self.gates(h)) * h # output h_out = self.readout_mlp(h) return h_out def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss def loss_proteins(self, pred, label): criterion = nn.BCEWithLogitsLoss() loss = criterion(pred, label.to(torch.float)) return loss
3,772
27.583333
93
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/nets/Planetoid_node_classification/mo_net.py
import torch import torch.nn as nn import torch.nn.functional as F from torch_scatter import scatter_add import dgl import numpy as np """ GMM: Gaussian Mixture Model Convolution layer Geometric Deep Learning on Graphs and Manifolds using Mixture Model CNNs (Federico Monti et al., CVPR 2017) https://arxiv.org/pdf/1611.08402.pdf """ from layers.gmm_layer import GMMLayer from layers.mlp_readout_layer import MLPReadout from torch_geometric.nn import GMMConv class MoNet(nn.Module): def __init__(self, net_params): super().__init__() self.name = 'MoNet' in_dim = net_params['in_dim'] hidden_dim = net_params['hidden_dim'] out_dim = net_params['out_dim'] kernel = net_params['kernel'] # for MoNet dim = net_params['pseudo_dim_MoNet'] # for MoNet n_classes = net_params['n_classes'] dropout = net_params['dropout'] n_layers = net_params['L'] self.readout = net_params['readout'] batch_norm = net_params['batch_norm'] residual = net_params['residual'] self.device = net_params['device'] self.n_classes = n_classes aggr_type = "sum" # default for MoNet self.embedding_h = nn.Embedding(in_dim, hidden_dim) self.layers = nn.ModuleList() self.pseudo_proj = nn.ModuleList() # Hidden layer for _ in range(n_layers-1): self.layers.append(GMMLayer(hidden_dim, hidden_dim, dim, kernel, aggr_type, dropout, batch_norm, residual)) self.pseudo_proj.append(nn.Sequential(nn.Linear(2, dim), nn.Tanh())) # Output layer self.layers.append(GMMLayer(hidden_dim, out_dim, dim, kernel, aggr_type, dropout, batch_norm, residual)) self.pseudo_proj.append(nn.Sequential(nn.Linear(2, dim), nn.Tanh())) self.MLP_layer = MLPReadout(out_dim, n_classes) def forward(self, g, h, e): h = self.embedding_h(h) # computing the 'pseudo' named tensor which depends on node degrees g.ndata['deg'] = g.in_degrees() g.apply_edges(self.compute_pseudo) pseudo = g.edata['pseudo'].to(self.device).float() for i in range(len(self.layers)): h = self.layers[i](g, h, self.pseudo_proj[i](pseudo)) return self.MLP_layer(h) def compute_pseudo(self, edges): # compute pseudo edge features for MoNet # to avoid zero division in case in_degree is 0, we add constant '1' in all node degrees denoting self-loop srcs = 1/np.sqrt(edges.src['deg']+1) dsts = 1/np.sqrt(edges.dst['deg']+1) pseudo = torch.cat((srcs.unsqueeze(-1), dsts.unsqueeze(-1)), dim=1) return {'pseudo': pseudo} def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss """ GMM: Gaussian Mixture Model Convolution layer Geometric Deep Learning on Graphs and Manifolds using Mixture Model CNNs (Federico Monti et al., CVPR 2017) https://arxiv.org/pdf/1611.08402.pdf """ class MoNetNet_pyg(nn.Module): def __init__(self, net_params): super().__init__() self.name = 'MoNet' in_dim_node = net_params['in_dim'] hidden_dim = net_params['hidden_dim'] out_dim = net_params['out_dim'] kernel = net_params['kernel'] # for MoNet dim = net_params['pseudo_dim_MoNet'] # for MoNet n_classes = net_params['n_classes'] self.dropout = net_params['dropout'] self.n_layers = net_params['L'] self.readout = net_params['readout'] self.batch_norm = net_params['batch_norm'] self.residual = net_params['residual'] self.device = net_params['device'] self.n_classes = n_classes self.dim = dim # aggr_type = "sum" # default for MoNet aggr_type = "mean" self.embedding_h = nn.Linear(in_dim_node, hidden_dim) # node feat is an integer # self.embedding_e = nn.Linear(1, dim) # edge feat is a float self.layers = nn.ModuleList() self.pseudo_proj = nn.ModuleList() self.batchnorm_h = nn.ModuleList() # Hidden layer for _ in range(self.n_layers - 1): self.layers.append(GMMConv(hidden_dim, hidden_dim, dim, kernel, separate_gaussians = False ,aggr = aggr_type, root_weight = True, bias = True)) if self.batch_norm: self.batchnorm_h.append(nn.BatchNorm1d(hidden_dim)) self.pseudo_proj.append(nn.Sequential(nn.Linear(2, dim), nn.Tanh())) # Output layer self.layers.append(GMMConv(hidden_dim, out_dim, dim, kernel, separate_gaussians = False ,aggr = aggr_type, root_weight = True, bias = True)) if self.batch_norm: self.batchnorm_h.append(nn.BatchNorm1d(out_dim)) self.pseudo_proj.append(nn.Sequential(nn.Linear(2, dim), nn.Tanh())) self.MLP_layer = MLPReadout(out_dim, n_classes) # to do def forward(self, h, edge_index, e): h = self.embedding_h(h) edge_weight = torch.ones((edge_index.size(1),), device = edge_index.device) row, col = edge_index[0], edge_index[1] deg = scatter_add(edge_weight, row, dim=0, dim_size=h.size(0)) deg_inv_sqrt = deg.pow_(-0.5) deg_inv_sqrt.masked_fill_(deg_inv_sqrt == float('inf'), 0) pseudo = torch.cat((deg_inv_sqrt[row].unsqueeze(-1), deg_inv_sqrt[col].unsqueeze(-1)), dim=1) for i in range(self.n_layers): h_in = h h = self.layers[i](h, edge_index, self.pseudo_proj[i](pseudo)) if self.batch_norm: h = self.batchnorm_h[i](h) # batch normalization h = F.relu(h) # non-linear activation if self.residual: h = h_in + h # residual connection h = F.dropout(h, self.dropout, training=self.training) return self.MLP_layer(h) def loss(self, pred, label): criterion = nn.CrossEntropyLoss() loss = criterion(pred, label) return loss
6,420
39.639241
121
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/graphsage_layer.py
import torch import torch.nn as nn import torch.nn.functional as F import dgl.function as fn from dgl.nn.pytorch import SAGEConv """ GraphSAGE: William L. Hamilton, Rex Ying, Jure Leskovec, Inductive Representation Learning on Large Graphs (NeurIPS 2017) https://cs.stanford.edu/people/jure/pubs/graphsage-nips17.pdf """ class GraphSageLayer(nn.Module): def __init__(self, in_feats, out_feats, activation, dropout, aggregator_type, batch_norm, residual=False, bias=True, dgl_builtin=False): super().__init__() self.in_channels = in_feats self.out_channels = out_feats self.aggregator_type = aggregator_type self.batch_norm = batch_norm self.residual = residual self.dgl_builtin = dgl_builtin if in_feats != out_feats: self.residual = False self.dropout = nn.Dropout(p=dropout) if dgl_builtin == False: self.nodeapply = NodeApply(in_feats, out_feats, activation, dropout, bias=bias) if aggregator_type == "maxpool": self.aggregator = MaxPoolAggregator(in_feats, in_feats, activation, bias) elif aggregator_type == "lstm": self.aggregator = LSTMAggregator(in_feats, in_feats) else: self.aggregator = MeanAggregator() else: self.sageconv = SAGEConv(in_feats, out_feats, aggregator_type, dropout, activation=activation) if self.batch_norm: self.batchnorm_h = nn.BatchNorm1d(out_feats) def forward(self, g, h): h_in = h # for residual connection if self.dgl_builtin == False: h = self.dropout(h) g.ndata['h'] = h g.update_all(fn.copy_src(src='h', out='m'), self.aggregator, self.nodeapply) h = g.ndata['h'] else: h = self.sageconv(g, h) if self.batch_norm: h = self.batchnorm_h(h) if self.residual: h = h_in + h # residual connection return h def __repr__(self): return '{}(in_channels={}, out_channels={}, aggregator={}, residual={})'.format(self.__class__.__name__, self.in_channels, self.out_channels, self.aggregator_type, self.residual) """ Aggregators for GraphSage """ class Aggregator(nn.Module): """ Base Aggregator class. """ def __init__(self): super().__init__() def forward(self, node): neighbour = node.mailbox['m'] c = self.aggre(neighbour) return {"c": c} def aggre(self, neighbour): # N x F raise NotImplementedError class MeanAggregator(Aggregator): """ Mean Aggregator for graphsage """ def __init__(self): super().__init__() def aggre(self, neighbour): mean_neighbour = torch.mean(neighbour, dim=1) return mean_neighbour class MaxPoolAggregator(Aggregator): """ Maxpooling aggregator for graphsage """ def __init__(self, in_feats, out_feats, activation, bias): super().__init__() self.linear = nn.Linear(in_feats, out_feats, bias=bias) self.activation = activation def aggre(self, neighbour): neighbour = self.linear(neighbour) if self.activation: neighbour = self.activation(neighbour) maxpool_neighbour = torch.max(neighbour, dim=1)[0] return maxpool_neighbour class LSTMAggregator(Aggregator): """ LSTM aggregator for graphsage """ def __init__(self, in_feats, hidden_feats): super().__init__() self.lstm = nn.LSTM(in_feats, hidden_feats, batch_first=True) self.hidden_dim = hidden_feats self.hidden = self.init_hidden() nn.init.xavier_uniform_(self.lstm.weight, gain=nn.init.calculate_gain('relu')) def init_hidden(self): """ Defaulted to initialite all zero """ return (torch.zeros(1, 1, self.hidden_dim), torch.zeros(1, 1, self.hidden_dim)) def aggre(self, neighbours): """ aggregation function """ # N X F rand_order = torch.randperm(neighbours.size()[1]) neighbours = neighbours[:, rand_order, :] (lstm_out, self.hidden) = self.lstm(neighbours.view(neighbours.size()[0], neighbours.size()[1], -1)) return lstm_out[:, -1, :] def forward(self, node): neighbour = node.mailbox['m'] c = self.aggre(neighbour) return {"c": c} class NodeApply(nn.Module): """ Works -> the node_apply function in DGL paradigm """ def __init__(self, in_feats, out_feats, activation, dropout, bias=True): super().__init__() self.dropout = nn.Dropout(p=dropout) self.linear = nn.Linear(in_feats * 2, out_feats, bias) self.activation = activation def concat(self, h, aggre_result): bundle = torch.cat((h, aggre_result), 1) bundle = self.linear(bundle) return bundle def forward(self, node): h = node.data['h'] c = node.data['c'] bundle = self.concat(h, c) bundle = F.normalize(bundle, p=2, dim=1) if self.activation: bundle = self.activation(bundle) return {"h": bundle} ############################################################## # # Additional layers for edge feature/representation analysis # ############################################################## class GraphSageLayerEdgeFeat(nn.Module): def __init__(self, in_feats, out_feats, activation, dropout, aggregator_type, batch_norm, residual=False, bias=True, dgl_builtin=False): super().__init__() self.in_channels = in_feats self.out_channels = out_feats self.batch_norm = batch_norm self.residual = residual if in_feats != out_feats: self.residual = False self.dropout = nn.Dropout(p=dropout) self.activation = activation self.A = nn.Linear(in_feats, out_feats, bias=bias) self.B = nn.Linear(in_feats, out_feats, bias=bias) self.nodeapply = NodeApply(in_feats, out_feats, activation, dropout, bias=bias) if self.batch_norm: self.batchnorm_h = nn.BatchNorm1d(out_feats) def message_func(self, edges): Ah_j = edges.src['Ah'] e_ij = edges.src['Bh'] + edges.dst['Bh'] # e_ij = Bhi + Bhj edges.data['e'] = e_ij return {'Ah_j' : Ah_j, 'e_ij' : e_ij} def reduce_func(self, nodes): # Anisotropic MaxPool aggregation Ah_j = nodes.mailbox['Ah_j'] e = nodes.mailbox['e_ij'] sigma_ij = torch.sigmoid(e) # sigma_ij = sigmoid(e_ij) Ah_j = sigma_ij * Ah_j if self.activation: Ah_j = self.activation(Ah_j) c = torch.max(Ah_j, dim=1)[0] return {'c' : c} def forward(self, g, h): h_in = h # for residual connection h = self.dropout(h) g.ndata['h'] = h g.ndata['Ah'] = self.A(h) g.ndata['Bh'] = self.B(h) g.update_all(self.message_func, self.reduce_func, self.nodeapply) h = g.ndata['h'] if self.batch_norm: h = self.batchnorm_h(h) if self.residual: h = h_in + h # residual connection return h def __repr__(self): return '{}(in_channels={}, out_channels={}, residual={})'.format( self.__class__.__name__, self.in_channels, self.out_channels, self.residual) ############################################################## class GraphSageLayerEdgeReprFeat(nn.Module): def __init__(self, in_feats, out_feats, activation, dropout, aggregator_type, batch_norm, residual=False, bias=True, dgl_builtin=False): super().__init__() self.in_channels = in_feats self.out_channels = out_feats self.batch_norm = batch_norm self.residual = residual if in_feats != out_feats: self.residual = False self.dropout = nn.Dropout(p=dropout) self.activation = activation self.A = nn.Linear(in_feats, out_feats, bias=bias) self.B = nn.Linear(in_feats, out_feats, bias=bias) self.C = nn.Linear(in_feats, out_feats, bias=bias) self.nodeapply = NodeApply(in_feats, out_feats, activation, dropout, bias=bias) if self.batch_norm: self.batchnorm_h = nn.BatchNorm1d(out_feats) self.batchnorm_e = nn.BatchNorm1d(out_feats) def message_func(self, edges): Ah_j = edges.src['Ah'] e_ij = edges.data['Ce'] + edges.src['Bh'] + edges.dst['Bh'] # e_ij = Ce_ij + Bhi + Bhj edges.data['e'] = e_ij return {'Ah_j' : Ah_j, 'e_ij' : e_ij} def reduce_func(self, nodes): # Anisotropic MaxPool aggregation Ah_j = nodes.mailbox['Ah_j'] e = nodes.mailbox['e_ij'] sigma_ij = torch.sigmoid(e) # sigma_ij = sigmoid(e_ij) Ah_j = sigma_ij * Ah_j if self.activation: Ah_j = self.activation(Ah_j) c = torch.max(Ah_j, dim=1)[0] return {'c' : c} def forward(self, g, h, e): h_in = h # for residual connection e_in = e h = self.dropout(h) g.ndata['h'] = h g.ndata['Ah'] = self.A(h) g.ndata['Bh'] = self.B(h) g.edata['e'] = e g.edata['Ce'] = self.C(e) g.update_all(self.message_func, self.reduce_func, self.nodeapply) h = g.ndata['h'] e = g.edata['e'] if self.activation: e = self.activation(e) # non-linear activation if self.batch_norm: h = self.batchnorm_h(h) e = self.batchnorm_e(e) if self.residual: h = h_in + h # residual connection e = e_in + e # residual connection return h, e def __repr__(self): return '{}(in_channels={}, out_channels={}, residual={})'.format( self.__class__.__name__, self.in_channels, self.out_channels, self.residual)
10,938
29.386111
114
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/mlp_readout_layer.py
import torch import torch.nn as nn import torch.nn.functional as F """ MLP Layer used after graph vector representation """ class MLPReadout(nn.Module): def __init__(self, input_dim, output_dim, L=2): #L=nb_hidden_layers super().__init__() list_FC_layers = [ nn.Linear( input_dim//2**l , input_dim//2**(l+1) , bias=True ) for l in range(L) ] list_FC_layers.append(nn.Linear( input_dim//2**L , output_dim , bias=True )) self.FC_layers = nn.ModuleList(list_FC_layers) self.L = L def forward(self, x): y = x for l in range(self.L): y = self.FC_layers[l](y) y = F.relu(y) y = self.FC_layers[self.L](y) return y # class MLPReadout(nn.Module): # # def __init__(self, input_dim, output_dim): # L=nb_hidden_layers # super().__init__() # FC_layers = nn.Linear(input_dim, output_dim, bias=True) # # # def forward(self, x): # y = x # y = self.FC_layers(y) # return y
1,026
26.756757
109
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/gated_gcn_layer.py
import torch import torch.nn as nn import torch.nn.functional as F from torch import Tensor from torch_geometric.typing import OptTensor from torch_scatter import scatter from torch_geometric.nn.conv import MessagePassing """ ResGatedGCN: Residual Gated Graph ConvNets An Experimental Study of Neural Networks for Variable Graphs (Xavier Bresson and Thomas Laurent, ICLR 2018) https://arxiv.org/pdf/1711.07553v2.pdf """ class GatedGCNLayer(nn.Module): """ Param: [] """ def __init__(self, input_dim, output_dim, dropout, batch_norm, residual=False): super().__init__() self.in_channels = input_dim self.out_channels = output_dim self.dropout = dropout self.batch_norm = batch_norm self.residual = residual if input_dim != output_dim: self.residual = False self.A = nn.Linear(input_dim, output_dim, bias=True) self.B = nn.Linear(input_dim, output_dim, bias=True) self.C = nn.Linear(input_dim, output_dim, bias=True) self.D = nn.Linear(input_dim, output_dim, bias=True) self.E = nn.Linear(input_dim, output_dim, bias=True) self.bn_node_h = nn.BatchNorm1d(output_dim) self.bn_node_e = nn.BatchNorm1d(output_dim) def message_func(self, edges): Bh_j = edges.src['Bh'] e_ij = edges.data['Ce'] + edges.src['Dh'] + edges.dst['Eh'] # e_ij = Ce_ij + Dhi + Ehj edges.data['e'] = e_ij return {'Bh_j' : Bh_j, 'e_ij' : e_ij} def reduce_func(self, nodes): Ah_i = nodes.data['Ah'] Bh_j = nodes.mailbox['Bh_j'] e = nodes.mailbox['e_ij'] sigma_ij = torch.sigmoid(e) # sigma_ij = sigmoid(e_ij) #h = Ah_i + torch.mean( sigma_ij * Bh_j, dim=1 ) # hi = Ahi + mean_j alpha_ij * Bhj h = Ah_i + torch.sum( sigma_ij * Bh_j, dim=1 ) / ( torch.sum( sigma_ij, dim=1 ) + 1e-6 ) # hi = Ahi + sum_j eta_ij/sum_j' eta_ij' * Bhj <= dense attention return {'h' : h} def forward(self, g, h, e): h_in = h # for residual connection e_in = e # for residual connection g.ndata['h'] = h g.ndata['Ah'] = self.A(h) g.ndata['Bh'] = self.B(h) g.ndata['Dh'] = self.D(h) g.ndata['Eh'] = self.E(h) g.edata['e'] = e g.edata['Ce'] = self.C(e) g.update_all(self.message_func,self.reduce_func) h = g.ndata['h'] # result of graph convolution e = g.edata['e'] # result of graph convolution if self.batch_norm: h = self.bn_node_h(h) # batch normalization e = self.bn_node_e(e) # batch normalization h = F.relu(h) # non-linear activation e = F.relu(e) # non-linear activation if self.residual: h = h_in + h # residual connection e = e_in + e # residual connection h = F.dropout(h, self.dropout, training=self.training) e = F.dropout(e, self.dropout, training=self.training) return h, e def __repr__(self): return '{}(in_channels={}, out_channels={})'.format(self.__class__.__name__, self.in_channels, self.out_channels) """ ResGatedGCN: Residual Gated Graph ConvNets for pyg implement, is made by myself An Experimental Study of Neural Networks for Variable Graphs (Xavier Bresson and Thomas Laurent, ICLR 2018) https://arxiv.org/pdf/1711.07553v2.pdf """ class ResGatedGCNLayer(MessagePassing): """ Param: [] """ def __init__(self, input_dim, output_dim, dropout, batch_norm, residual=False): super().__init__() self.in_channels = input_dim self.out_channels = output_dim self.dropout = dropout self.batch_norm = batch_norm self.residual = residual if input_dim != output_dim: self.residual = False self.A = nn.Linear(input_dim, output_dim, bias=True) self.B = nn.Linear(input_dim, output_dim, bias=True) self.C = nn.Linear(input_dim, output_dim, bias=True) self.D = nn.Linear(input_dim, output_dim, bias=True) self.E = nn.Linear(input_dim, output_dim, bias=True) def message(self, x_j: Tensor, alpha_j: Tensor, alpha_i: Tensor, Ah: Tensor ,edge_weight: OptTensor): e_ij = edge_weight + alpha_j + alpha_i # e_ij = edges.data['Ce'] + edges.src['Dh'] + edges.dst['Eh'] # e_ij = Ce_ij + Dhi + Ehj return [x_j, e_ij, Ah] def aggregate(self, inputs, index, ptr=None, dim_size=None): Ah_i = inputs[2] Bh_j = inputs[0] sigma_ij = torch.sigmoid(inputs[1]) e = inputs[1] # aa=scatter(sigma_ij * Bh_j, index, dim=self.node_dim, dim_size=dim_size, # reduce='add') h = Ah_i + scatter(sigma_ij*Bh_j, index, dim= self.node_dim, dim_size=dim_size, reduce='add') / (scatter(sigma_ij, index, dim=self.node_dim, dim_size=dim_size, reduce='sum') + 1e-6) return [h, e] # hi = Ahi + sum_j eta_ij/sum_j' eta_ij' * Bhj <= dense attention def forward(self, h, edge_index, edge_weight): # h = conv(h, edge_index, e)g, h, e h_in = h # for residual connection e_in = edge_weight # for residual connection Ah = self.A(h) Bh = self.B(h) Dh = self.D(h) Eh = self.E(h) Ce = self.C(edge_weight) # g.update_all(self.message_func, self.reduce_func) m = self.propagate(edge_index, x=(Bh,Bh), alpha=(Dh,Eh), Ah=Ah, edge_weight=Ce, size=None) h = m[0] # result of graph convolution e = m[1] # result of graph convolution return h, e def __repr__(self): return '{}(in_channels={}, out_channels={})'.format(self.__class__.__name__, self.in_channels, self.out_channels) ############################################################## # # Additional layers for edge feature/representation analysis # ############################################################## class GatedGCNLayerEdgeFeatOnly(nn.Module): """ Param: [] """ def __init__(self, input_dim, output_dim, dropout, batch_norm, residual=False): super().__init__() self.in_channels = input_dim self.out_channels = output_dim self.dropout = dropout self.batch_norm = batch_norm self.residual = residual if input_dim != output_dim: self.residual = False self.A = nn.Linear(input_dim, output_dim, bias=True) self.B = nn.Linear(input_dim, output_dim, bias=True) self.D = nn.Linear(input_dim, output_dim, bias=True) self.E = nn.Linear(input_dim, output_dim, bias=True) self.bn_node_h = nn.BatchNorm1d(output_dim) def message_func(self, edges): Bh_j = edges.src['Bh'] e_ij = edges.src['Dh'] + edges.dst['Eh'] # e_ij = Dhi + Ehj edges.data['e'] = e_ij return {'Bh_j' : Bh_j, 'e_ij' : e_ij} def reduce_func(self, nodes): Ah_i = nodes.data['Ah'] Bh_j = nodes.mailbox['Bh_j'] e = nodes.mailbox['e_ij'] sigma_ij = torch.sigmoid(e) # sigma_ij = sigmoid(e_ij) h = Ah_i + torch.sum( sigma_ij * Bh_j, dim=1 ) / ( torch.sum( sigma_ij, dim=1 ) + 1e-6 ) # hi = Ahi + sum_j eta_ij/sum_j' eta_ij' * Bhj <= dense attention return {'h' : h} def forward(self, g, h, e): h_in = h # for residual connection g.ndata['h'] = h g.ndata['Ah'] = self.A(h) g.ndata['Bh'] = self.B(h) g.ndata['Dh'] = self.D(h) g.ndata['Eh'] = self.E(h) g.update_all(self.message_func,self.reduce_func) h = g.ndata['h'] # result of graph convolution if self.batch_norm: h = self.bn_node_h(h) # batch normalization h = F.relu(h) # non-linear activation if self.residual: h = h_in + h # residual connection h = F.dropout(h, self.dropout, training=self.training) return h, e def __repr__(self): return '{}(in_channels={}, out_channels={})'.format(self.__class__.__name__, self.in_channels, self.out_channels) ############################################################## class GatedGCNLayerIsotropic(nn.Module): """ Param: [] """ def __init__(self, input_dim, output_dim, dropout, batch_norm, residual=False): super().__init__() self.in_channels = input_dim self.out_channels = output_dim self.dropout = dropout self.batch_norm = batch_norm self.residual = residual if input_dim != output_dim: self.residual = False self.A = nn.Linear(input_dim, output_dim, bias=True) self.B = nn.Linear(input_dim, output_dim, bias=True) self.bn_node_h = nn.BatchNorm1d(output_dim) def message_func(self, edges): Bh_j = edges.src['Bh'] return {'Bh_j' : Bh_j} def reduce_func(self, nodes): Ah_i = nodes.data['Ah'] Bh_j = nodes.mailbox['Bh_j'] h = Ah_i + torch.sum( Bh_j, dim=1 ) # hi = Ahi + sum_j Bhj return {'h' : h} def forward(self, g, h, e): h_in = h # for residual connection g.ndata['h'] = h g.ndata['Ah'] = self.A(h) g.ndata['Bh'] = self.B(h) g.update_all(self.message_func,self.reduce_func) h = g.ndata['h'] # result of graph convolution if self.batch_norm: h = self.bn_node_h(h) # batch normalization h = F.relu(h) # non-linear activation if self.residual: h = h_in + h # residual connection h = F.dropout(h, self.dropout, training=self.training) return h, e def __repr__(self): return '{}(in_channels={}, out_channels={})'.format(self.__class__.__name__, self.in_channels, self.out_channels)
10,484
35.40625
170
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/gat_layer.py
import torch import torch.nn as nn import torch.nn.functional as F from dgl.nn.pytorch import GATConv """ GAT: Graph Attention Network Graph Attention Networks (Veličković et al., ICLR 2018) https://arxiv.org/abs/1710.10903 """ class GATLayer(nn.Module): """ Parameters ---------- in_dim : Number of input features. out_dim : Number of output features. num_heads : int Number of heads in Multi-Head Attention. dropout : Required for dropout of attn and feat in GATConv batch_norm : boolean flag for batch_norm layer. residual : If True, use residual connection inside this layer. Default: ``False``. activation : callable activation function/layer or None, optional. If not None, applies an activation function to the updated node features. Using dgl builtin GATConv by default: https://github.com/graphdeeplearning/benchmarking-gnns/commit/206e888ecc0f8d941c54e061d5dffcc7ae2142fc """ def __init__(self, in_dim, out_dim, num_heads, dropout, batch_norm, residual=False, activation=F.elu): super().__init__() self.residual = residual self.activation = activation self.batch_norm = batch_norm if in_dim != (out_dim*num_heads): self.residual = False self.gatconv = GATConv(in_dim, out_dim, num_heads, dropout, dropout) if self.batch_norm: self.batchnorm_h = nn.BatchNorm1d(out_dim * num_heads) def forward(self, g, h): h_in = h # for residual connection h = self.gatconv(g, h).flatten(1) if self.batch_norm: h = self.batchnorm_h(h) if self.activation: h = self.activation(h) if self.residual: h = h_in + h # residual connection return h ############################################################## # # Additional layers for edge feature/representation analysis # ############################################################## class CustomGATHeadLayer(nn.Module): def __init__(self, in_dim, out_dim, dropout, batch_norm): super().__init__() self.dropout = dropout self.batch_norm = batch_norm self.fc = nn.Linear(in_dim, out_dim, bias=False) self.attn_fc = nn.Linear(2 * out_dim, 1, bias=False) self.batchnorm_h = nn.BatchNorm1d(out_dim) def edge_attention(self, edges): z2 = torch.cat([edges.src['z'], edges.dst['z']], dim=1) a = self.attn_fc(z2) return {'e': F.leaky_relu(a)} def message_func(self, edges): return {'z': edges.src['z'], 'e': edges.data['e']} def reduce_func(self, nodes): alpha = F.softmax(nodes.mailbox['e'], dim=1) alpha = F.dropout(alpha, self.dropout, training=self.training) h = torch.sum(alpha * nodes.mailbox['z'], dim=1) return {'h': h} def forward(self, g, h): z = self.fc(h) g.ndata['z'] = z g.apply_edges(self.edge_attention) g.update_all(self.message_func, self.reduce_func) h = g.ndata['h'] if self.batch_norm: h = self.batchnorm_h(h) h = F.elu(h) h = F.dropout(h, self.dropout, training=self.training) return h class CustomGATLayer(nn.Module): """ Param: [in_dim, out_dim, n_heads] """ def __init__(self, in_dim, out_dim, num_heads, dropout, batch_norm, residual=True): super().__init__() self.in_channels = in_dim self.out_channels = out_dim self.num_heads = num_heads self.residual = residual if in_dim != (out_dim*num_heads): self.residual = False self.heads = nn.ModuleList() for i in range(num_heads): self.heads.append(CustomGATHeadLayer(in_dim, out_dim, dropout, batch_norm)) self.merge = 'cat' def forward(self, g, h, e): h_in = h # for residual connection head_outs = [attn_head(g, h) for attn_head in self.heads] if self.merge == 'cat': h = torch.cat(head_outs, dim=1) else: h = torch.mean(torch.stack(head_outs)) if self.residual: h = h_in + h # residual connection return h, e def __repr__(self): return '{}(in_channels={}, out_channels={}, heads={}, residual={})'.format(self.__class__.__name__, self.in_channels, self.out_channels, self.num_heads, self.residual) ############################################################## class CustomGATHeadLayerEdgeReprFeat(nn.Module): def __init__(self, in_dim, out_dim, dropout, batch_norm): super().__init__() self.dropout = dropout self.batch_norm = batch_norm self.fc_h = nn.Linear(in_dim, out_dim, bias=False) self.fc_e = nn.Linear(in_dim, out_dim, bias=False) self.fc_proj = nn.Linear(3* out_dim, out_dim) self.attn_fc = nn.Linear(3* out_dim, 1, bias=False) self.batchnorm_h = nn.BatchNorm1d(out_dim) self.batchnorm_e = nn.BatchNorm1d(out_dim) def edge_attention(self, edges): z = torch.cat([edges.data['z_e'], edges.src['z_h'], edges.dst['z_h']], dim=1) e_proj = self.fc_proj(z) attn = F.leaky_relu(self.attn_fc(z)) return {'attn': attn, 'e_proj': e_proj} def message_func(self, edges): return {'z': edges.src['z_h'], 'attn': edges.data['attn']} def reduce_func(self, nodes): alpha = F.softmax(nodes.mailbox['attn'], dim=1) h = torch.sum(alpha * nodes.mailbox['z'], dim=1) return {'h': h} def forward(self, g, h, e): z_h = self.fc_h(h) z_e = self.fc_e(e) g.ndata['z_h'] = z_h g.edata['z_e'] = z_e g.apply_edges(self.edge_attention) g.update_all(self.message_func, self.reduce_func) h = g.ndata['h'] e = g.edata['e_proj'] if self.batch_norm: h = self.batchnorm_h(h) e = self.batchnorm_e(e) h = F.elu(h) e = F.elu(e) h = F.dropout(h, self.dropout, training=self.training) e = F.dropout(e, self.dropout, training=self.training) return h, e class CustomGATLayerEdgeReprFeat(nn.Module): """ Param: [in_dim, out_dim, n_heads] """ def __init__(self, in_dim, out_dim, num_heads, dropout, batch_norm, residual=True): super().__init__() self.in_channels = in_dim self.out_channels = out_dim self.num_heads = num_heads self.residual = residual if in_dim != (out_dim*num_heads): self.residual = False self.heads = nn.ModuleList() for i in range(num_heads): self.heads.append(CustomGATHeadLayerEdgeReprFeat(in_dim, out_dim, dropout, batch_norm)) self.merge = 'cat' def forward(self, g, h, e): h_in = h # for residual connection e_in = e head_outs_h = [] head_outs_e = [] for attn_head in self.heads: h_temp, e_temp = attn_head(g, h, e) head_outs_h.append(h_temp) head_outs_e.append(e_temp) if self.merge == 'cat': h = torch.cat(head_outs_h, dim=1) e = torch.cat(head_outs_e, dim=1) else: raise NotImplementedError if self.residual: h = h_in + h # residual connection e = e_in + e return h, e def __repr__(self): return '{}(in_channels={}, out_channels={}, heads={}, residual={})'.format(self.__class__.__name__, self.in_channels, self.out_channels, self.num_heads, self.residual) ############################################################## class CustomGATHeadLayerIsotropic(nn.Module): def __init__(self, in_dim, out_dim, dropout, batch_norm): super().__init__() self.dropout = dropout self.batch_norm = batch_norm self.fc = nn.Linear(in_dim, out_dim, bias=False) self.batchnorm_h = nn.BatchNorm1d(out_dim) def message_func(self, edges): return {'z': edges.src['z']} def reduce_func(self, nodes): h = torch.sum(nodes.mailbox['z'], dim=1) return {'h': h} def forward(self, g, h): z = self.fc(h) g.ndata['z'] = z g.update_all(self.message_func, self.reduce_func) h = g.ndata['h'] if self.batch_norm: h = self.batchnorm_h(h) h = F.elu(h) h = F.dropout(h, self.dropout, training=self.training) return h class CustomGATLayerIsotropic(nn.Module): """ Param: [in_dim, out_dim, n_heads] """ def __init__(self, in_dim, out_dim, num_heads, dropout, batch_norm, residual=True): super().__init__() self.in_channels = in_dim self.out_channels = out_dim self.num_heads = num_heads self.residual = residual if in_dim != (out_dim*num_heads): self.residual = False self.heads = nn.ModuleList() for i in range(num_heads): self.heads.append(CustomGATHeadLayerIsotropic(in_dim, out_dim, dropout, batch_norm)) self.merge = 'cat' def forward(self, g, h, e): h_in = h # for residual connection head_outs = [attn_head(g, h) for attn_head in self.heads] if self.merge == 'cat': h = torch.cat(head_outs, dim=1) else: h = torch.mean(torch.stack(head_outs)) if self.residual: h = h_in + h # residual connection return h, e def __repr__(self): return '{}(in_channels={}, out_channels={}, heads={}, residual={})'.format(self.__class__.__name__, self.in_channels, self.out_channels, self.num_heads, self.residual)
10,303
29.850299
107
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/gin_layer.py
import torch import torch.nn as nn import torch.nn.functional as F import dgl.function as fn """ GIN: Graph Isomorphism Networks HOW POWERFUL ARE GRAPH NEURAL NETWORKS? (Keyulu Xu, Weihua Hu, Jure Leskovec and Stefanie Jegelka, ICLR 2019) https://arxiv.org/pdf/1810.00826.pdf """ class GINLayer(nn.Module): """ [!] code adapted from dgl implementation of GINConv Parameters ---------- apply_func : callable activation function/layer or None If not None, apply this function to the updated node feature, the :math:`f_\Theta` in the formula. aggr_type : Aggregator type to use (``sum``, ``max`` or ``mean``). out_dim : Rquired for batch norm layer; should match out_dim of apply_func if not None. dropout : Required for dropout of output features. batch_norm : boolean flag for batch_norm layer. residual : boolean flag for using residual connection. init_eps : optional Initial :math:`\epsilon` value, default: ``0``. learn_eps : bool, optional If True, :math:`\epsilon` will be a learnable parameter. """ def __init__(self, apply_func, aggr_type, dropout, batch_norm, residual=False, init_eps=0, learn_eps=False): super().__init__() self.apply_func = apply_func if aggr_type == 'sum': self._reducer = fn.sum elif aggr_type == 'max': self._reducer = fn.max elif aggr_type == 'mean': self._reducer = fn.mean else: raise KeyError('Aggregator type {} not recognized.'.format(aggr_type)) self.batch_norm = batch_norm self.residual = residual self.dropout = dropout in_dim = apply_func.mlp.input_dim out_dim = apply_func.mlp.output_dim if in_dim != out_dim: self.residual = False # to specify whether eps is trainable or not. if learn_eps: self.eps = torch.nn.Parameter(torch.FloatTensor([init_eps])) else: self.register_buffer('eps', torch.FloatTensor([init_eps])) self.bn_node_h = nn.BatchNorm1d(out_dim) def forward(self, g, h): h_in = h # for residual connection g = g.local_var() g.ndata['h'] = h g.update_all(fn.copy_u('h', 'm'), self._reducer('m', 'neigh')) h = (1 + self.eps) * h + g.ndata['neigh'] if self.apply_func is not None: h = self.apply_func(h) if self.batch_norm: h = self.bn_node_h(h) # batch normalization h = F.relu(h) # non-linear activation if self.residual: h = h_in + h # residual connection h = F.dropout(h, self.dropout, training=self.training) return h class ApplyNodeFunc(nn.Module): """ This class is used in class GINNet Update the node feature hv with MLP """ def __init__(self, mlp): super().__init__() self.mlp = mlp def forward(self, h): h = self.mlp(h) return h class MLP(nn.Module): """MLP with linear output""" def __init__(self, num_layers, input_dim, hidden_dim, output_dim): super().__init__() self.linear_or_not = True # default is linear model self.num_layers = num_layers self.output_dim = output_dim self.input_dim = input_dim if num_layers < 1: raise ValueError("number of layers should be positive!") elif num_layers == 1: # Linear model self.linear = nn.Linear(input_dim, output_dim) else: # Multi-layer model self.linear_or_not = False self.linears = torch.nn.ModuleList() self.batch_norms = torch.nn.ModuleList() self.linears.append(nn.Linear(input_dim, hidden_dim)) for layer in range(num_layers - 2): self.linears.append(nn.Linear(hidden_dim, hidden_dim)) self.linears.append(nn.Linear(hidden_dim, output_dim)) for layer in range(num_layers - 1): self.batch_norms.append(nn.BatchNorm1d((hidden_dim))) def forward(self, x): if self.linear_or_not: # If linear model return self.linear(x) else: # If MLP h = x for i in range(self.num_layers - 1): h = F.relu(self.batch_norms[i](self.linears[i](h))) return self.linears[-1](h)
4,598
30.9375
113
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/gmm_layer.py
import torch import torch.nn as nn import torch.nn.functional as F from torch.nn import init import dgl.function as fn """ GMM: Gaussian Mixture Model Convolution layer Geometric Deep Learning on Graphs and Manifolds using Mixture Model CNNs (Federico Monti et al., CVPR 2017) https://arxiv.org/pdf/1611.08402.pdf """ class GMMLayer(nn.Module): """ [!] code adapted from dgl implementation of GMMConv Parameters ---------- in_dim : Number of input features. out_dim : Number of output features. dim : Dimensionality of pseudo-coordinte. kernel : Number of kernels :math:`K`. aggr_type : Aggregator type (``sum``, ``mean``, ``max``). dropout : Required for dropout of output features. batch_norm : boolean flag for batch_norm layer. residual : If True, use residual connection inside this layer. Default: ``False``. bias : If True, adds a learnable bias to the output. Default: ``True``. """ def __init__(self, in_dim, out_dim, dim, kernel, aggr_type, dropout, batch_norm, residual=False, bias=True): super().__init__() self.in_dim = in_dim self.out_dim = out_dim self.dim = dim self.kernel = kernel self.batch_norm = batch_norm self.residual = residual self.dropout = dropout if aggr_type == 'sum': self._reducer = fn.sum elif aggr_type == 'mean': self._reducer = fn.mean elif aggr_type == 'max': self._reducer = fn.max else: raise KeyError("Aggregator type {} not recognized.".format(aggr_type)) self.mu = nn.Parameter(torch.Tensor(kernel, dim)) self.inv_sigma = nn.Parameter(torch.Tensor(kernel, dim)) self.fc = nn.Linear(in_dim, kernel * out_dim, bias=False) self.bn_node_h = nn.BatchNorm1d(out_dim) if in_dim != out_dim: self.residual = False if bias: self.bias = nn.Parameter(torch.Tensor(out_dim)) else: self.register_buffer('bias', None) self.reset_parameters() def reset_parameters(self): """Reinitialize learnable parameters.""" gain = init.calculate_gain('relu') init.xavier_normal_(self.fc.weight, gain=gain) init.normal_(self.mu.data, 0, 0.1) init.constant_(self.inv_sigma.data, 1) if self.bias is not None: init.zeros_(self.bias.data) def forward(self, g, h, pseudo): h_in = h # for residual connection g = g.local_var() g.ndata['h'] = self.fc(h).view(-1, self.kernel, self.out_dim) E = g.number_of_edges() # compute gaussian weight gaussian = -0.5 * ((pseudo.view(E, 1, self.dim) - self.mu.view(1, self.kernel, self.dim)) ** 2) gaussian = gaussian * (self.inv_sigma.view(1, self.kernel, self.dim) ** 2) gaussian = torch.exp(gaussian.sum(dim=-1, keepdim=True)) # (E, K, 1) g.edata['w'] = gaussian g.update_all(fn.u_mul_e('h', 'w', 'm'), self._reducer('m', 'h')) h = g.ndata['h'].sum(1) if self.batch_norm: h = self.bn_node_h(h) # batch normalization h = F.relu(h) # non-linear activation if self.residual: h = h_in + h # residual connection if self.bias is not None: h = h + self.bias h = F.dropout(h, self.dropout, training=self.training) return h
3,680
31.289474
111
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/gcn_layer.py
import torch import torch.nn as nn import torch.nn.functional as F import dgl.function as fn from dgl.nn.pytorch import GraphConv """ GCN: Graph Convolutional Networks Thomas N. Kipf, Max Welling, Semi-Supervised Classification with Graph Convolutional Networks (ICLR 2017) http://arxiv.org/abs/1609.02907 """ # Sends a message of node feature h # Equivalent to => return {'m': edges.src['h']} msg = fn.copy_src(src='h', out='m') def reduce(nodes): accum = torch.mean(nodes.mailbox['m'], 1) return {'h': accum} class NodeApplyModule(nn.Module): # Update node feature h_v with (Wh_v+b) def __init__(self, in_dim, out_dim): super().__init__() self.linear = nn.Linear(in_dim, out_dim) def forward(self, node): h = self.linear(node.data['h']) return {'h': h} class GCNLayer(nn.Module): """ Param: [in_dim, out_dim] """ def __init__(self, in_dim, out_dim, activation, dropout, batch_norm, residual=False, dgl_builtin=False): super().__init__() self.in_channels = in_dim self.out_channels = out_dim self.batch_norm = batch_norm self.residual = residual self.dgl_builtin = dgl_builtin if in_dim != out_dim: self.residual = False self.batchnorm_h = nn.BatchNorm1d(out_dim) self.activation = activation self.dropout = nn.Dropout(dropout) if self.dgl_builtin == False: self.apply_mod = NodeApplyModule(in_dim, out_dim) else: self.conv = GraphConv(in_dim, out_dim) def forward(self, g, feature): h_in = feature # to be used for residual connection if self.dgl_builtin == False: g.ndata['h'] = feature g.update_all(msg, reduce) g.apply_nodes(func=self.apply_mod) h = g.ndata['h'] # result of graph convolution else: h = self.conv(g, feature) if self.batch_norm: h = self.batchnorm_h(h) # batch normalization if self.activation: h = self.activation(h) if self.residual: h = h_in + h # residual connection h = self.dropout(h) return h def __repr__(self): return '{}(in_channels={}, out_channels={}, residual={})'.format(self.__class__.__name__, self.in_channels, self.out_channels, self.residual)
2,561
29.86747
109
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/ring_gnn_equiv_layer.py
import torch import torch.nn as nn import torch.nn.functional as F """ Ring-GNN equi 2 to 2 layer file On the equivalence between graph isomorphism testing and function approximation with GNNs (Chen et al, 2019) https://arxiv.org/pdf/1905.12560v1.pdf CODE ADPATED FROM https://github.com/leichen2018/Ring-GNN/ """ class RingGNNEquivLayer(nn.Module): def __init__(self, device, input_dim, output_dim, layer_norm, residual, dropout, normalization='inf', normalization_val=1.0, radius=2, k2_init = 0.1): super().__init__() self.device = device basis_dimension = 15 self.radius = radius self.layer_norm = layer_norm self.residual = residual self.dropout = dropout coeffs_values = lambda i, j, k: torch.randn([i, j, k]) * torch.sqrt(2. / (i + j).float()) self.diag_bias_list = nn.ParameterList([]) for i in range(radius): for j in range(i+1): self.diag_bias_list.append(nn.Parameter(torch.zeros(1, output_dim, 1, 1))) self.all_bias = nn.Parameter(torch.zeros(1, output_dim, 1, 1)) self.coeffs_list = nn.ParameterList([]) for i in range(radius): for j in range(i+1): self.coeffs_list.append(nn.Parameter(coeffs_values(input_dim, output_dim, basis_dimension))) self.switch = nn.ParameterList([nn.Parameter(torch.FloatTensor([1])), nn.Parameter(torch.FloatTensor([k2_init]))]) self.output_dim = output_dim self.normalization = normalization self.normalization_val = normalization_val if self.layer_norm: self.ln_x = LayerNorm(output_dim.item()) if self.residual: self.res_x = nn.Linear(input_dim.item(), output_dim.item()) def forward(self, inputs): m = inputs.size()[3] ops_out = ops_2_to_2(inputs, m, normalization=self.normalization) ops_out = torch.stack(ops_out, dim = 2) output_list = [] for i in range(self.radius): for j in range(i+1): output_i = torch.einsum('dsb,ndbij->nsij', self.coeffs_list[i*(i+1)//2 + j], ops_out) mat_diag_bias = torch.eye(inputs.size()[3]).unsqueeze(0).unsqueeze(0).to(self.device) * self.diag_bias_list[i*(i+1)//2 + j] # mat_diag_bias = torch.eye(inputs.size()[3]).to('cuda:0').unsqueeze(0).unsqueeze(0) * self.diag_bias_list[i*(i+1)//2 + j] if j == 0: output = output_i + mat_diag_bias else: output = torch.einsum('abcd,abde->abce', output_i, output) output_list.append(output) output = 0 for i in range(self.radius): output += output_list[i] * self.switch[i] output = output + self.all_bias if self.layer_norm: # Now, changing shapes from [1xdxnxn] to [nxnxd] for BN output = output.permute(3,2,1,0).squeeze() # output = self.bn_x(output.reshape(m*m, self.output_dim.item())) # batch normalization output = self.ln_x(output) # layer normalization # Returning output back to original shape output = output.reshape(m, m, self.output_dim.item()) output = output.permute(2,1,0).unsqueeze(0) output = F.relu(output) # non-linear activation if self.residual: # Now, changing shapes from [1xdxnxn] to [nxnxd] for Linear() layer inputs, output = inputs.permute(3,2,1,0).squeeze(), output.permute(3,2,1,0).squeeze() residual_ = self.res_x(inputs) output = residual_ + output # residual connection # Returning output back to original shape output = output.permute(2,1,0).unsqueeze(0) output = F.dropout(output, self.dropout, training=self.training) return output def ops_2_to_2(inputs, dim, normalization='inf', normalization_val=1.0): # N x D x m x m # input: N x D x m x m diag_part = torch.diagonal(inputs, dim1 = 2, dim2 = 3) # N x D x m sum_diag_part = torch.sum(diag_part, dim=2, keepdim = True) # N x D x 1 sum_of_rows = torch.sum(inputs, dim=3) # N x D x m sum_of_cols = torch.sum(inputs, dim=2) # N x D x m sum_all = torch.sum(sum_of_rows, dim=2) # N x D # op1 - (1234) - extract diag op1 = torch.diag_embed(diag_part) # N x D x m x m # op2 - (1234) + (12)(34) - place sum of diag on diag op2 = torch.diag_embed(sum_diag_part.repeat(1, 1, dim)) # op3 - (1234) + (123)(4) - place sum of row i on diag ii op3 = torch.diag_embed(sum_of_rows) # op4 - (1234) + (124)(3) - place sum of col i on diag ii op4 = torch.diag_embed(sum_of_cols) # op5 - (1234) + (124)(3) + (123)(4) + (12)(34) + (12)(3)(4) - place sum of all entries on diag op5 = torch.diag_embed(sum_all.unsqueeze(2).repeat(1, 1, dim)) # op6 - (14)(23) + (13)(24) + (24)(1)(3) + (124)(3) + (1234) - place sum of col i on row i op6 = sum_of_cols.unsqueeze(3).repeat(1, 1, 1, dim) # op7 - (14)(23) + (23)(1)(4) + (234)(1) + (123)(4) + (1234) - place sum of row i on row i op7 = sum_of_rows.unsqueeze(3).repeat(1, 1, 1, dim) # op8 - (14)(2)(3) + (134)(2) + (14)(23) + (124)(3) + (1234) - place sum of col i on col i op8 = sum_of_cols.unsqueeze(2).repeat(1, 1, dim, 1) # op9 - (13)(24) + (13)(2)(4) + (134)(2) + (123)(4) + (1234) - place sum of row i on col i op9 = sum_of_rows.unsqueeze(2).repeat(1, 1, dim, 1) # op10 - (1234) + (14)(23) - identity op10 = inputs # op11 - (1234) + (13)(24) - transpose op11 = torch.transpose(inputs, -2, -1) # op12 - (1234) + (234)(1) - place ii element in row i op12 = diag_part.unsqueeze(3).repeat(1, 1, 1, dim) # op13 - (1234) + (134)(2) - place ii element in col i op13 = diag_part.unsqueeze(2).repeat(1, 1, dim, 1) # op14 - (34)(1)(2) + (234)(1) + (134)(2) + (1234) + (12)(34) - place sum of diag in all entries op14 = sum_diag_part.unsqueeze(3).repeat(1, 1, dim, dim) # op15 - sum of all ops - place sum of all entries in all entries op15 = sum_all.unsqueeze(2).unsqueeze(3).repeat(1, 1, dim, dim) #A_2 = torch.einsum('abcd,abde->abce', inputs, inputs) #A_4 = torch.einsum('abcd,abde->abce', A_2, A_2) #op16 = torch.where(A_4>1, torch.ones(A_4.size()), A_4) if normalization is not None: float_dim = float(dim) if normalization is 'inf': op2 = torch.div(op2, float_dim) op3 = torch.div(op3, float_dim) op4 = torch.div(op4, float_dim) op5 = torch.div(op5, float_dim**2) op6 = torch.div(op6, float_dim) op7 = torch.div(op7, float_dim) op8 = torch.div(op8, float_dim) op9 = torch.div(op9, float_dim) op14 = torch.div(op14, float_dim) op15 = torch.div(op15, float_dim**2) #return [op1, op2, op3, op4, op5, op6, op7, op8, op9, op10, op11, op12, op13, op14, op15, op16] ''' l = [op1, op2, op3, op4, op5, op6, op7, op8, op9, op10, op11, op12, op13, op14, op15] for i, ls in enumerate(l): print(i+1) print(torch.sum(ls)) print("$%^&*(*&^%$#$%^&*(*&^%$%^&*(*&^%$%^&*(") ''' return [op1, op2, op3, op4, op5, op6, op7, op8, op9, op10, op11, op12, op13, op14, op15] class LayerNorm(nn.Module): def __init__(self, d): super().__init__() self.a = nn.Parameter(torch.ones(d).unsqueeze(0).unsqueeze(0)) # shape is 1 x 1 x d self.b = nn.Parameter(torch.zeros(d).unsqueeze(0).unsqueeze(0)) # shape is 1 x 1 x d def forward(self, x): # x tensor of the shape n x n x d mean = x.mean(dim=(0,1), keepdim=True) var = x.var(dim=(0,1), keepdim=True, unbiased=False) x = self.a * (x - mean) / torch.sqrt(var + 1e-6) + self.b # shape is n x n x d return x
8,076
39.385
139
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/layers/three_wl_gnn_layers.py
import torch import torch.nn as nn import torch.nn.functional as F """ Layers used for 3WLGNN Provably Powerful Graph Networks (Maron et al., 2019) https://papers.nips.cc/paper/8488-provably-powerful-graph-networks.pdf CODE adapted from https://github.com/hadarser/ProvablyPowerfulGraphNetworks_torch/ """ class RegularBlock(nn.Module): """ Imputs: N x input_depth x m x m Take the input through 2 parallel MLP routes, multiply the result, and add a skip-connection at the end. At the skip-connection, reduce the dimension back to output_depth """ def __init__(self, depth_of_mlp, in_features, out_features, residual=False): super().__init__() self.residual = residual self.mlp1 = MlpBlock(in_features, out_features, depth_of_mlp) self.mlp2 = MlpBlock(in_features, out_features, depth_of_mlp) self.skip = SkipConnection(in_features+out_features, out_features) if self.residual: self.res_x = nn.Linear(in_features, out_features) def forward(self, inputs): mlp1 = self.mlp1(inputs) mlp2 = self.mlp2(inputs) mult = torch.matmul(mlp1, mlp2) out = self.skip(in1=inputs, in2=mult) if self.residual: # Now, changing shapes from [1xdxnxn] to [nxnxd] for Linear() layer inputs, out = inputs.permute(3,2,1,0).squeeze(), out.permute(3,2,1,0).squeeze() residual_ = self.res_x(inputs) out = residual_ + out # residual connection # Returning output back to original shape out = out.permute(2,1,0).unsqueeze(0) return out class MlpBlock(nn.Module): """ Block of MLP layers with activation function after each (1x1 conv layers). """ def __init__(self, in_features, out_features, depth_of_mlp, activation_fn=nn.functional.relu): super().__init__() self.activation = activation_fn self.convs = nn.ModuleList() for i in range(depth_of_mlp): self.convs.append(nn.Conv2d(in_features, out_features, kernel_size=1, padding=0, bias=True)) _init_weights(self.convs[-1]) in_features = out_features def forward(self, inputs): out = inputs for conv_layer in self.convs: out = self.activation(conv_layer(out)) return out class SkipConnection(nn.Module): """ Connects the two given inputs with concatenation :param in1: earlier input tensor of shape N x d1 x m x m :param in2: later input tensor of shape N x d2 x m x m :param in_features: d1+d2 :param out_features: output num of features :return: Tensor of shape N x output_depth x m x m """ def __init__(self, in_features, out_features): super().__init__() self.conv = nn.Conv2d(in_features, out_features, kernel_size=1, padding=0, bias=True) _init_weights(self.conv) def forward(self, in1, in2): # in1: N x d1 x m x m # in2: N x d2 x m x m out = torch.cat((in1, in2), dim=1) out = self.conv(out) return out class FullyConnected(nn.Module): def __init__(self, in_features, out_features, activation_fn=nn.functional.relu): super().__init__() self.fc = nn.Linear(in_features, out_features) _init_weights(self.fc) self.activation = activation_fn def forward(self, input): out = self.fc(input) if self.activation is not None: out = self.activation(out) return out def diag_offdiag_maxpool(input): N = input.shape[-1] max_diag = torch.max(torch.diagonal(input, dim1=-2, dim2=-1), dim=2)[0] # BxS # with torch.no_grad(): max_val = torch.max(max_diag) min_val = torch.max(-1 * input) val = torch.abs(torch.add(max_val, min_val)) min_mat = torch.mul(val, torch.eye(N, device=input.device)).view(1, 1, N, N) max_offdiag = torch.max(torch.max(input - min_mat, dim=3)[0], dim=2)[0] # BxS return torch.cat((max_diag, max_offdiag), dim=1) # output Bx2S def _init_weights(layer): """ Init weights of the layer :param layer: :return: """ nn.init.xavier_uniform_(layer.weight) # nn.init.xavier_normal_(layer.weight) if layer.bias is not None: nn.init.zeros_(layer.bias) class LayerNorm(nn.Module): def __init__(self, d): super().__init__() self.a = nn.Parameter(torch.ones(d).unsqueeze(0).unsqueeze(0)) # shape is 1 x 1 x d self.b = nn.Parameter(torch.zeros(d).unsqueeze(0).unsqueeze(0)) # shape is 1 x 1 x d def forward(self, x): # x tensor of the shape n x n x d mean = x.mean(dim=(0,1), keepdim=True) var = x.var(dim=(0,1), keepdim=True, unbiased=False) x = self.a * (x - mean) / torch.sqrt(var + 1e-6) + self.b # shape is n x n x d return x
4,983
31.154839
108
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/train/train_Planetoid_node_classification.py
""" Utility functions for training one epoch and evaluating one epoch """ import torch import torch.nn as nn import math from train.metrics import accuracy_TU as accuracy """ For GCNs """ def train_epoch_sparse(model, optimizer, device, dataset, train_idx): model.train() epoch_loss = 0 epoch_train_acc = 0 nb_data = 0 gpu_mem = 0 # for iter, (batch_graphs, batch_labels) in enumerate(data_loader): batch_x = dataset.dataset[0].x.to(device) batch_e = dataset.edge_attr.to(device) batch_labels = dataset.dataset[0].y.long().to(device) edge_index = dataset.dataset[0].edge_index.long().to(device) train_idx = train_idx.to(device) optimizer.zero_grad() batch_scores = model.forward(batch_x, edge_index, batch_e)[train_idx] loss = model.loss(batch_scores, batch_labels[train_idx]).to(torch.float) loss.backward() optimizer.step() epoch_loss = loss.detach().item() epoch_train_acc = accuracy(batch_scores, batch_labels[train_idx]) / train_idx.size(0) return epoch_loss, epoch_train_acc, optimizer def evaluate_network_sparse(model, device, dataset, val_idx): model.eval() epoch_test_loss = 0 epoch_test_acc = 0 nb_data = 0 with torch.no_grad(): batch_x = dataset.dataset[0].x.to(device) batch_e = dataset.edge_attr.to(device) batch_labels = dataset.dataset[0].y.long().to(device) edge_index = dataset.dataset[0].edge_index.long().to(device) val_idx = val_idx.to(device) batch_scores = model.forward(batch_x, edge_index, batch_e)[val_idx] loss = model.loss(batch_scores, batch_labels[val_idx]).to(torch.float) epoch_test_loss = loss.detach().item() epoch_test_acc = accuracy(batch_scores, batch_labels[val_idx]) / val_idx.size(0) return epoch_test_loss, epoch_test_acc # """ # For WL-GNNs # """ # def train_epoch_dense(model, optimizer, device, data_loader, epoch, batch_size): # model.train() # epoch_loss = 0 # epoch_train_acc = 0 # nb_data = 0 # gpu_mem = 0 # optimizer.zero_grad() # for iter, (x_with_node_feat, labels) in enumerate(data_loader): # x_with_node_feat = x_with_node_feat.to(device) # labels = labels.to(device) # # scores = model.forward(x_with_node_feat) # loss = model.loss(scores, labels) # loss.backward() # # if not (iter%batch_size): # optimizer.step() # optimizer.zero_grad() # # epoch_loss += loss.detach().item() # epoch_train_acc += accuracy(scores, labels) # nb_data += labels.size(0) # epoch_loss /= (iter + 1) # epoch_train_acc /= nb_data # # return epoch_loss, epoch_train_acc, optimizer # # def evaluate_network_dense(model, device, data_loader, epoch): # model.eval() # epoch_test_loss = 0 # epoch_test_acc = 0 # nb_data = 0 # with torch.no_grad(): # for iter, (x_with_node_feat, labels) in enumerate(data_loader): # x_with_node_feat = x_with_node_feat.to(device) # labels = labels.to(device) # # scores = model.forward(x_with_node_feat) # loss = model.loss(scores, labels) # epoch_test_loss += loss.detach().item() # epoch_test_acc += accuracy(scores, labels) # nb_data += labels.size(0) # epoch_test_loss /= (iter + 1) # epoch_test_acc /= nb_data # # return epoch_test_loss, epoch_test_acc def check_patience(all_losses, best_loss, best_epoch, curr_loss, curr_epoch, counter): if curr_loss < best_loss: counter = 0 best_loss = curr_loss best_epoch = curr_epoch else: counter += 1 return best_loss, best_epoch, counter
3,774
30.991525
89
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/train/metrics.py
import torch import torch.nn as nn import torch.nn.functional as F from sklearn.metrics import confusion_matrix from sklearn.metrics import f1_score import numpy as np def MAE(scores, targets): MAE = F.l1_loss(scores, targets) MAE = MAE.detach().item() return MAE # it is the original one to calculate the value, have found the ogb evaluator use the same way. There we use this one to calculate. def accuracy_TU(scores, targets): scores = scores.detach().argmax(dim=1) acc = (scores==targets).float().sum().item() return acc def accuracy_MNIST_CIFAR(scores, targets): scores = scores.detach().argmax(dim=1) acc = (scores==targets).float().sum().item() return acc def accuracy_CITATION_GRAPH(scores, targets): scores = scores.detach().argmax(dim=1) acc = (scores==targets).float().sum().item() acc = acc / len(targets) return acc # it takes into account the case of each class, is the sum of the accuracy of each class(the right divide the total in each class) then # divided by the total number of classes def accuracy_SBM(scores, targets): S = targets.cpu().numpy() C = np.argmax( torch.nn.Softmax(dim=1)(scores).cpu().detach().numpy() , axis=1 ) CM = confusion_matrix(S,C).astype(np.float32) nb_classes = CM.shape[0] targets = targets.cpu().detach().numpy() nb_non_empty_classes = 0 pr_classes = np.zeros(nb_classes) for r in range(nb_classes): cluster = np.where(targets==r)[0] if cluster.shape[0] != 0: pr_classes[r] = CM[r,r]/ float(cluster.shape[0]) if CM[r,r]>0: nb_non_empty_classes += 1 else: pr_classes[r] = 0.0 acc = 100.* np.sum(pr_classes)/ float(nb_classes) return acc def accuracy_ogb(y_pred, y_true): acc_list = [] # y_true = data.y # y_pred = y_pred.argmax(dim=-1, keepdim=True) # for i in range(y_true.shape[0]): # is_labeled = y_true[:, i] == y_true[:, i] # correct = y_true[is_labeled, i] == y_pred[is_labeled, i] # acc_list.append(float(np.sum(correct)) / len(correct)) y_pred = y_pred.detach().argmax(dim=1) acc = (y_pred == y_true).float().sum().item() return acc def binary_f1_score(scores, targets): """Computes the F1 score using scikit-learn for binary class labels. Returns the F1 score for the positive class, i.e. labelled '1'. """ y_true = targets.cpu().numpy() y_pred = scores.argmax(dim=1).cpu().numpy() return f1_score(y_true, y_pred, average='binary') def accuracy_VOC(scores, targets): scores = scores.detach().argmax(dim=1).cpu() targets = targets.cpu().detach().numpy() acc = f1_score(scores, targets, average='weighted') return acc
2,754
31.797619
135
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/train/train_ogb_node_classification.py
""" Utility functions for training one epoch and evaluating one epoch """ import torch import torch.nn as nn import math import dgl from tqdm import tqdm from train.metrics import accuracy_SBM as accuracy from train.metrics import accuracy_ogb from ogb.nodeproppred import Evaluator """ For GCNs """ def train_epoch(model, optimizer, device, train_loader, epoch=None): model.train() # pbar = tqdm(total=len(train_loader)) # pbar.set_description(f'Training epoch: {epoch:04d}') total_loss = total_examples = 0 for data in train_loader: optimizer.zero_grad() data = data.to(device) try: batch_pos_enc = data.pos_enc.to(device) sign_flip = torch.rand(batch_pos_enc.size(1)).to(device) sign_flip[sign_flip >= 0.5] = 1.0 sign_flip[sign_flip < 0.5] = -1.0 batch_pos_enc = batch_pos_enc * sign_flip.unsqueeze(0) batch_scores = model.forward(data.x, data.edge_index, data.edge_attr,batch_pos_enc) except: batch_scores = model(data.x, data.edge_index, data.edge_attr) loss = model.loss(batch_scores[data.train_mask], data.y.view(-1)[data.train_mask]).to(torch.float) loss.backward() optimizer.step() total_loss += float(loss) * int(data.train_mask.sum()) total_examples += int(data.train_mask.sum()) # pbar.update(1) # # pbar.close() return total_loss / total_examples # model.train() # # # for iter, (batch_graphs, batch_labels) in enumerate(data_loader): # batch_x = dataset.x.to(device) # batch_e = dataset.edge_attr.to(device) # # batch_e = dataset.edge_attr # batch_labels = dataset.y.long().to(device) # edge_index = dataset.edge_index.long().to(device) # train_idx = train_idx.to(device) # # optimizer.zero_grad() # batch_scores = model.forward(batch_x, edge_index, batch_e)[train_idx] # loss = model.loss(batch_scores, batch_labels.view(-1)[train_idx]).to(torch.float) # loss.backward() # optimizer.step() # epoch_loss = loss.detach().item() # # return epoch_loss def train_epoch_arxiv(model, optimizer, device, dataset, train_idx): model.train() # for iter, (batch_graphs, batch_labels) in enumerate(data_loader): batch_x = dataset.x.to(device) batch_e = dataset.edge_attr.to(device) # batch_e = dataset.edge_attr batch_labels = dataset.y.long().to(device) edge_index = dataset.edge_index.long().to(device) train_idx = train_idx.to(device) optimizer.zero_grad() try: batch_pos_enc = dataset.pos_enc.to(device) sign_flip = torch.rand(batch_pos_enc.size(1)).to(device) sign_flip[sign_flip >= 0.5] = 1.0 sign_flip[sign_flip < 0.5] = -1.0 batch_pos_enc = batch_pos_enc * sign_flip.unsqueeze(0) batch_scores = model.forward(batch_x, edge_index, batch_e, batch_pos_enc)[train_idx] except: batch_scores = model.forward(batch_x, edge_index, batch_e)[train_idx] loss = model.loss(batch_scores, batch_labels.view(-1)[train_idx]).to(torch.float) loss.backward() optimizer.step() epoch_loss = loss.detach().item() return epoch_loss def train_epoch_proteins(model, optimizer, device, train_loader, epoch=None): model.train() # pbar = tqdm(total=len(train_loader)) # pbar.set_description(f'Training epoch: {epoch:04d}') total_loss = total_examples = 0 for data in train_loader: optimizer.zero_grad() data = data.to(device) try: batch_pos_enc = data.pos_enc.to(device) sign_flip = torch.rand(batch_pos_enc.size(1)).to(device) sign_flip[sign_flip >= 0.5] = 1.0 sign_flip[sign_flip < 0.5] = -1.0 batch_pos_enc = batch_pos_enc * sign_flip.unsqueeze(0) out = model.forward(data.x, data.edge_index, data.edge_attr,batch_pos_enc) except: out = model(data.x, data.edge_index, data.edge_attr) loss = model.loss_proteins(out[data.train_mask], data.y[data.train_mask]) loss.backward() optimizer.step() total_loss += float(loss) * int(data.train_mask.sum()) total_examples += int(data.train_mask.sum()) # pbar.update(1) # # pbar.close() return total_loss / total_examples @torch.no_grad() def evaluate_network(model, device, test_loader, evaluator, epoch): model.eval() y_true = {'train': [], 'valid': [], 'test': []} y_pred = {'train': [], 'valid': [], 'test': []} # pbar = tqdm(total=len(test_loader)) # pbar.set_description(f'Evaluating epoch: {epoch:04d}') total_loss = total_examples = 0 for data in test_loader: data = data.to(device) try: batch_pos_enc = data.pos_enc.to(device) out = model.forward(data.x, data.edge_index.long(), data.edge_attr, batch_pos_enc) except: out = model.forward(data.x, data.edge_index.long(), data.edge_attr) # out = model(data.x, data.edge_index.long(), data.edge_attr) for split in y_true.keys(): mask = data[f'{split}_mask'] y_true[split].append(data.y[mask].cpu()) y_pred[split].append(out[mask].argmax(dim=-1, keepdim=True).cpu()) loss = model.loss(out[data.valid_mask], data.y.view(-1)[data.valid_mask]) total_loss += float(loss) * int(data.valid_mask.sum()) total_examples += int(data.valid_mask.sum()) # pbar.update(1) # pbar.close() train_acc = evaluator.eval({ 'y_true': torch.cat(y_true['train'], dim=0), 'y_pred': torch.cat(y_pred['train'], dim=0), })['acc'] valid_acc = evaluator.eval({ 'y_true': torch.cat(y_true['valid'], dim=0), 'y_pred': torch.cat(y_pred['valid'], dim=0), })['acc'] test_acc = evaluator.eval({ 'y_true': torch.cat(y_true['test'], dim=0), 'y_pred': torch.cat(y_pred['test'], dim=0), })['acc'] return train_acc, valid_acc, test_acc, total_loss / total_examples # model.train() # # # for iter, (batch_graphs, batch_labels) in enumerate(data_loader): # batch_x = dataset.x.to(device) # batch_e = dataset.edge_attr.to(device) # batch_labels = dataset.y.long().to(device) # edge_index = dataset.edge_index.long().to(device) # train_idx = train_idx.to(device) # # optimizer.zero_grad() # batch_scores = model.forward(batch_x, edge_index, batch_e)[train_idx] # loss = model.loss_proteins(batch_scores, batch_labels[train_idx]).to(torch.float) # loss.backward() # optimizer.step() # epoch_loss = loss.detach().item() # # return epoch_loss @torch.no_grad() def evaluate_network_arxiv(model, device, dataset, evaluator): model.eval() batch_x = dataset.dataset[0].x.to(device) y_true = dataset.dataset[0].y.long().to(device) split_idx = dataset.split_idx batch_e = dataset.dataset[0].edge_attr.to(device) edge_index = dataset.dataset[0].edge_index.long().to(device) try: batch_pos_enc = dataset.dataset[0].pos_enc.to(device) batch_scores = model.forward(batch_x, edge_index, batch_e, batch_pos_enc) except: batch_scores = model.forward(batch_x, edge_index, batch_e) # batch_scores = model.forward(batch_x, edge_index, batch_e) loss = model.loss(batch_scores[split_idx['valid']], y_true.view(-1)[split_idx['valid']]).to(torch.float) epoch_valid_loss = loss.detach().item() # y_pred = batch_scores y_pred = batch_scores.argmax(dim=-1, keepdim=True) # y_true = y_true.view(-1, 1) y_true = y_true.view(-1, 1) train_acc = evaluator.eval({ 'y_true': y_true[split_idx['train']], 'y_pred': y_pred[split_idx['train']], })['acc'] valid_acc = evaluator.eval({ 'y_true': y_true[split_idx['valid']], 'y_pred': y_pred[split_idx['valid']], })['acc'] test_acc = evaluator.eval({ 'y_true': y_true[split_idx['test']], 'y_pred': y_pred[split_idx['test']], })['acc'] return train_acc, valid_acc, test_acc, epoch_valid_loss @torch.no_grad() def evaluate_network_proteins(model, device, test_loader, evaluator, epoch = None): model.eval() y_true = {'train': [], 'valid': [], 'test': []} y_pred = {'train': [], 'valid': [], 'test': []} # pbar = tqdm(total=len(test_loader)) # pbar.set_description(f'Evaluating epoch: {epoch:04d}') total_loss = total_examples = 0 for data in test_loader: data = data.to(device) try: batch_pos_enc = data.pos_enc.to(device) out = model.forward(data.x, data.edge_index, data.edge_attr, batch_pos_enc) except: out = model.forward(data.x, data.edge_index, data.edge_attr) # out = model(data.x, data.edge_index, data.edge_attr) for split in y_true.keys(): mask = data[f'{split}_mask'] y_true[split].append(data.y[mask].cpu()) y_pred[split].append(out[mask].cpu()) loss = model.loss_proteins(out[data.valid_mask], data.y[data.valid_mask]) total_loss += float(loss) * int(data.valid_mask.sum()) total_examples += int(data.valid_mask.sum()) # pbar.update(1) # pbar.close() train_rocauc = evaluator.eval({ 'y_true': torch.cat(y_true['train'], dim=0), 'y_pred': torch.cat(y_pred['train'], dim=0), })['rocauc'] valid_rocauc = evaluator.eval({ 'y_true': torch.cat(y_true['valid'], dim=0), 'y_pred': torch.cat(y_pred['valid'], dim=0), })['rocauc'] test_rocauc = evaluator.eval({ 'y_true': torch.cat(y_true['test'], dim=0), 'y_pred': torch.cat(y_pred['test'], dim=0), })['rocauc'] return train_rocauc, valid_rocauc, test_rocauc, total_loss / total_examples # # model.eval() # batch_x = dataset.dataset[0].x.to(device) # y_true = dataset.dataset[0].y.long().to(device) # split_idx = dataset.split_idx # batch_e = dataset.dataset[0].edge_attr.to(device) # edge_index = dataset.dataset[0].edge_index.long().to(device) # # batch_scores = model.forward(batch_x, edge_index, batch_e) # loss = model.loss_proteins(batch_scores[split_idx['valid']], y_true[split_idx['valid']]).to(torch.float) # epoch_valid_loss = loss.detach().item() # y_pred = batch_scores # # y_pred = batch_scores.argmax(dim=-1, keepdim=True) # # y_true = y_true.view(-1, 1) # train_acc = evaluator.eval({ # 'y_true': y_true[split_idx['train']], # 'y_pred': y_pred[split_idx['train']], # })['rocauc'] # valid_acc = evaluator.eval({ # 'y_true': y_true[split_idx['valid']], # 'y_pred': y_pred[split_idx['valid']], # })['rocauc'] # test_acc = evaluator.eval({ # 'y_true': y_true[split_idx['test']], # 'y_pred': y_pred[split_idx['test']], # })['rocauc'] # # return train_acc, valid_acc, test_acc, epoch_valid_loss
11,004
35.440397
110
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/train/train_SBMs_node_classification.py
""" Utility functions for training one epoch and evaluating one epoch """ import torch import torch.nn as nn import math import dgl from train.metrics import accuracy_SBM as accuracy from train.metrics import accuracy_ogb """ For GCNs """ def train_epoch_sparse(model, optimizer, device, data_loader, epoch, framework = 'pyg'): model.train() epoch_loss = 0 epoch_train_acc = 0 epoch_train_acc_ogb = 0 nb_data = 0 gpu_mem = 0 if framework == 'pyg': for iter, batch_graphs in enumerate(data_loader): batch_x = batch_graphs.x.to(device) # num x feat edge_index = batch_graphs.edge_index.to(device) batch_e = batch_graphs.edge_attr.to(device) batch_labels = batch_graphs.y.long().to(device) # batch_graphs = batch_graphs.to(device) # add to satisfy the version to put the graph to the cuda optimizer.zero_grad() try: batch_pos_enc = batch_graphs.pos_enc.to(device) sign_flip = torch.rand(batch_pos_enc.size(1)).to(device) sign_flip[sign_flip >= 0.5] = 1.0 sign_flip[sign_flip < 0.5] = -1.0 batch_pos_enc = batch_pos_enc * sign_flip.unsqueeze(0) batch_scores = model.forward(batch_x, edge_index, batch_e, batch_pos_enc) except: # batch_scores = model.forward(batch_graphs.x, batch_graphs.edge_index) batch_scores = model.forward(batch_x, edge_index, batch_e) loss = model.loss(batch_scores, batch_labels) loss.backward() optimizer.step() epoch_loss += loss.detach().item() epoch_train_acc += accuracy(batch_scores, batch_labels) # epoch_train_acc_ogb += accuracy_ogb(batch_scores, batch_labels) # nb_data += batch_labels.size(0) # print("Number: ", iter) epoch_loss /= (iter + 1) epoch_train_acc /= (iter + 1) # epoch_train_acc_ogb /= nb_data return epoch_loss, epoch_train_acc, optimizer elif framework == 'dgl': for iter, (batch_graphs, batch_labels) in enumerate(data_loader): batch_x = batch_graphs.ndata['feat'].to(device) # num x feat batch_e = batch_graphs.edata['feat'].to(device) batch_labels = batch_labels.to(device) batch_graphs = batch_graphs.to(device) #add to satisfy the version to put the graph to the cuda optimizer.zero_grad() try: batch_pos_enc = batch_graphs.ndata['pos_enc'].to(device) sign_flip = torch.rand(batch_pos_enc.size(1)).to(device) sign_flip[sign_flip>=0.5] = 1.0; sign_flip[sign_flip<0.5] = -1.0 batch_pos_enc = batch_pos_enc * sign_flip.unsqueeze(0) batch_scores = model.forward(batch_graphs, batch_x, batch_e, batch_pos_enc) except: batch_scores = model.forward(batch_graphs, batch_x, batch_e) loss = model.loss(batch_scores, batch_labels) loss.backward() optimizer.step() epoch_loss += loss.detach().item() epoch_train_acc += accuracy(batch_scores, batch_labels) # print("Number: ", iter) epoch_loss /= (iter + 1) epoch_train_acc /= (iter + 1) return epoch_loss, epoch_train_acc, optimizer def evaluate_network_sparse(model, device, data_loader, epoch, framework = 'pyg'): model.eval() epoch_test_loss = 0 epoch_test_acc = 0 nb_data = 0 if framework == 'pyg': with torch.no_grad(): for iter, batch_graphs in enumerate(data_loader): batch_x = batch_graphs.x.to(device) # num x feat edge_index = batch_graphs.edge_index.to(device) batch_e = batch_graphs.edge_attr.to(device) batch_labels = batch_graphs.y.long().to(device) try: batch_pos_enc = batch_graphs.pos_enc.to(device) batch_scores = model.forward(batch_x, edge_index,batch_e, batch_pos_enc) except: batch_scores = model.forward(batch_x, edge_index,batch_e) loss = model.loss(batch_scores, batch_labels) epoch_test_loss += loss.detach().item() epoch_test_acc += accuracy(batch_scores, batch_labels) epoch_test_loss /= (iter + 1) epoch_test_acc /= (iter + 1) return epoch_test_loss, epoch_test_acc elif framework == 'dgl': with torch.no_grad(): for iter, (batch_graphs, batch_labels) in enumerate(data_loader): batch_x = batch_graphs.ndata['feat'].to(device) batch_e = batch_graphs.edata['feat'].to(device) batch_labels = batch_labels.to(device) batch_graphs = batch_graphs.to(device) # add to satisfy the version to put the graph to the cuda try: batch_pos_enc = batch_graphs.ndata['pos_enc'].to(device) batch_scores = model.forward(batch_graphs, batch_x, batch_e, batch_pos_enc) except: batch_scores = model.forward(batch_graphs, batch_x, batch_e) loss = model.loss(batch_scores, batch_labels) epoch_test_loss += loss.detach().item() epoch_test_acc += accuracy(batch_scores, batch_labels) epoch_test_loss /= (iter + 1) epoch_test_acc /= (iter + 1) return epoch_test_loss, epoch_test_acc """ For WL-GNNs """ # def train_epoch_dense(model, optimizer, device, data_loader, epoch, batch_size): # # model.train() # epoch_loss = 0 # epoch_train_acc = 0 # nb_data = 0 # gpu_mem = 0 # optimizer.zero_grad() # for iter, (x_with_node_feat, labels) in enumerate(data_loader): # x_with_node_feat = x_with_node_feat.to(device) # labels = labels.to(device) # # scores = model.forward(x_with_node_feat) # loss = model.loss(scores, labels) # loss.backward() # # if not (iter%batch_size): # optimizer.step() # optimizer.zero_grad() # # epoch_loss += loss.detach().item() # epoch_train_acc += accuracy(scores, labels) # epoch_loss /= (iter + 1) # epoch_train_acc /= (iter + 1) # # return epoch_loss, epoch_train_acc, optimizer # # # # def evaluate_network_dense(model, device, data_loader, epoch): # # model.eval() # epoch_test_loss = 0 # epoch_test_acc = 0 # nb_data = 0 # with torch.no_grad(): # for iter, (x_with_node_feat, labels) in enumerate(data_loader): # x_with_node_feat = x_with_node_feat.to(device) # labels = labels.to(device) # # scores = model.forward(x_with_node_feat) # loss = model.loss(scores, labels) # epoch_test_loss += loss.detach().item() # epoch_test_acc += accuracy(scores, labels) # epoch_test_loss /= (iter + 1) # epoch_test_acc /= (iter + 1) # # return epoch_test_loss, epoch_test_acc
7,186
38.489011
113
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/utils/cleaner_main.py
# Clean the main.py file after conversion from notebook. # Any notebook code is removed from the main.py file. import subprocess def cleaner_main(filename): # file names file_notebook = filename + '.ipynb' file_python = filename + '.py' # convert notebook to python file print('Convert ' + file_notebook + ' to ' + file_python) subprocess.check_output('jupyter nbconvert --to script ' + str(file_notebook) , shell=True) print('Clean ' + file_python) # open file with open(file_python, "r") as f_in: lines_in = f_in.readlines() # remove cell indices lines_in = [ line for i,line in enumerate(lines_in) if '# In[' not in line ] # remove comments lines_in = [ line for i,line in enumerate(lines_in) if line[0]!='#' ] # remove "in_ipynb()" function idx_start_fnc = next((i for i, x in enumerate(lines_in) if 'def in_ipynb' in x), None) if idx_start_fnc!=None: idx_end_fnc = idx_start_fnc + next((i for i, x in enumerate(lines_in[idx_start_fnc+1:]) if x[:4] not in ['\n',' ']), None) lines_in = [ line for i,line in enumerate(lines_in) if i not in range(idx_start_fnc,idx_end_fnc+1) ] list_elements_to_remove = ['in_ipynb()', 'print(notebook_mode)'] for elem in list_elements_to_remove: lines_in = [ line for i,line in enumerate(lines_in) if elem not in line ] # unindent "if notebook_mode==False" block idx_start_fnc = next((i for i, x in enumerate(lines_in) if 'if notebook_mode==False' in x), None) if idx_start_fnc!=None: idx_end_fnc = idx_start_fnc + next((i for i, x in enumerate(lines_in[idx_start_fnc+1:]) if x[:8] not in ['\n',' ']), None) for i in range(idx_start_fnc,idx_end_fnc+1): lines_in[i] = lines_in[i][4:] lines_in.pop(idx_start_fnc) list_elements_to_remove = ['# notebook mode', '# terminal mode'] for elem in list_elements_to_remove: lines_in = [ line for i,line in enumerate(lines_in) if elem not in line ] # remove remaining "if notebook_mode==True" blocks - single indent run = True while run: idx_start_fnc = next((i for i, x in enumerate(lines_in) if x[:16]=='if notebook_mode'), None) if idx_start_fnc!=None: idx_end_fnc = idx_start_fnc + next((i for i, x in enumerate(lines_in[idx_start_fnc+1:]) if x[:4] not in ['\n',' ']), None) lines_in = [ line for i,line in enumerate(lines_in) if i not in range(idx_start_fnc,idx_end_fnc+1) ] else: run = False # remove "if notebook_mode==True" block - double indents idx_start_fnc = next((i for i, x in enumerate(lines_in) if x[:20]==' if notebook_mode'), None) if idx_start_fnc!=None: idx_end_fnc = idx_start_fnc + next((i for i, x in enumerate(lines_in[idx_start_fnc+1:]) if x[:8] not in ['\n',' ']), None) lines_in = [ line for i,line in enumerate(lines_in) if i not in range(idx_start_fnc,idx_end_fnc+1) ] # prepare main() for terminal mode idx = next((i for i, x in enumerate(lines_in) if 'def main' in x), None) if idx!=None: lines_in[idx] = 'def main():' idx = next((i for i, x in enumerate(lines_in) if x[:5]=='else:'), None) if idx!=None: lines_in.pop(idx) idx = next((i for i, x in enumerate(lines_in) if x[:10]==' main()'), None) if idx!=None: lines_in[idx] = 'main()' # remove notebook variables idx = next((i for i, x in enumerate(lines_in) if 'use_gpu = True' in x), None) if idx!=None: lines_in.pop(idx) idx = next((i for i, x in enumerate(lines_in) if 'gpu_id = -1' in x), None) if idx!=None: lines_in.pop(idx) idx = next((i for i, x in enumerate(lines_in) if 'device = None' in x), None) if idx!=None: lines_in.pop(idx) run = True while run: idx = next((i for i, x in enumerate(lines_in) if x[:10]=='MODEL_NAME'), None) if idx!=None: lines_in.pop(idx) else: run = False # save clean file lines_out = str() for line in lines_in: lines_out += line with open(file_python, 'w') as f_out: f_out.write(lines_out) print('Done. ')
3,939
37.252427
136
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/ogbn.py
import time import os import pickle import numpy as np import os.path as osp import dgl import torch from torch_scatter import scatter from scipy import sparse as sp import numpy as np from tqdm import tqdm from torch_geometric.data import InMemoryDataset from torch_geometric.data import Data from scipy.sparse import csr_matrix from ogb.nodeproppred import PygNodePropPredDataset, Evaluator from torch_geometric.utils import get_laplacian # to_undirected from torch_geometric.utils.num_nodes import maybe_num_nodes from torch_sparse import coalesce class load_SBMsDataSetDGL(torch.utils.data.Dataset): def __init__(self, data_dir, name, split): self.split = split self.is_test = split.lower() in ['test', 'val'] with open(os.path.join(data_dir, name + '_%s.pkl' % self.split), 'rb') as f: self.dataset = pickle.load(f) self.node_labels = [] self.graph_lists = [] self.n_samples = len(self.dataset) self._prepare() def _prepare(self): print("preparing %d graphs for the %s set..." % (self.n_samples, self.split.upper())) for data in self.dataset: node_features = data.node_feat edge_list = (data.W != 0).nonzero() # converting adj matrix to edge_list # Create the DGL Graph g = dgl.DGLGraph() g.add_nodes(node_features.size(0)) g.ndata['feat'] = node_features.long() for src, dst in edge_list: g.add_edges(src.item(), dst.item()) # adding edge features for Residual Gated ConvNet #edge_feat_dim = g.ndata['feat'].size(1) # dim same as node feature dim edge_feat_dim = 1 # dim same as node feature dim g.edata['feat'] = torch.ones(g.number_of_edges(), edge_feat_dim) self.graph_lists.append(g) self.node_labels.append(data.node_label) def __len__(self): """Return the number of graphs in the dataset.""" return self.n_samples def __getitem__(self, idx): """ Get the idx^th sample. Parameters --------- idx : int The sample index. Returns ------- (dgl.DGLGraph, int) DGLGraph with node feature stored in `feat` field And its label. """ return self.graph_lists[idx], self.node_labels[idx] class PygNodeSBMsDataset(InMemoryDataset): def __init__(self, data_dir, name, split, transform = None, pre_transform = None, meta_dict = None ): self.split = split self.root = data_dir self.name = name self.is_test = split.lower() in ['test', 'val'] self.node_labels = [] self.graph_lists = [] super(PygNodeSBMsDataset, self).__init__(self.root, transform) self.data, self.slices = torch.load(self.processed_paths[0]) @property def raw_dir(self): return osp.join(self.root) @property def num_classes(self): return self.__num_classes__ @property def raw_file_names(self): return [os.path.join(self.name + '_%s.pkl' % self.split)] @property def processed_file_names(self): return 'geometric_data_processed' + self.name + self.split + '.pt' def download(self): r"""Downloads the dataset to the :obj:`self.raw_dir` folder.""" raise NotImplementedError def process(self): with open(os.path.join(self.root, self.name + '_%s.pkl' % self.split), 'rb') as f: self.dataset = pickle.load(f) # self.n_samples = len(self.dataset) print("preparing graphs for the %s set..." % (self.split.upper())) print('Converting graphs into PyG objects...') pyg_graph_list = [] for data in tqdm(self.dataset): node_features = data.node_feat edge_list = (data.W != 0).nonzero() # converting adj matrix to edge_list g = Data() g.__num_nodes__ = node_features.size(0) g.edge_index = edge_list.T #g.edge_index = torch.from_numpy(edge_list) g.x = node_features.long() # adding edge features for Residual Gated ConvNet edge_feat_dim = 1 g.edge_attr = torch.ones(g.num_edges, edge_feat_dim) g.y = data.node_label.to(torch.float32) pyg_graph_list.append(g) del self.dataset data, slices = self.collate(pyg_graph_list) print('Saving...') torch.save((data, slices), self.processed_paths[0]) def __repr__(self): return '{}()'.format(self.__class__.__name__) # def size_repr(key, item, indent=0): # indent_str = ' ' * indent # if torch.is_tensor(item) and item.dim() == 0: # out = item.item() # elif torch.is_tensor(item): # out = str(list(item.size())) # elif isinstance(item, list) or isinstance(item, tuple): # out = str([len(item)]) # elif isinstance(item, dict): # lines = [indent_str + size_repr(k, v, 2) for k, v in item.items()] # out = '{\n' + ',\n'.join(lines) + '\n' + indent_str + '}' # elif isinstance(item, str): # out = f'"{item}"' # else: # out = str(item) # # return f'{indent_str}{key}={out}' # # class PygNodeSBMsDataset(InMemoryDataset): # # def __init__(self, # data_dir, # name, # split, # transform = None, # pre_transform = None, # meta_dict = None # ): # # self.split = split # self.root = data_dir # self.name = name # self.is_test = split.lower() in ['test', 'val'] # # self.node_labels = [] # self.graph_lists = [] # super(PygNodeSBMsDataset, self).__init__(self.root, transform) # self.data, self.slices = torch.load(self.processed_paths[0]) # # # @property # def raw_dir(self): # return osp.join(self.root) # # @property # def num_classes(self): # return self.__num_classes__ # # @property # def raw_file_names(self): # return [os.path.join(self.name + '_%s.pkl' % self.split)] # # @property # def processed_file_names(self): # return 'geometric_data_processed' + self.name + self.split + '.pt' # # def download(self): # r"""Downloads the dataset to the :obj:`self.raw_dir` folder.""" # raise NotImplementedError # # def process(self): # with open(os.path.join(self.root, self.name + '_%s.pkl' % self.split), 'rb') as f: # self.dataset = pickle.load(f) # # self.n_samples = len(self.dataset) # print("preparing graphs for the %s set..." % (self.split.upper())) # print('Converting graphs into PyG objects...') # pyg_graph_list = [] # for data in tqdm(self.dataset): # node_features = data.node_feat # edge_list = (data.W != 0).nonzero() # converting adj matrix to edge_list # g = Data() # g.__num_nodes__ = node_features.size(0) # g.edge_index = edge_list.T # #g.edge_index = torch.from_numpy(edge_list) # g.x = node_features.long() # # adding edge features for Residual Gated ConvNet # edge_feat_dim = 1 # g.edge_attr = torch.ones(g.num_edges, edge_feat_dim) # g.y = data.node_label.to(torch.float32) # pyg_graph_list.append(g) # del self.dataset # data, slices = self.collate(pyg_graph_list) # print('Saving...') # torch.save((data, slices), self.processed_paths[0]) # # def __repr__(self): # return '{}()'.format(self.__class__.__name__) class Data_idx(object): def __init__(self, x=None, edge_index=None, edge_attr=None, y=None, pos_enc = None, **kwargs): self.x = x self.edge_index = edge_index self.y = y self.edge_attr = edge_attr self.pos_enc = pos_enc def __len__(self): r"""The number of examples in the dataset.""" return 1 def __getitem__(self, idx): # data = self.data.__class__() dataset = Data( x = self.x, edge_index = self.edge_index, y=self.y, edge_attr=self.edge_attr) if self.pos_enc == None else Data(x = self.x, edge_index = self.edge_index, y=self.y, edge_attr=self.edge_attr, pos_enc=self.pos_enc) return dataset # # def __repr__(self): # cls = str(self.__class__.__name__) # has_dict = any([isinstance(item, dict) for _, item in self]) # # if not has_dict: # info = [size_repr(key, item) for key, item in self] # return '{}({})'.format(cls, ', '.join(info)) # else: # info = [size_repr(key, item, indent=2) for key, item in self] # return '{}(\n{}\n)'.format(cls, ',\n'.join(info)) def to_undirected(edge_index, num_nodes=None): r"""Converts the graph given by :attr:`edge_index` to an undirected graph, so that :math:`(j,i) \in \mathcal{E}` for every edge :math:`(i,j) \in \mathcal{E}`. Args: edge_index (LongTensor): The edge indices. num_nodes (int, optional): The number of nodes, *i.e.* :obj:`max_val + 1` of :attr:`edge_index`. (default: :obj:`None`) :rtype: :class:`LongTensor` """ num_nodes = maybe_num_nodes(edge_index, num_nodes) row, col = edge_index value_ori = torch.full([row.size(0)], 2) value_add = torch.full([col.size(0)], 1) row, col = torch.cat([row, col], dim=0), torch.cat([col, row], dim=0) edge_attr = torch.cat([value_ori, value_add], dim=0) edge_index = torch.stack([row, col], dim=0) edge_index, edge_attr = coalesce(edge_index, edge_attr, num_nodes, num_nodes) return edge_index, edge_attr.view(-1,1).type(torch.float32) class ogbnDatasetpyg(InMemoryDataset): def __init__(self, name, use_node_embedding = False): """ TODO """ start = time.time() print("[I] Loading data ...") self.name = name data_dir = 'data/ogbn' # edge_index = self.dataset[0].edge_index # dataset = PygNodePropPredDataset(name=name, root=data_dir) if name == 'ogbn-arxiv': self.dataset = PygNodePropPredDataset(name=name, root=data_dir) self.split_idx = self.dataset.get_idx_split() self.dataset.data.edge_index, self.dataset.data.edge_attr = to_undirected(self.dataset[0].edge_index, self.dataset[0].num_nodes) self.dataset.data.y = self.dataset.data.y.squeeze(1) self.dataset.slices['edge_index'] = torch.tensor([0,self.dataset.data.edge_index.size(1)],dtype=torch.long) self.dataset.slices['edge_attr'] = torch.tensor([0, self.dataset.data.edge_index.size(1)], dtype=torch.long) if name == 'ogbn-proteins': self.dataset = PygNodePropPredDataset(name=name, root=data_dir) self.split_idx = self.dataset.get_idx_split() self.dataset.data.x = scatter(self.dataset[0].edge_attr, self.dataset[0].edge_index[0], dim=0, dim_size=self.dataset[0].num_nodes, reduce='mean').to('cpu') self.dataset.slices['x'] = torch.tensor([0, self.dataset.data.y.size(0)], dtype=torch.long) # self.edge_attr = self.dataset[0].edge_attr edge_feat_dim = 1 self.edge_attr = torch.ones(self.dataset[0].num_edges, edge_feat_dim).type(torch.float32) if name == 'ogbn-mag': dataset = PygNodePropPredDataset(name=name, root=data_dir) self.split_idx = dataset.get_idx_split() rel_data = dataset[0] edge_index, self.edge_attr = to_undirected(rel_data.edge_index_dict[('paper', 'cites', 'paper')], rel_data.num_nodes['paper']) # We are only interested in paper <-> paper relations. self.dataset = Data_idx( x=rel_data.x_dict['paper'], edge_index=edge_index, y=rel_data.y_dict['paper'], edge_attr = self.edge_attr) if name == 'ogbn-products': self.dataset = PygNodePropPredDataset(name=name, root=data_dir) self.split_idx = self.dataset.get_idx_split() self.dataset.slices['edge_attr'] = torch.tensor([0, self.dataset.data.edge_index.size(1)], dtype=torch.long) edge_feat_dim = 1 self.dataset.edge_attr = torch.ones(self.dataset[0].num_edges, edge_feat_dim).type(torch.float32) if use_node_embedding: print("use_node_embedding ...") embedding = torch.load('data/ogbn/embedding_' + name[5:] + '.pt', map_location='cpu') self.dataset.data.x = torch.cat([self.dataset.data.x, embedding], dim=-1) # self.dataset.data.embedding = embedding#torch.cat([self.dataset.data.x, embedding], dim=-1) # self.dataset.slices['embedding'] = torch.tensor([0, self.dataset.data.x.size(0)], # dtype=torch.long) # edge_attr.type = # edge_feat_dim = 1 # self.edge_attr = torch.ones(self.dataset[0].num_edges, edge_feat_dim) print("[I] Finished loading.") print("[I] Data load time: {:.4f}s".format(time.time()-start)) def _add_positional_encodings(self, pos_enc_dim, DATASET_NAME = None): # Graph positional encoding v/ Laplacian eigenvectors # self.train.graph_lists = [positional_encoding(g, pos_enc_dim) for g in self.train.graph_lists] # iter(self.train) self.graph_lists = [positional_encoding(self.dataset[0], pos_enc_dim, DATASET_NAME = DATASET_NAME)] self.dataset.data, self.dataset.slices = self.collate(self.graph_lists) def __repr__(self): return '{}()'.format(self.__class__.__name__) class SBMsDatasetDGL(torch.utils.data.Dataset): def __init__(self, name): """ TODO """ start = time.time() print("[I] Loading data ...") self.name = name data_dir = 'data/SBMs' self.train = load_SBMsDataSetDGL(data_dir, name, split='train') self.test = load_SBMsDataSetDGL(data_dir, name, split='test') self.val = load_SBMsDataSetDGL(data_dir, name, split='val') print("[I] Finished loading.") print("[I] Data load time: {:.4f}s".format(time.time()-start)) def self_loop(g): """ Utility function only, to be used only when necessary as per user self_loop flag : Overwriting the function dgl.transform.add_self_loop() to not miss ndata['feat'] and edata['feat'] This function is called inside a function in SBMsDataset class. """ new_g = dgl.DGLGraph() new_g.add_nodes(g.number_of_nodes()) new_g.ndata['feat'] = g.ndata['feat'] src, dst = g.all_edges(order="eid") src = dgl.backend.zerocopy_to_numpy(src) dst = dgl.backend.zerocopy_to_numpy(dst) non_self_edges_idx = src != dst nodes = np.arange(g.number_of_nodes()) new_g.add_edges(src[non_self_edges_idx], dst[non_self_edges_idx]) new_g.add_edges(nodes, nodes) # This new edata is not used since this function gets called only for GCN, GAT # However, we need this for the generic requirement of ndata and edata new_g.edata['feat'] = torch.zeros(new_g.number_of_edges()) return new_g def positional_encoding(g, pos_enc_dim, DATASET_NAME = None): """ Graph positional encoding v/ Laplacian eigenvectors """ # Laplacian,for the pyg try: g.pos_enc = torch.load('data/ogbn/laplacian_'+DATASET_NAME[5:]+'.pt', map_location='cpu') if g.pos_enc.size(1) != pos_enc_dim: os.remove('data/ogbn/laplacian_'+DATASET_NAME[5:]+'.pt') L = get_laplacian(g.edge_index, normalization='sym', dtype=torch.float64) L = csr_matrix((L[1], (L[0][0], L[0][1])), shape=(g.num_nodes, g.num_nodes)) # Eigenvectors with scipy # EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim+1, which='SR') EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim + 1, which='SR', tol=1e-2) # for 40 PEs EigVec = EigVec[:, EigVal.argsort()] # increasing order g.pos_enc = torch.from_numpy(EigVec[:, 1:pos_enc_dim + 1].astype(np.float32)).float() torch.save(g.pos_enc.cpu(), 'data/ogbn/laplacian_' + DATASET_NAME[5:] + '.pt') except: L = get_laplacian(g.edge_index,normalization='sym',dtype = torch.float64) L = csr_matrix((L[1], (L[0][0], L[0][1])), shape=(g.num_nodes, g.num_nodes)) # Eigenvectors with scipy # EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim+1, which='SR') EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim + 1, which='SR', tol=1e-2) # for 40 PEs EigVec = EigVec[:, EigVal.argsort()] # increasing order g.pos_enc = torch.from_numpy(EigVec[:, 1:pos_enc_dim + 1].astype(np.float32)).float() torch.save(g.pos_enc.cpu(), 'data/ogbn/laplacian_'+DATASET_NAME[5:]+'.pt') # add astype to discards the imaginary part to satisfy the version change pytorch1.5.0 # # Eigenvectors with numpy # EigVal, EigVec = np.linalg.eig(L.toarray()) # idx = EigVal.argsort() # increasing order # EigVal, EigVec = EigVal[idx], np.real(EigVec[:,idx]) # g.ndata['pos_enc'] = torch.from_numpy(np.abs(EigVec[:,1:pos_enc_dim+1])).float() return g class SBMsDataset(torch.utils.data.Dataset): def __init__(self, name): """ Loading SBM datasets """ start = time.time() print("[I] Loading dataset %s..." % (name)) self.name = name data_dir = 'data/SBMs/' with open(data_dir+name+'.pkl',"rb") as f: f = pickle.load(f) self.train = f[0] self.val = f[1] self.test = f[2] print('train, test, val sizes :',len(self.train),len(self.test),len(self.val)) print("[I] Finished loading.") print("[I] Data load time: {:.4f}s".format(time.time()-start)) # form a mini batch from a given list of samples = [(graph, label) pairs] def collate(self, samples): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.cat(labels).long() #tab_sizes_n = [ graphs[i].number_of_nodes() for i in range(len(graphs))] #tab_snorm_n = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_n ] #snorm_n = torch.cat(tab_snorm_n).sqrt() #tab_sizes_e = [ graphs[i].number_of_edges() for i in range(len(graphs))] #tab_snorm_e = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_e ] #snorm_e = torch.cat(tab_snorm_e).sqrt() batched_graph = dgl.batch(graphs) return batched_graph, labels # prepare dense tensors for GNNs which use; such as RingGNN and 3WLGNN def collate_dense_gnn(self, samples): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.cat(labels).long() #tab_sizes_n = [ graphs[i].number_of_nodes() for i in range(len(graphs))] #tab_snorm_n = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_n ] #snorm_n = tab_snorm_n[0][0].sqrt() #batched_graph = dgl.batch(graphs) g = graphs[0] adj = self._sym_normalize_adj(g.adjacency_matrix().to_dense()) """ Adapted from https://github.com/leichen2018/Ring-GNN/ Assigning node and edge feats:: we have the adjacency matrix in R^{n x n}, the node features in R^{d_n} and edge features R^{d_e}. Then we build a zero-initialized tensor, say T, in R^{(1 + d_n + d_e) x n x n}. T[0, :, :] is the adjacency matrix. The diagonal T[1:1+d_n, i, i], i = 0 to n-1, store the node feature of node i. The off diagonal T[1+d_n:, i, j] store edge features of edge(i, j). """ zero_adj = torch.zeros_like(adj) if self.name == 'SBM_CLUSTER': self.num_node_type = 7 elif self.name == 'SBM_PATTERN': self.num_node_type = 3 # use node feats to prepare adj adj_node_feat = torch.stack([zero_adj for j in range(self.num_node_type)]) adj_node_feat = torch.cat([adj.unsqueeze(0), adj_node_feat], dim=0) for node, node_label in enumerate(g.ndata['feat']): adj_node_feat[node_label.item()+1][node][node] = 1 x_node_feat = adj_node_feat.unsqueeze(0) return x_node_feat, labels def _sym_normalize_adj(self, adj): deg = torch.sum(adj, dim = 0)#.squeeze() deg_inv = torch.where(deg>0, 1./torch.sqrt(deg), torch.zeros(deg.size())) deg_inv = torch.diag(deg_inv) return torch.mm(deg_inv, torch.mm(adj, deg_inv)) def _add_self_loops(self): # function for adding self loops # this function will be called only if self_loop flag is True self.train.graph_lists = [self_loop(g) for g in self.train.graph_lists] self.val.graph_lists = [self_loop(g) for g in self.val.graph_lists] self.test.graph_lists = [self_loop(g) for g in self.test.graph_lists] def _add_positional_encodings(self, pos_enc_dim): # Graph positional encoding v/ Laplacian eigenvectors self.train.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'dgl') for g in self.train.graph_lists] self.val.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'dgl') for g in self.val.graph_lists] self.test.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'dgl') for g in self.test.graph_lists]
22,737
38.407279
140
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/molecules.py
import torch import pickle import torch.utils.data import time import os import numpy as np import csv import dgl from scipy import sparse as sp import numpy as np from torch_geometric.data import Data from torch_geometric.data import InMemoryDataset from tqdm import tqdm # *NOTE # The dataset pickle and index files are in ./zinc_molecules/ dir # [<split>.pickle and <split>.index; for split 'train', 'val' and 'test'] class MoleculeDGL(torch.utils.data.Dataset): def __init__(self, data_dir, split, num_graphs): self.data_dir = data_dir self.split = split self.num_graphs = num_graphs with open(data_dir + "/%s.pickle" % self.split,"rb") as f: self.data = pickle.load(f) # loading the sampled indices from file ./zinc_molecules/<split>.index with open(data_dir + "/%s.index" % self.split,"r") as f: data_idx = [list(map(int, idx)) for idx in csv.reader(f)] self.data = [ self.data[i] for i in data_idx[0] ] assert len(self.data)==num_graphs, "Sample num_graphs again; available idx: train/val/test => 10k/1k/1k" """ data is a list of Molecule dict objects with following attributes molecule = data[idx] ; molecule['num_atom'] : nb of atoms, an integer (N) ; molecule['atom_type'] : tensor of size N, each element is an atom type, an integer between 0 and num_atom_type ; molecule['bond_type'] : tensor of size N x N, each element is a bond type, an integer between 0 and num_bond_type ; molecule['logP_SA_cycle_normalized'] : the chemical property to regress, a float variable """ self.graph_lists = [] self.graph_labels = [] self.n_samples = len(self.data) self._prepare() def _prepare(self): print("preparing %d graphs for the %s set..." % (self.num_graphs, self.split.upper())) for molecule in self.data: node_features = molecule['atom_type'].long() adj = molecule['bond_type'] edge_list = (adj != 0).nonzero() # converting adj matrix to edge_list edge_idxs_in_adj = edge_list.split(1, dim=1) edge_features = adj[edge_idxs_in_adj].reshape(-1).long() # Create the DGL Graph g = dgl.DGLGraph() g.add_nodes(molecule['num_atom']) g.ndata['feat'] = node_features for src, dst in edge_list: g.add_edges(src.item(), dst.item()) g.edata['feat'] = edge_features self.graph_lists.append(g) self.graph_labels.append(molecule['logP_SA_cycle_normalized']) def __len__(self): """Return the number of graphs in the dataset.""" return self.n_samples def __getitem__(self, idx): """ Get the idx^th sample. Parameters --------- idx : int The sample index. Returns ------- (dgl.DGLGraph, int) DGLGraph with node feature stored in `feat` field And its label. """ return self.graph_lists[idx], self.graph_labels[idx] class Moleculepyg(InMemoryDataset): def __init__(self, data_dir, split, num_graphs, name, root = 'dataset', transform=None, pre_transform=None, meta_dict = None): self.data_dir = data_dir self.split = split self.num_graphs = num_graphs self.root = root self.name = name with open(data_dir + "/%s.pickle" % self.split, "rb") as f: self.data = pickle.load(f) # loading the sampled indices from file ./zinc_molecules/<split>.index with open(data_dir + "/%s.index" % self.split, "r") as f: data_idx = [list(map(int, idx)) for idx in csv.reader(f)] self.data = [self.data[i] for i in data_idx[0]] assert len(self.data) == num_graphs, "Sample num_graphs again; available idx: train/val/test => 10k/1k/1k" """ data is a list of Molecule dict objects with following attributes molecule = data[idx] ; molecule['num_atom'] : nb of atoms, an integer (N) ; molecule['atom_type'] : tensor of size N, each element is an atom type, an integer between 0 and num_atom_type ; molecule['bond_type'] : tensor of size N x N, each element is a bond type, an integer between 0 and num_bond_type ; molecule['logP_SA_cycle_normalized'] : the chemical property to regress, a float variable """ self.n_samples = len(self.data) super(Moleculepyg, self).__init__(self.root, transform) self.data, self.slices = torch.load(self.processed_paths[0]) #self.process() @property def processed_file_names(self): return 'geometric_data_processed' + self.name + self.split + '.pt' def process(self): print("preparing %d graphs for the %s set..." % (self.num_graphs, self.split.upper())) print('Converting graphs into PyG objects...') pyg_graph_list = [] graph_labels = [] for graph in tqdm(self.data): node_features = graph['atom_type'].long() adj = graph['bond_type'] edge_list = (adj != 0).nonzero() # converting adj matrix to edge_list edge_idxs_in_adj = edge_list.split(1, dim=1) edge_features = adj[edge_idxs_in_adj].reshape(-1).long() g = Data() g.__num_nodes__ = graph['num_atom'] g.edge_index = edge_list.T #g.edge_index = torch.from_numpy(edge_list) g.edge_attr = edge_features del graph['bond_type'] if graph['atom_type'] is not None: g.x = graph['atom_type'].long() del graph['atom_type'] graph_labels.append(graph['logP_SA_cycle_normalized']) pyg_graph_list.append(g) for i, g in enumerate(pyg_graph_list): # if 'classification' in self.task_type: # if has_nan: g.y = graph_labels[i].to(torch.float32) # else: # g.y = torch.from_numpy(graph_label[i]).view(1,-1).to(torch.long) # else: # g.y = torch.from_numpy(graph_label[i]).view(1,-1).to(torch.float32) data, slices = self.collate(pyg_graph_list) print('Saving...') torch.save((data, slices), self.processed_paths[0]) class MoleculeDatasetDGL(torch.utils.data.Dataset): def __init__(self, name='Zinc', framework = 'pyg'): t0 = time.time() self.name = name self.num_atom_type = 28 # known meta-info about the zinc dataset; can be calculated as well self.num_bond_type = 4 # known meta-info about the zinc dataset; can be calculated as well data_dir='./data/molecules' self.train = MoleculeDGL(data_dir, 'train', num_graphs=10000) self.val = MoleculeDGL(data_dir, 'val', num_graphs=1000) self.test = MoleculeDGL(data_dir, 'test', num_graphs=1000) print("Time taken: {:.4f}s".format(time.time()-t0)) class MoleculeDatasetpyg(InMemoryDataset): def __init__(self, name='ZINC'): t0 = time.time() self.name = name self.num_atom_type = 28 # known meta-info about the zinc dataset; can be calculated as well self.num_bond_type = 4 # known meta-info about the zinc dataset; can be calculated as well data_dir = './data/molecules' self.train = Moleculepyg(data_dir, 'train', num_graphs=10000, name='ZINC') self.val = Moleculepyg(data_dir, 'val', num_graphs=1000, name='ZINC') self.test = Moleculepyg(data_dir, 'test', num_graphs=1000, name='ZINC') print("Time taken: {:.4f}s".format(time.time() - t0)) def self_loop(g): """ Utility function only, to be used only when necessary as per user self_loop flag : Overwriting the function dgl.transform.add_self_loop() to not miss ndata['feat'] and edata['feat'] This function is called inside a function in MoleculeDataset class. """ new_g = dgl.DGLGraph() new_g.add_nodes(g.number_of_nodes()) new_g.ndata['feat'] = g.ndata['feat'] src, dst = g.all_edges(order="eid") src = dgl.backend.zerocopy_to_numpy(src) dst = dgl.backend.zerocopy_to_numpy(dst) non_self_edges_idx = src != dst nodes = np.arange(g.number_of_nodes()) new_g.add_edges(src[non_self_edges_idx], dst[non_self_edges_idx]) new_g.add_edges(nodes, nodes) # This new edata is not used since this function gets called only for GCN, GAT # However, we need this for the generic requirement of ndata and edata new_g.edata['feat'] = torch.zeros(new_g.number_of_edges()) return new_g def positional_encoding(g, pos_enc_dim): """ Graph positional encoding v/ Laplacian eigenvectors """ # Laplacian A = g.adjacency_matrix_scipy(return_edge_ids=False).astype(float) N = sp.diags(dgl.backend.asnumpy(g.in_degrees()).clip(1) ** -0.5, dtype=float) L = sp.eye(g.number_of_nodes()) - N * A * N # Eigenvectors with numpy EigVal, EigVec = np.linalg.eig(L.toarray()) idx = EigVal.argsort() # increasing order EigVal, EigVec = EigVal[idx], np.real(EigVec[:,idx]) g.ndata['pos_enc'] = torch.from_numpy(EigVec[:,1:pos_enc_dim+1]).float() # # Eigenvectors with scipy # EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim+1, which='SR') # EigVec = EigVec[:, EigVal.argsort()] # increasing order # g.ndata['pos_enc'] = torch.from_numpy(np.abs(EigVec[:,1:pos_enc_dim+1])).float() return g class MoleculeDataset(torch.utils.data.Dataset): def __init__(self, name): """ Loading SBM datasets """ start = time.time() print("[I] Loading dataset %s..." % (name)) self.name = name data_dir = 'data/molecules/' with open(data_dir+name+'.pkl',"rb") as f: f = pickle.load(f) self.train = f[0] self.val = f[1] self.test = f[2] self.num_atom_type = f[3] self.num_bond_type = f[4] print('train, test, val sizes :',len(self.train),len(self.test),len(self.val)) print("[I] Finished loading.") print("[I] Data load time: {:.4f}s".format(time.time()-start)) # form a mini batch from a given list of samples = [(graph, label) pairs] def collate(self, samples): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.tensor(np.array(labels)).unsqueeze(1) #tab_sizes_n = [ graphs[i].number_of_nodes() for i in range(len(graphs))] #tab_snorm_n = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_n ] #snorm_n = torch.cat(tab_snorm_n).sqrt() #tab_sizes_e = [ graphs[i].number_of_edges() for i in range(len(graphs))] #tab_snorm_e = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_e ] #snorm_e = torch.cat(tab_snorm_e).sqrt() batched_graph = dgl.batch(graphs) return batched_graph, labels # prepare dense tensors for GNNs using them; such as RingGNN, 3WLGNN def collate_dense_gnn(self, samples, edge_feat): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.tensor(np.array(labels)).unsqueeze(1) #tab_sizes_n = [ graphs[i].number_of_nodes() for i in range(len(graphs))] #tab_snorm_n = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_n ] #snorm_n = tab_snorm_n[0][0].sqrt() #batched_graph = dgl.batch(graphs) g = graphs[0] adj = self._sym_normalize_adj(g.adjacency_matrix().to_dense()) """ Adapted from https://github.com/leichen2018/Ring-GNN/ Assigning node and edge feats:: we have the adjacency matrix in R^{n x n}, the node features in R^{d_n} and edge features R^{d_e}. Then we build a zero-initialized tensor, say T, in R^{(1 + d_n + d_e) x n x n}. T[0, :, :] is the adjacency matrix. The diagonal T[1:1+d_n, i, i], i = 0 to n-1, store the node feature of node i. The off diagonal T[1+d_n:, i, j] store edge features of edge(i, j). """ zero_adj = torch.zeros_like(adj) if edge_feat: # use edge feats also to prepare adj adj_with_edge_feat = torch.stack([zero_adj for j in range(self.num_atom_type + self.num_bond_type)]) adj_with_edge_feat = torch.cat([adj.unsqueeze(0), adj_with_edge_feat], dim=0) us, vs = g.edges() for idx, edge_label in enumerate(g.edata['feat']): adj_with_edge_feat[edge_label.item()+1+self.num_atom_type][us[idx]][vs[idx]] = 1 for node, node_label in enumerate(g.ndata['feat']): adj_with_edge_feat[node_label.item()+1][node][node] = 1 x_with_edge_feat = adj_with_edge_feat.unsqueeze(0) return None, x_with_edge_feat, labels else: # use only node feats to prepare adj adj_no_edge_feat = torch.stack([zero_adj for j in range(self.num_atom_type)]) adj_no_edge_feat = torch.cat([adj.unsqueeze(0), adj_no_edge_feat], dim=0) for node, node_label in enumerate(g.ndata['feat']): adj_no_edge_feat[node_label.item()+1][node][node] = 1 x_no_edge_feat = adj_no_edge_feat.unsqueeze(0) return x_no_edge_feat, None, labels def _sym_normalize_adj(self, adj): deg = torch.sum(adj, dim = 0)#.squeeze() deg_inv = torch.where(deg>0, 1./torch.sqrt(deg), torch.zeros(deg.size())) deg_inv = torch.diag(deg_inv) return torch.mm(deg_inv, torch.mm(adj, deg_inv)) def _add_self_loops(self): # function for adding self loops # this function will be called only if self_loop flag is True self.train.graph_lists = [self_loop(g) for g in self.train.graph_lists] self.val.graph_lists = [self_loop(g) for g in self.val.graph_lists] self.test.graph_lists = [self_loop(g) for g in self.test.graph_lists] def _add_positional_encodings(self, pos_enc_dim): # Graph positional encoding v/ Laplacian eigenvectors self.train.graph_lists = [positional_encoding(g, pos_enc_dim) for g in self.train.graph_lists] self.val.graph_lists = [positional_encoding(g, pos_enc_dim) for g in self.val.graph_lists] self.test.graph_lists = [positional_encoding(g, pos_enc_dim) for g in self.test.graph_lists]
14,989
39.404313
130
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/data.py
""" File to load dataset based on user control from main file """ from data.molecules import * from data.SBMs import SBMsDataset, SBMsDatasetpyg from data.planetoids import PlanetoidDataset from data.ogbn import ogbnDatasetpyg def LoadData(DATASET_NAME, use_node_embedding = False, framework = 'dgl'): """ This function is called in the main.py file returns: ; dataset object """ # handling for MNIST or CIFAR Superpixels # handling for (ZINC) molecule dataset if DATASET_NAME == 'ZINC': return MoleculeDataset(DATASET_NAME) if 'dgl' == framework else MoleculeDatasetpyg(DATASET_NAME) # handling for SBM datasets SBM_DATASETS = ['SBM_CLUSTER', 'SBM_PATTERN'] if DATASET_NAME in SBM_DATASETS: return SBMsDatasetpyg(DATASET_NAME) if 'pyg' == framework else SBMsDataset(DATASET_NAME) if DATASET_NAME in ['Cora', 'Citeseer', 'Pubmed']: return PlanetoidDataset(DATASET_NAME, use_node_embedding = use_node_embedding) if DATASET_NAME in ['ogbn-arxiv', 'ogbn-proteins', 'ogbn-mag', 'ogbn-products']: return ogbnDatasetpyg(name=DATASET_NAME, use_node_embedding = use_node_embedding)
1,193
33.114286
104
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/node2vec_citeseer.py
import argparse import torch from torch_geometric.nn import Node2Vec from torch_geometric.utils import to_undirected import torch_geometric as pyg from ogb.nodeproppred import PygNodePropPredDataset import os.path as osp def save_embedding(model): torch.save(model.embedding.weight.data.cpu(), 'data/planetoid/embedding_Citeseer.pt') def main(): parser = argparse.ArgumentParser(description='OGBN-citeseer (Node2Vec)') parser.add_argument('--device', type=int, default=0) parser.add_argument('--embedding_dim', type=int, default=256) parser.add_argument('--walk_length', type=int, default=80) parser.add_argument('--context_size', type=int, default=20) parser.add_argument('--walks_per_node', type=int, default=10) parser.add_argument('--batch_size', type=int, default=256) parser.add_argument('--lr', type=float, default=0.01) parser.add_argument('--epochs', type=int, default=20) parser.add_argument('--log_steps', type=int, default=1) args = parser.parse_args() device = f'cuda:{args.device}' if torch.cuda.is_available() else 'cpu' device = torch.device(device) # root = osp.join(osp.dirname(osp.realpath(__file__)), '..', 'data', 'arxiv') data_dir = 'planetoid' dataset = pyg.datasets.Planetoid(name='Citeseer',root = data_dir) data = dataset[0] data.edge_index = to_undirected(data.edge_index, data.num_nodes) model = Node2Vec(data.edge_index, args.embedding_dim, args.walk_length, args.context_size, args.walks_per_node, sparse=True).to(device) loader = model.loader(batch_size=args.batch_size, shuffle=True, num_workers=4) optimizer = torch.optim.SparseAdam(model.parameters(), lr=args.lr) model.train() for epoch in range(1, args.epochs + 1): for i, (pos_rw, neg_rw) in enumerate(loader): optimizer.zero_grad() loss = model.loss(pos_rw.to(device), neg_rw.to(device)) loss.backward() optimizer.step() if (i + 1) % args.log_steps == 0: print(f'Epoch: {epoch:02d}, Step: {i+1:03d}/{len(loader)}, ' f'Loss: {loss:.4f}') if (i + 1) % 100 == 0: # Save model every 100 steps. save_embedding(model) save_embedding(model) if __name__ == "__main__": main()
2,380
36.793651
89
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/SBMs.py
import time import os import pickle import numpy as np import os.path as osp import dgl import torch from ogb.utils.url import decide_download, download_url, extract_zip from scipy import sparse as sp import numpy as np from tqdm import tqdm from torch_geometric.data import InMemoryDataset from torch_geometric.data import Data from scipy.sparse import csr_matrix from torch_geometric.utils import get_laplacian class load_SBMsDataSetDGL(torch.utils.data.Dataset): def __init__(self, data_dir, name, split): self.split = split self.is_test = split.lower() in ['test', 'val'] with open(os.path.join(data_dir, name + '_%s.pkl' % self.split), 'rb') as f: self.dataset = pickle.load(f) self.node_labels = [] self.graph_lists = [] self.n_samples = len(self.dataset) self._prepare() def _prepare(self): print("preparing %d graphs for the %s set..." % (self.n_samples, self.split.upper())) for data in self.dataset: node_features = data.node_feat edge_list = (data.W != 0).nonzero() # converting adj matrix to edge_list # Create the DGL Graph g = dgl.DGLGraph() g.add_nodes(node_features.size(0)) g.ndata['feat'] = node_features.long() for src, dst in edge_list: g.add_edges(src.item(), dst.item()) # adding edge features for Residual Gated ConvNet #edge_feat_dim = g.ndata['feat'].size(1) # dim same as node feature dim edge_feat_dim = 1 # dim same as node feature dim g.edata['feat'] = torch.ones(g.number_of_edges(), edge_feat_dim) self.graph_lists.append(g) self.node_labels.append(data.node_label) def __len__(self): """Return the number of graphs in the dataset.""" return self.n_samples def __getitem__(self, idx): """ Get the idx^th sample. Parameters --------- idx : int The sample index. Returns ------- (dgl.DGLGraph, int) DGLGraph with node feature stored in `feat` field And its label. """ return self.graph_lists[idx], self.node_labels[idx] class PygNodeSBMsDataset(InMemoryDataset): def __init__(self, data_dir, name, split, transform = None, pre_transform = None, meta_dict = None ): self.url = '' self.split = split self.root = data_dir self.original_root = data_dir self.name = name self.is_test = split.lower() in ['test', 'val'] self.node_labels = [] self.graph_lists = [] super(PygNodeSBMsDataset, self).__init__(self.root, transform) self.data, self.slices = torch.load(self.processed_paths[0]) @property def raw_dir(self): return osp.join(self.root) @property def num_classes(self): return self.__num_classes__ @property def raw_file_names(self): return [os.path.join(self.name + '_%s.pkl' % self.split)] @property def processed_file_names(self): return 'geometric_data_processed' + self.name + self.split + '.pt' # def download(self): # r"""Downloads the dataset to the :obj:`self.raw_dir` folder.""" # raise NotImplementedError def download(self): url = self.url if decide_download(url): path = download_url(url, self.original_root) extract_zip(path, self.original_root) os.unlink(path) shutil.rmtree(self.root) shutil.move(osp.join(self.original_root, self.download_name), self.root) else: print('Stop downloading.') shutil.rmtree(self.root) exit(-1) def process(self): with open(os.path.join(self.root, self.name + '_%s.pkl' % self.split), 'rb') as f: self.dataset = pickle.load(f) # self.n_samples = len(self.dataset) print("preparing graphs for the %s set..." % (self.split.upper())) print('Converting graphs into PyG objects...') pyg_graph_list = [] for data in tqdm(self.dataset): node_features = data.node_feat edge_list = (data.W != 0).nonzero() # converting adj matrix to edge_list g = Data() g.__num_nodes__ = node_features.size(0) g.edge_index = edge_list.T #g.edge_index = torch.from_numpy(edge_list) g.x = node_features.long() # adding edge features for Residual Gated ConvNet edge_feat_dim = 1 g.edge_attr = torch.ones(g.num_edges, edge_feat_dim) g.y = data.node_label.to(torch.float32) pyg_graph_list.append(g) del self.dataset data, slices = self.collate(pyg_graph_list) print('Saving...') torch.save((data, slices), self.processed_paths[0]) def __repr__(self): return '{}()'.format(self.__class__.__name__) class SBMsDatasetpyg(InMemoryDataset): def __init__(self, name): """ TODO """ start = time.time() print("[I] Loading data ...") self.name = name data_dir = 'data/SBMs' self.train = PygNodeSBMsDataset(data_dir, name, split='train') self.test = PygNodeSBMsDataset(data_dir, name, split='test') self.val = PygNodeSBMsDataset(data_dir, name, split='val') print("[I] Finished loading.") print("[I] Data load time: {:.4f}s".format(time.time()-start)) def _add_positional_encodings(self, pos_enc_dim): # Graph positional encoding v/ Laplacian eigenvectors # self.train.graph_lists = [positional_encoding(g, pos_enc_dim) for g in self.train.graph_lists] # iter(self.train) self.train.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'pyg') for _, g in enumerate(self.train)] self.val.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'pyg') for _, g in enumerate(self.val)] self.test.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'pyg') for _, g in enumerate(self.test)] # self.train.data.pos_enc = [torch.cat(g.pos_enc,dim=0) for _, g in enumerate(self.train.graph_lists)] self.train.data, self.train.slices = self.collate(self.train.graph_lists) self.val.data, self.val.slices = self.collate(self.val.graph_lists) self.test.data, self.test.slices = self.collate(self.test.graph_lists) class SBMsDatasetDGL(torch.utils.data.Dataset): def __init__(self, name): """ TODO """ start = time.time() print("[I] Loading data ...") self.name = name data_dir = 'data/SBMs' self.train = load_SBMsDataSetDGL(data_dir, name, split='train') self.test = load_SBMsDataSetDGL(data_dir, name, split='test') self.val = load_SBMsDataSetDGL(data_dir, name, split='val') print("[I] Finished loading.") print("[I] Data load time: {:.4f}s".format(time.time()-start)) def self_loop(g): """ Utility function only, to be used only when necessary as per user self_loop flag : Overwriting the function dgl.transform.add_self_loop() to not miss ndata['feat'] and edata['feat'] This function is called inside a function in SBMsDataset class. """ new_g = dgl.DGLGraph() new_g.add_nodes(g.number_of_nodes()) new_g.ndata['feat'] = g.ndata['feat'] src, dst = g.all_edges(order="eid") src = dgl.backend.zerocopy_to_numpy(src) dst = dgl.backend.zerocopy_to_numpy(dst) non_self_edges_idx = src != dst nodes = np.arange(g.number_of_nodes()) new_g.add_edges(src[non_self_edges_idx], dst[non_self_edges_idx]) new_g.add_edges(nodes, nodes) # This new edata is not used since this function gets called only for GCN, GAT # However, we need this for the generic requirement of ndata and edata new_g.edata['feat'] = torch.zeros(new_g.number_of_edges()) return new_g def positional_encoding(g, pos_enc_dim, framework = 'dgl'): """ Graph positional encoding v/ Laplacian eigenvectors """ # Laplacian,for the pyg if framework == 'pyg': L = get_laplacian(g.edge_index,normalization='sym',dtype = torch.float64) L = csr_matrix((L[1], (L[0][0], L[0][1])), shape=(g.num_nodes, g.num_nodes)) # Eigenvectors with scipy # EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim+1, which='SR') EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim + 1, which='SR', tol=1e-2) # for 40 PEs EigVec = EigVec[:, EigVal.argsort()] # increasing order g.pos_enc = torch.from_numpy(EigVec[:, 1:pos_enc_dim + 1].astype(np.float32)).float() # add astype to discards the imaginary part to satisfy the version change pytorch1.5.0 elif framework == 'dgl': A = g.adjacency_matrix_scipy(return_edge_ids=False).astype(float) N = sp.diags(dgl.backend.asnumpy(g.in_degrees()).clip(1) ** -0.5, dtype=float) L = sp.eye(g.number_of_nodes()) - N * A * N # Eigenvectors with scipy # EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim+1, which='SR') EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim + 1, which='SR', tol=1e-2) # for 40 PEs EigVec = EigVec[:, EigVal.argsort()] # increasing order g.ndata['pos_enc'] = torch.from_numpy(EigVec[:, 1:pos_enc_dim + 1].astype(np.float32)).float() # add astype to discards the imaginary part to satisfy the version change pytorch1.5.0 # # Eigenvectors with numpy # EigVal, EigVec = np.linalg.eig(L.toarray()) # idx = EigVal.argsort() # increasing order # EigVal, EigVec = EigVal[idx], np.real(EigVec[:,idx]) # g.ndata['pos_enc'] = torch.from_numpy(np.abs(EigVec[:,1:pos_enc_dim+1])).float() return g class SBMsDataset(torch.utils.data.Dataset): def __init__(self, name): """ Loading SBM datasets """ start = time.time() print("[I] Loading dataset %s..." % (name)) self.name = name data_dir = 'data/SBMs/' with open(data_dir+name+'.pkl',"rb") as f: f = pickle.load(f) self.train = f[0] self.val = f[1] self.test = f[2] print('train, test, val sizes :',len(self.train),len(self.test),len(self.val)) print("[I] Finished loading.") print("[I] Data load time: {:.4f}s".format(time.time()-start)) # form a mini batch from a given list of samples = [(graph, label) pairs] def collate(self, samples): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.cat(labels).long() #tab_sizes_n = [ graphs[i].number_of_nodes() for i in range(len(graphs))] #tab_snorm_n = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_n ] #snorm_n = torch.cat(tab_snorm_n).sqrt() #tab_sizes_e = [ graphs[i].number_of_edges() for i in range(len(graphs))] #tab_snorm_e = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_e ] #snorm_e = torch.cat(tab_snorm_e).sqrt() batched_graph = dgl.batch(graphs) return batched_graph, labels # prepare dense tensors for GNNs which use; such as RingGNN and 3WLGNN def collate_dense_gnn(self, samples): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.cat(labels).long() #tab_sizes_n = [ graphs[i].number_of_nodes() for i in range(len(graphs))] #tab_snorm_n = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_n ] #snorm_n = tab_snorm_n[0][0].sqrt() #batched_graph = dgl.batch(graphs) g = graphs[0] adj = self._sym_normalize_adj(g.adjacency_matrix().to_dense()) """ Adapted from https://github.com/leichen2018/Ring-GNN/ Assigning node and edge feats:: we have the adjacency matrix in R^{n x n}, the node features in R^{d_n} and edge features R^{d_e}. Then we build a zero-initialized tensor, say T, in R^{(1 + d_n + d_e) x n x n}. T[0, :, :] is the adjacency matrix. The diagonal T[1:1+d_n, i, i], i = 0 to n-1, store the node feature of node i. The off diagonal T[1+d_n:, i, j] store edge features of edge(i, j). """ zero_adj = torch.zeros_like(adj) if self.name == 'SBM_CLUSTER': self.num_node_type = 7 elif self.name == 'SBM_PATTERN': self.num_node_type = 3 # use node feats to prepare adj adj_node_feat = torch.stack([zero_adj for j in range(self.num_node_type)]) adj_node_feat = torch.cat([adj.unsqueeze(0), adj_node_feat], dim=0) for node, node_label in enumerate(g.ndata['feat']): adj_node_feat[node_label.item()+1][node][node] = 1 x_node_feat = adj_node_feat.unsqueeze(0) return x_node_feat, labels def _sym_normalize_adj(self, adj): deg = torch.sum(adj, dim = 0)#.squeeze() deg_inv = torch.where(deg>0, 1./torch.sqrt(deg), torch.zeros(deg.size())) deg_inv = torch.diag(deg_inv) return torch.mm(deg_inv, torch.mm(adj, deg_inv)) def _add_self_loops(self): # function for adding self loops # this function will be called only if self_loop flag is True self.train.graph_lists = [self_loop(g) for g in self.train.graph_lists] self.val.graph_lists = [self_loop(g) for g in self.val.graph_lists] self.test.graph_lists = [self_loop(g) for g in self.test.graph_lists] def _add_positional_encodings(self, pos_enc_dim): # Graph positional encoding v/ Laplacian eigenvectors self.train.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'dgl') for g in self.train.graph_lists] self.val.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'dgl') for g in self.val.graph_lists] self.test.graph_lists = [positional_encoding(g, pos_enc_dim, framework = 'dgl') for g in self.test.graph_lists]
14,591
37.70557
127
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/node2vec_proteins.py
import argparse import torch from torch_geometric.nn import Node2Vec from ogb.nodeproppred import PygNodePropPredDataset def save_embedding(model): torch.save(model.embedding.weight.data.cpu(), 'ogbn/embedding_proteins.pt') def main(): parser = argparse.ArgumentParser(description='OGBN-Proteins (Node2Vec)') parser.add_argument('--device', type=int, default=0) parser.add_argument('--embedding_dim', type=int, default=128) parser.add_argument('--walk_length', type=int, default=80) parser.add_argument('--context_size', type=int, default=20) parser.add_argument('--walks_per_node', type=int, default=10) parser.add_argument('--batch_size', type=int, default=256) parser.add_argument('--lr', type=float, default=0.01) parser.add_argument('--epochs', type=int, default=1) parser.add_argument('--log_steps', type=int, default=1) args = parser.parse_args() device = f'cuda:{args.device}' if torch.cuda.is_available() else 'cpu' device = torch.device(device) dataset = PygNodePropPredDataset(name='ogbn-proteins',root = 'ogbn') data = dataset[0] model = Node2Vec(data.edge_index, args.embedding_dim, args.walk_length, args.context_size, args.walks_per_node, sparse=True).to(device) loader = model.loader(batch_size=args.batch_size, shuffle=True, num_workers=4) optimizer = torch.optim.SparseAdam(model.parameters(), lr=args.lr) model.train() for epoch in range(1, args.epochs + 1): for i, (pos_rw, neg_rw) in enumerate(loader): optimizer.zero_grad() loss = model.loss(pos_rw.to(device), neg_rw.to(device)) loss.backward() optimizer.step() if (i + 1) % args.log_steps == 0: print(f'Epoch: {epoch:02d}, Step: {i+1:03d}/{len(loader)}, ' f'Loss: {loss:.4f}') if (i + 1) % 100 == 0: # Save model every 100 steps. save_embedding(model) save_embedding(model) if __name__ == "__main__": main()
2,096
34.542373
79
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/CSL.py
import numpy as np, time, pickle, random, csv import torch from torch.utils.data import DataLoader, Dataset import os import pickle import numpy as np import dgl from sklearn.model_selection import StratifiedKFold, train_test_split random.seed(42) from scipy import sparse as sp class DGLFormDataset(torch.utils.data.Dataset): """ DGLFormDataset wrapping graph list and label list as per pytorch Dataset. *lists (list): lists of 'graphs' and 'labels' with same len(). """ def __init__(self, *lists): assert all(len(lists[0]) == len(li) for li in lists) self.lists = lists self.graph_lists = lists[0] self.graph_labels = lists[1] def __getitem__(self, index): return tuple(li[index] for li in self.lists) def __len__(self): return len(self.lists[0]) def format_dataset(dataset): """ Utility function to recover data, INTO-> dgl/pytorch compatible format """ graphs = [data[0] for data in dataset] labels = [data[1] for data in dataset] for graph in graphs: #graph.ndata['feat'] = torch.FloatTensor(graph.ndata['feat']) graph.ndata['feat'] = graph.ndata['feat'].float() # dgl 4.0 # adding edge features for Residual Gated ConvNet, if not there if 'feat' not in graph.edata.keys(): edge_feat_dim = graph.ndata['feat'].shape[1] # dim same as node feature dim graph.edata['feat'] = torch.ones(graph.number_of_edges(), edge_feat_dim) return DGLFormDataset(graphs, labels) def get_all_split_idx(dataset): """ - Split total number of graphs into 3 (train, val and test) in 3:1:1 - Stratified split proportionate to original distribution of data with respect to classes - Using sklearn to perform the split and then save the indexes - Preparing 5 such combinations of indexes split to be used in Graph NNs - As with KFold, each of the 5 fold have unique test set. """ root_idx_dir = './data/CSL/' if not os.path.exists(root_idx_dir): os.makedirs(root_idx_dir) all_idx = {} # If there are no idx files, do the split and store the files if not (os.path.exists(root_idx_dir + dataset.name + '_train.index')): print("[!] Splitting the data into train/val/test ...") # Using 5-fold cross val as used in RP-GNN paper k_splits = 5 cross_val_fold = StratifiedKFold(n_splits=k_splits, shuffle=True) k_data_splits = [] # this is a temporary index assignment, to be used below for val splitting for i in range(len(dataset.graph_lists)): dataset[i][0].a = lambda: None setattr(dataset[i][0].a, 'index', i) for indexes in cross_val_fold.split(dataset.graph_lists, dataset.graph_labels): remain_index, test_index = indexes[0], indexes[1] remain_set = format_dataset([dataset[index] for index in remain_index]) # Gets final 'train' and 'val' train, val, _, __ = train_test_split(remain_set, range(len(remain_set.graph_lists)), test_size=0.25, stratify=remain_set.graph_labels) train, val = format_dataset(train), format_dataset(val) test = format_dataset([dataset[index] for index in test_index]) # Extracting only idxs idx_train = [item[0].a.index for item in train] idx_val = [item[0].a.index for item in val] idx_test = [item[0].a.index for item in test] f_train_w = csv.writer(open(root_idx_dir + dataset.name + '_train.index', 'a+')) f_val_w = csv.writer(open(root_idx_dir + dataset.name + '_val.index', 'a+')) f_test_w = csv.writer(open(root_idx_dir + dataset.name + '_test.index', 'a+')) f_train_w.writerow(idx_train) f_val_w.writerow(idx_val) f_test_w.writerow(idx_test) print("[!] Splitting done!") # reading idx from the files for section in ['train', 'val', 'test']: with open(root_idx_dir + dataset.name + '_'+ section + '.index', 'r') as f: reader = csv.reader(f) all_idx[section] = [list(map(int, idx)) for idx in reader] return all_idx class CSL(torch.utils.data.Dataset): """ Circular Skip Link Graphs: Source: https://github.com/PurdueMINDS/RelationalPooling/ """ def __init__(self, path="data/CSL/"): self.name = "CSL" self.adj_list = pickle.load(open(os.path.join(path, 'graphs_Kary_Deterministic_Graphs.pkl'), 'rb')) self.graph_labels = torch.load(os.path.join(path, 'y_Kary_Deterministic_Graphs.pt')) self.graph_lists = [] self.n_samples = len(self.graph_labels) self.num_node_type = 1 #41 self.num_edge_type = 1 #164 self._prepare() def _prepare(self): t0 = time.time() print("[I] Preparing Circular Skip Link Graphs v4 ...") for sample in self.adj_list: _g = dgl.DGLGraph() _g.from_scipy_sparse_matrix(sample) g = dgl.transform.remove_self_loop(_g) g.ndata['feat'] = torch.zeros(g.number_of_nodes()).long() #g.ndata['feat'] = torch.arange(0, g.number_of_nodes()).long() # v1 #g.ndata['feat'] = torch.randperm(g.number_of_nodes()).long() # v3 # adding edge features as generic requirement g.edata['feat'] = torch.zeros(g.number_of_edges()).long() #g.edata['feat'] = torch.arange(0, g.number_of_edges()).long() # v1 #g.edata['feat'] = torch.ones(g.number_of_edges()).long() # v2 # NOTE: come back here, to define edge features as distance between the indices of the edges ################################################################### # srcs, dsts = new_g.edges() # edge_feat = [] # for edge in range(len(srcs)): # a = srcs[edge].item() # b = dsts[edge].item() # edge_feat.append(abs(a-b)) # g.edata['feat'] = torch.tensor(edge_feat, dtype=torch.int).long() ################################################################### self.graph_lists.append(g) self.num_node_type = self.graph_lists[0].ndata['feat'].size(0) self.num_edge_type = self.graph_lists[0].edata['feat'].size(0) print("[I] Finished preparation after {:.4f}s".format(time.time()-t0)) def __len__(self): return self.n_samples def __getitem__(self, idx): return self.graph_lists[idx], self.graph_labels[idx] def self_loop(g): """ Utility function only, to be used only when necessary as per user self_loop flag : Overwriting the function dgl.transform.add_self_loop() to not miss ndata['feat'] and edata['feat'] This function is called inside a function in TUsDataset class. """ new_g = dgl.DGLGraph() new_g.add_nodes(g.number_of_nodes()) new_g.ndata['feat'] = g.ndata['feat'] src, dst = g.all_edges(order="eid") src = dgl.backend.zerocopy_to_numpy(src) dst = dgl.backend.zerocopy_to_numpy(dst) non_self_edges_idx = src != dst nodes = np.arange(g.number_of_nodes()) new_g.add_edges(src[non_self_edges_idx], dst[non_self_edges_idx]) new_g.add_edges(nodes, nodes) # This new edata is not used since this function gets called only for GCN, GAT # However, we need this for the generic requirement of ndata and edata new_g.edata['feat'] = torch.zeros(new_g.number_of_edges()) return new_g def positional_encoding(g, pos_enc_dim): """ Graph positional encoding v/ Laplacian eigenvectors """ n = g.number_of_nodes() # Laplacian A = g.adjacency_matrix_scipy(return_edge_ids=False).astype(float) N = sp.diags(dgl.backend.asnumpy(g.in_degrees()).clip(1) ** -0.5, dtype=float) L = sp.eye(n) - N * A * N # Eigenvectors EigVal, EigVec = np.linalg.eig(L.toarray()) idx = EigVal.argsort() # increasing order EigVal, EigVec = EigVal[idx], np.real(EigVec[:,idx]) g.ndata['pos_enc'] = torch.from_numpy(EigVec[:,1:pos_enc_dim+1]).float() return g class CSLDataset(torch.utils.data.Dataset): def __init__(self, name='CSL'): t0 = time.time() self.name = name dataset = CSL() print("[!] Dataset: ", self.name) # this function splits data into train/val/test and returns the indices self.all_idx = get_all_split_idx(dataset) self.all = dataset self.train = [self.format_dataset([dataset[idx] for idx in self.all_idx['train'][split_num]]) for split_num in range(5)] self.val = [self.format_dataset([dataset[idx] for idx in self.all_idx['val'][split_num]]) for split_num in range(5)] self.test = [self.format_dataset([dataset[idx] for idx in self.all_idx['test'][split_num]]) for split_num in range(5)] print("Time taken: {:.4f}s".format(time.time()-t0)) def format_dataset(self, dataset): """ Utility function to recover data, INTO-> dgl/pytorch compatible format """ graphs = [data[0] for data in dataset] labels = [data[1] for data in dataset] return DGLFormDataset(graphs, labels) # form a mini batch from a given list of samples = [(graph, label) pairs] def collate(self, samples): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.tensor(np.array(labels)) batched_graph = dgl.batch(graphs) return batched_graph, labels # prepare dense tensors for GNNs using them; such as RingGNN, 3WLGNN def collate_dense_gnn(self, samples, pos_enc): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.tensor(np.array(labels)) g = graphs[0] adj = self._sym_normalize_adj(g.adjacency_matrix().to_dense()) """ Adapted from https://github.com/leichen2018/Ring-GNN/ Assigning node and edge feats:: we have the adjacency matrix in R^{n x n}, the node features in R^{d_n} and edge features R^{d_e}. Then we build a zero-initialized tensor, say T, in R^{(1 + d_n + d_e) x n x n}. T[0, :, :] is the adjacency matrix. The diagonal T[1:1+d_n, i, i], i = 0 to n-1, store the node feature of node i. The off diagonal T[1+d_n:, i, j] store edge features of edge(i, j). """ zero_adj = torch.zeros_like(adj) if pos_enc: in_dim = g.ndata['pos_enc'].shape[1] # use node feats to prepare adj adj_node_feat = torch.stack([zero_adj for j in range(in_dim)]) adj_node_feat = torch.cat([adj.unsqueeze(0), adj_node_feat], dim=0) for node, node_feat in enumerate(g.ndata['pos_enc']): adj_node_feat[1:, node, node] = node_feat x_node_feat = adj_node_feat.unsqueeze(0) return x_node_feat, labels else: # no node features here in_dim = 1 # use node feats to prepare adj adj_node_feat = torch.stack([zero_adj for j in range(in_dim)]) adj_node_feat = torch.cat([adj.unsqueeze(0), adj_node_feat], dim=0) for node, node_feat in enumerate(g.ndata['feat']): adj_node_feat[1:, node, node] = node_feat x_no_node_feat = adj_node_feat.unsqueeze(0) return x_no_node_feat, labels def _sym_normalize_adj(self, adj): deg = torch.sum(adj, dim = 0)#.squeeze() deg_inv = torch.where(deg>0, 1./torch.sqrt(deg), torch.zeros(deg.size())) deg_inv = torch.diag(deg_inv) return torch.mm(deg_inv, torch.mm(adj, deg_inv)) def _add_self_loops(self): # function for adding self loops # this function will be called only if self_loop flag is True for split_num in range(5): self.train[split_num].graph_lists = [self_loop(g) for g in self.train[split_num].graph_lists] self.val[split_num].graph_lists = [self_loop(g) for g in self.val[split_num].graph_lists] self.test[split_num].graph_lists = [self_loop(g) for g in self.test[split_num].graph_lists] for split_num in range(5): self.train[split_num] = DGLFormDataset(self.train[split_num].graph_lists, self.train[split_num].graph_labels) self.val[split_num] = DGLFormDataset(self.val[split_num].graph_lists, self.val[split_num].graph_labels) self.test[split_num] = DGLFormDataset(self.test[split_num].graph_lists, self.test[split_num].graph_labels) def _add_positional_encodings(self, pos_enc_dim): # Graph positional encoding v/ Laplacian eigenvectors for split_num in range(5): self.train[split_num].graph_lists = [positional_encoding(g, pos_enc_dim) for g in self.train[split_num].graph_lists] self.val[split_num].graph_lists = [positional_encoding(g, pos_enc_dim) for g in self.val[split_num].graph_lists] self.test[split_num].graph_lists = [positional_encoding(g, pos_enc_dim) for g in self.test[split_num].graph_lists]
13,727
40.101796
128
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/node2vec-products.py
import argparse import torch from torch_geometric.nn import Node2Vec from ogb.nodeproppred import PygNodePropPredDataset def save_embedding(model): torch.save(model.embedding.weight.data.cpu(), 'ogbn/embedding_products.pt') def main(): parser = argparse.ArgumentParser(description='OGBN-Products (Node2Vec)') parser.add_argument('--device', type=int, default=0) parser.add_argument('--embedding_dim', type=int, default=128) parser.add_argument('--walk_length', type=int, default=40) parser.add_argument('--context_size', type=int, default=20) parser.add_argument('--walks_per_node', type=int, default=10) parser.add_argument('--batch_size', type=int, default=256) parser.add_argument('--lr', type=float, default=0.01) parser.add_argument('--epochs', type=int, default=1) parser.add_argument('--log_steps', type=int, default=1) args = parser.parse_args() device = f'cuda:{args.device}' if torch.cuda.is_available() else 'cpu' device = torch.device(device) dataset = PygNodePropPredDataset(name='ogbn-products',root='ogbn') data = dataset[0] model = Node2Vec(data.edge_index, args.embedding_dim, args.walk_length, args.context_size, args.walks_per_node, sparse=True).to(device) loader = model.loader(batch_size=args.batch_size, shuffle=True, num_workers=4) #change from 4 to 0 for convient debug optimizer = torch.optim.SparseAdam(model.parameters(), lr=args.lr) model.train() for epoch in range(1, args.epochs + 1): for i, (pos_rw, neg_rw) in enumerate(loader): optimizer.zero_grad() loss = model.loss(pos_rw.to(device), neg_rw.to(device)) loss.backward() optimizer.step() if (i + 1) % args.log_steps == 0: print(f'Epoch: {epoch:02d}, Step: {i+1:03d}/{len(loader)}, ' f'Loss: {loss:.4f}') if (i + 1) % 100 == 0: # Save model every 100 steps. save_embedding(model) save_embedding(model) if __name__ == "__main__": main()
2,133
35.169492
79
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/planetoids.py
import torch import pickle import torch.utils.data import time import os import numpy as np from torch_geometric.utils import get_laplacian import csv from scipy import sparse as sp import dgl from dgl.data import TUDataset from dgl.data import LegacyTUDataset import torch_geometric as pyg from scipy.sparse import csr_matrix import random random.seed(42) from sklearn.model_selection import StratifiedKFold, train_test_split from torch_geometric.data import InMemoryDataset import csv import json class pygFormDataset(torch.utils.data.Dataset): """ DGLFormDataset wrapping graph list and label list as per pytorch Dataset. *lists (list): lists of 'graphs' and 'labels' with same len(). """ def __init__(self, *lists): assert all(len(lists[0]) == len(li) for li in lists) self.lists = lists self.node_lists = lists[0] self.node_labels = lists[1] def __getitem__(self, index): return tuple(li[index] for li in self.lists) def __len__(self): return len(self.lists[0]) def format_dataset(dataset): """ Utility function to recover data, INTO-> dgl/pytorch compatible format """ nodes = [data[0] for data in dataset] labels = [data[1] for data in dataset] return pygFormDataset(nodes, labels) class NumpyEncoder(json.JSONEncoder): def default(self, obj): if isinstance(obj, np.ndarray): return obj.tolist() return json.JSONEncoder.default(self, obj) def get_all_split_idx(dataset): """ - Split total number of graphs into 3 (train, val and test) in 80:10:10 - Stratified split proportionate to original distribution of data with respect to classes - Using sklearn to perform the split and then save the indexes - Preparing 10 such combinations of indexes split to be used in Graph NNs - As with KFold, each of the 10 fold have unique test set. """ root_idx_dir = './data/planetoid/' if not os.path.exists(root_idx_dir): os.makedirs(root_idx_dir) # If there are no idx files, do the split and store the files if not os.path.exists(root_idx_dir + f"{dataset.name}_splits.json"): print("[!] Splitting the data into train/val/test ...") all_idxs = np.arange(dataset[0].num_nodes) # Using 10-fold cross val to compare with benchmark papers k_splits = 10 cross_val_fold = StratifiedKFold(n_splits=k_splits, shuffle=True) k_data_splits = [] split = {"train": [], "val": [], "test": []} for train_ok_split, test_ok_split in cross_val_fold.split(X = all_idxs, y = dataset[0].y): # split = {"train": [], "val": [], "test": all_idxs[test_ok_split]} train_ok_targets = dataset[0].y[train_ok_split] # Gets final 'train' and 'val' train_i_split, val_i_split = train_test_split(train_ok_split, test_size=0.111, stratify=train_ok_targets) # Extracting only idxs split['train'].append(train_i_split) split['val'].append(val_i_split) split['test'].append(all_idxs[test_ok_split]) filename = root_idx_dir + f"{dataset.name}_splits.json" with open(filename, "w") as f: json.dump(split, f, cls=NumpyEncoder) # , cls=NumpyEncoder print("[!] Splitting done!") # reading idx from the files with open(root_idx_dir + f"{dataset.name}_splits.json", "r") as fp: all_idx = json.load(fp) return all_idx class DGLFormDataset(torch.utils.data.Dataset): """ DGLFormDataset wrapping graph list and label list as per pytorch Dataset. *lists (list): lists of 'graphs' and 'labels' with same len(). """ def __init__(self, *lists): assert all(len(lists[0]) == len(li) for li in lists) self.lists = lists self.graph_lists = lists[0] self.graph_labels = lists[1] def __getitem__(self, index): return tuple(li[index] for li in self.lists) def __len__(self): return len(self.lists[0]) def self_loop(g): """ Utility function only, to be used only when necessary as per user self_loop flag : Overwriting the function dgl.transform.add_self_loop() to not miss ndata['feat'] and edata['feat'] This function is called inside a function in TUsDataset class. """ new_g = dgl.DGLGraph() new_g.add_nodes(g.number_of_nodes()) new_g.ndata['feat'] = g.ndata['feat'] src, dst = g.all_edges(order="eid") src = dgl.backend.zerocopy_to_numpy(src) dst = dgl.backend.zerocopy_to_numpy(dst) non_self_edges_idx = src != dst nodes = np.arange(g.number_of_nodes()) new_g.add_edges(src[non_self_edges_idx], dst[non_self_edges_idx]) new_g.add_edges(nodes, nodes) # This new edata is not used since this function gets called only for GCN, GAT # However, we need this for the generic requirement of ndata and edata new_g.edata['feat'] = torch.zeros(new_g.number_of_edges()) return new_g def positional_encoding(g, pos_enc_dim, framework = 'pyg'): """ Graph positional encoding v/ Laplacian eigenvectors """ # Laplacian,for the pyg if framework == 'pyg': L = get_laplacian(g.edge_index,normalization='sym',dtype = torch.float64) L = csr_matrix((L[1], (L[0][0], L[0][1])), shape=(g.num_nodes, g.num_nodes)) # Eigenvectors with scipy # EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim+1, which='SR') EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim + 1, which='SR', tol=1e-2) # for 40 PEs EigVec = EigVec[:, EigVal.argsort()] # increasing order pos_enc = torch.from_numpy(EigVec[:, 1:pos_enc_dim + 1].astype(np.float32)).float() return pos_enc # add astype to discards the imaginary part to satisfy the version change pytorch1.5.0 elif framework == 'dgl': A = g.adjacency_matrix_scipy(return_edge_ids=False).astype(float) N = sp.diags(dgl.backend.asnumpy(g.in_degrees()).clip(1) ** -0.5, dtype=float) L = sp.eye(g.number_of_nodes()) - N * A * N # Eigenvectors with scipy # EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim+1, which='SR') EigVal, EigVec = sp.linalg.eigs(L, k=pos_enc_dim + 1, which='SR', tol=1e-2) # for 40 PEs EigVec = EigVec[:, EigVal.argsort()] # increasing order g.ndata['pos_enc'] = torch.from_numpy(EigVec[:, 1:pos_enc_dim + 1].astype(np.float32)).float() # add astype to discards the imaginary part to satisfy the version change pytorch1.5.0 class PlanetoidDataset(InMemoryDataset): def __init__(self, name, use_node_embedding = False): t0 = time.time() self.name = name data_dir = 'data/planetoid' #dataset = TUDataset(self.name, hidden_size=1) # dataset = LegacyTUDataset(self.name, hidden_size=1) # dgl 4.0 self.dataset = pyg.datasets.Planetoid(root=data_dir, name= name ,split = 'full') print("[!] Dataset: ", self.name) if use_node_embedding: embedding = torch.load(data_dir + '/embedding_'+name + '.pt', map_location='cpu') # self.dataset.data.x = embedding # self.laplacian = positional_encoding(self.dataset[0], 200, framework = 'pyg') self.dataset.data.x = torch.cat([self.dataset.data.x, embedding], dim=-1) # this function splits data into train/val/test and returns the indices self.all_idx = get_all_split_idx(self.dataset) edge_feat_dim = 1 self.edge_attr = torch.ones(self.dataset[0].num_edges, edge_feat_dim) # self.all = dataset # dataset.train[split_number] self.train_idx = [torch.tensor(self.all_idx['train'][split_num], dtype=torch.long) for split_num in range(10)] self.val_idx = [torch.tensor(self.all_idx['val'][split_num], dtype=torch.long) for split_num in range(10)] self.test_idx = [torch.tensor(self.all_idx['test'][split_num], dtype=torch.long) for split_num in range(10)] # self.train = [self.format_dataset([dataset[idx] for idx in self.all_idx['train'][split_num]]) for split_num in range(10)] # self.val = [self.format_dataset([dataset[idx] for idx in self.all_idx['val'][split_num]]) for split_num in range(10)] # self.test = [self.format_dataset([dataset[idx] for idx in self.all_idx['test'][split_num]]) for split_num in range(10)] print("Time taken: {:.4f}s".format(time.time()-t0)) def format_dataset(self, dataset): """ Utility function to recover data, INTO-> dgl/pytorch compatible format """ graphs = [data[0] for data in dataset] labels = [data[1] for data in dataset] for graph in graphs: #graph.ndata['feat'] = torch.FloatTensor(graph.ndata['feat']) graph.ndata['feat'] = graph.ndata['feat'].float() # dgl 4.0 # adding edge features for Residual Gated ConvNet, if not there if 'feat' not in graph.edata.keys(): edge_feat_dim = graph.ndata['feat'].shape[1] # dim same as node feature dim graph.edata['feat'] = torch.ones(graph.number_of_edges(), edge_feat_dim) return DGLFormDataset(graphs, labels) # form a mini batch from a given list of samples = [(graph, label) pairs] def collate(self, samples): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.tensor(np.array(labels)) #tab_sizes_n = [ graphs[i].number_of_nodes() for i in range(len(graphs))] #tab_snorm_n = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_n ] #snorm_n = torch.cat(tab_snorm_n).sqrt() #tab_sizes_e = [ graphs[i].number_of_edges() for i in range(len(graphs))] #tab_snorm_e = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_e ] #snorm_e = torch.cat(tab_snorm_e).sqrt() batched_graph = dgl.batch(graphs) return batched_graph, labels # prepare dense tensors for GNNs using them; such as RingGNN, 3WLGNN def collate_dense_gnn(self, samples): # The input samples is a list of pairs (graph, label). graphs, labels = map(list, zip(*samples)) labels = torch.tensor(np.array(labels)) #tab_sizes_n = [ graphs[i].number_of_nodes() for i in range(len(graphs))] #tab_snorm_n = [ torch.FloatTensor(size,1).fill_(1./float(size)) for size in tab_sizes_n ] #snorm_n = tab_snorm_n[0][0].sqrt() #batched_graph = dgl.batch(graphs) g = graphs[0] adj = self._sym_normalize_adj(g.adjacency_matrix().to_dense()) """ Adapted from https://github.com/leichen2018/Ring-GNN/ Assigning node and edge feats:: we have the adjacency matrix in R^{n x n}, the node features in R^{d_n} and edge features R^{d_e}. Then we build a zero-initialized tensor, say T, in R^{(1 + d_n + d_e) x n x n}. T[0, :, :] is the adjacency matrix. The diagonal T[1:1+d_n, i, i], i = 0 to n-1, store the node feature of node i. The off diagonal T[1+d_n:, i, j] store edge features of edge(i, j). """ zero_adj = torch.zeros_like(adj) in_dim = g.ndata['feat'].shape[1] # use node feats to prepare adj adj_node_feat = torch.stack([zero_adj for j in range(in_dim)]) adj_node_feat = torch.cat([adj.unsqueeze(0), adj_node_feat], dim=0) for node, node_feat in enumerate(g.ndata['feat']): adj_node_feat[1:, node, node] = node_feat x_node_feat = adj_node_feat.unsqueeze(0) return x_node_feat, labels def _sym_normalize_adj(self, adj): deg = torch.sum(adj, dim = 0)#.squeeze() deg_inv = torch.where(deg>0, 1./torch.sqrt(deg), torch.zeros(deg.size())) deg_inv = torch.diag(deg_inv) return torch.mm(deg_inv, torch.mm(adj, deg_inv)) def _add_self_loops(self): # function for adding self loops # this function will be called only if self_loop flag is True for split_num in range(10): self.train[split_num].graph_lists = [self_loop(g) for g in self.train[split_num].graph_lists] self.val[split_num].graph_lists = [self_loop(g) for g in self.val[split_num].graph_lists] self.test[split_num].graph_lists = [self_loop(g) for g in self.test[split_num].graph_lists] for split_num in range(10): self.train[split_num] = DGLFormDataset(self.train[split_num].graph_lists, self.train[split_num].graph_labels) self.val[split_num] = DGLFormDataset(self.val[split_num].graph_lists, self.val[split_num].graph_labels) self.test[split_num] = DGLFormDataset(self.test[split_num].graph_lists, self.test[split_num].graph_labels)
13,158
43.60678
131
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/node2vec_arxiv.py
import argparse import torch from torch_geometric.nn import Node2Vec from torch_geometric.utils import to_undirected from ogb.nodeproppred import PygNodePropPredDataset import os.path as osp def save_embedding(model): torch.save(model.embedding.weight.data.cpu(), 'ogbn/embedding_arxiv.pt') def main(): parser = argparse.ArgumentParser(description='OGBN-Arxiv (Node2Vec)') parser.add_argument('--device', type=int, default=0) parser.add_argument('--embedding_dim', type=int, default=128) parser.add_argument('--walk_length', type=int, default=80) parser.add_argument('--context_size', type=int, default=20) parser.add_argument('--walks_per_node', type=int, default=10) parser.add_argument('--batch_size', type=int, default=256) parser.add_argument('--lr', type=float, default=0.01) parser.add_argument('--epochs', type=int, default=5) parser.add_argument('--log_steps', type=int, default=1) args = parser.parse_args() device = f'cuda:{args.device}' if torch.cuda.is_available() else 'cpu' device = torch.device(device) # root = osp.join(osp.dirname(osp.realpath(__file__)), '..', 'data', 'arxiv') dataset = PygNodePropPredDataset(name='ogbn-arxiv',root = 'ogbn') data = dataset[0] data.edge_index = to_undirected(data.edge_index, data.num_nodes) model = Node2Vec(data.edge_index, args.embedding_dim, args.walk_length, args.context_size, args.walks_per_node, sparse=True).to(device) loader = model.loader(batch_size=args.batch_size, shuffle=True, num_workers=4) optimizer = torch.optim.SparseAdam(model.parameters(), lr=args.lr) model.train() for epoch in range(1, args.epochs + 1): for i, (pos_rw, neg_rw) in enumerate(loader): optimizer.zero_grad() loss = model.loss(pos_rw.to(device), neg_rw.to(device)) loss.backward() optimizer.step() if (i + 1) % args.log_steps == 0: print(f'Epoch: {epoch:02d}, Step: {i+1:03d}/{len(loader)}, ' f'Loss: {loss:.4f}') if (i + 1) % 100 == 0: # Save model every 100 steps. save_embedding(model) save_embedding(model) if __name__ == "__main__": main()
2,306
36.819672
81
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/data/molecules/prepare_molecules.py
#!/usr/bin/env python # coding: utf-8 # # Notebook for preparing and saving MOLECULAR graphs # In[1]: import numpy as np import torch import pickle import time import os from IPython import get_ipython #get_ipython().run_line_magic('matplotlib', 'inline') import matplotlib.pyplot as plt # In[2]: print(torch.__version__) # # Download ZINC dataset # In[1]: #!unzip molecules.zip -d ../ # In[3]: if not os.path.isfile('molecules.zip'): print('downloading..') os.system('curl https://www.dropbox.com/s/feo9qle74kg48gy/molecules.zip?dl=1 -o molecules.zip -J -L -k') os.system('unzip molecules.zip -d ../') # !tar -xvf molecules.zip -C ../ else: print('File already downloaded') # # Convert to DGL format and save with pickle # In[4]: import os os.chdir('../../') # go to root folder of the project print(os.getcwd()) # In[5]: import pickle # get_ipython().run_line_magic('load_ext', 'autoreload') # get_ipython().run_line_magic('autoreload', '2') from data.molecules import MoleculeDatasetDGL ,MoleculeDatasetpyg from data.data import LoadData from torch.utils.data import DataLoader from data.molecules import MoleculeDataset # In[6]: framwork = 'pyg' DATASET_NAME = 'ZINC' dataset = MoleculeDatasetDGL(DATASET_NAME) if 'dgl' == framwork else MoleculeDatasetpyg(DATASET_NAME) # In[7]: def plot_histo_graphs(dataset, title): # histogram of graph sizes graph_sizes = [] for graph in dataset: graph_sizes.append(graph.num_nodes) if framwork == 'pyg' else graph_sizes.append(graph[0].number_of_nodes()) plt.figure(1) plt.hist(graph_sizes, bins=20) plt.title(title) plt.show() graph_sizes = torch.Tensor(graph_sizes) print('min/max :',graph_sizes.min().long().item(),graph_sizes.max().long().item()) #plot_histo_graphs(dataset.train,'trainset') plot_histo_graphs(dataset.val,'valset') plot_histo_graphs(dataset.test,'testset') # In[8]: #print(len(dataset.train)) print(len(dataset.val)) print(len(dataset.test)) #print(dataset.train[0]) print(dataset.val[0]) print(dataset.test[0]) # In[9]: num_atom_type = 28 num_bond_type = 4 # In[10]: # start = time.time() # # with open('data/molecules/ZINC_dgl.pkl','wb') as f: # pickle.dump([dataset.train,dataset.val,dataset.test,num_atom_type,num_bond_type],f) # print('Time (sec):',time.time() - start) # # Test load function # In[11]: # DATASET_NAME = 'ZINC' # dataset = LoadData(DATASET_NAME, framwork) # trainset, valset, testset = dataset.train, dataset.val, dataset.test # In[12]: from torch_geometric.data import DataLoader loader = DataLoader(dataset.val, batch_size=32, shuffle=True) for batch in loader: print(batch) print(batch.y) print(len(batch.y)) # batch_size = 10 # collate = MoleculeDataset.collate # print(MoleculeDataset) # train_loader = DataLoader(trainset, batch_size=batch_size, shuffle=True, collate_fn=collate) # In[ ]: # In[ ]:
2,951
17.110429
116
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/out/log2csv-planetoid.py
import os import re import numpy as np import csv def write2csv(path): # path='Planetoid_node_classification/results/result_GAT_pyg_Citeseer_GPU0_23h12m32s_on_Oct_28_2020.txt' csv_file=open('results.csv','w',encoding='gbk',newline='') csv_writer=csv.writer(csv_file) csv_writer.writerow(['data','model','L','params','train','val','test','epoch']) totals = [] for path in findAllFile(path): print(path) file=open(path) iterf=iter(file) for line in iterf: a = line.find('Dataset:') b = line.find('net_params={\'L\':') c=line.find('Model:') d=line.find('Total Parameters:') e=line.find('TEST ACCURACY averaged:') h = line.find('val ACCURACY averaged:') f=line.find('TRAIN ACCURACY averaged:') g=line.find(' Average Convergence Time (Epochs):') # h=line.find('params={\'seed\':') # print(g) if a == 0: dataset = line[line.index(':') + 2:line.index(',')] if b == 0: net = line[line.index(':') + 2:line.index(',')] if c == 0: model = line[line.index(':')+2:line.index('_')] if d == 0: Parameters = line[line.index(':')+2:line.index('\n')] if e == 0: TEST = line[line.index(':')+2:line.index('w')-1] if h == 0: val = line[line.index(':')+2:line.index('w')-1] if f == 0: TRAIN = line[line.index(':') + 2:line.index('w') - 1] # if h == 0: # seed = line[line.index(':') + 2:line.index(',')] if g == 0: Epochs = line[line.index(':') + 2:line.index('w') - 1] totals.append([dataset, model, net, Parameters, TRAIN, val,TEST, Epochs]) # csv_writer.writerow([dataset, model, net, Parameters, TRAIN, TEST, Epochs]) break totals.sort(key=lambda x: ((x[0]), (x[1]), int(x[2])), reverse=False) out = [] calculate = [] for i in range(totals.__len__()): out.append(totals[i]) csv_writer.writerow(out[i]) if (i+1)%4 == 0: avg_train_acc = np.array(totals[i-3:i+1])[:,4] avg_val_acc = np.array(totals[i-3:i+1])[:,5] avg_test_acc = np.array(totals[i-3:i+1])[:,6] # avg_test_acc [totals[i-4:i][0][4], totals[:4][1][4], totals[:4][2][4], totals[:4][3][4]] avg_epoch = np.array(totals[i-3:i+1])[:,7] train_acc=str(np.around(np.mean(np.array(avg_train_acc, dtype=np.float32)),decimals=4))+'±'+str(np.around(np.std(np.array(avg_train_acc, dtype=np.float32),ddof = 1),decimals=4)) val_acc = str(np.around(np.mean(np.array(avg_val_acc, dtype=np.float32)),decimals=4)) + '±' + str(np.around(np.std(np.array(avg_val_acc, dtype=np.float32), ddof=1),decimals=4)) test_acc= str(np.around(np.mean(np.array(avg_test_acc, dtype=np.float32)),decimals=4))+'±'+str(np.around(np.std(np.array(avg_test_acc, dtype=np.float32),ddof = 1),decimals=4)) Epochs_acc = str(np.around(np.mean(np.array(avg_epoch, dtype=np.float32)),decimals=4))+'±'+str(np.around(np.std(np.array(avg_epoch, dtype=np.float32),ddof = 1),decimals=4)) calculate.append([out[i-1][0], out[i-1][1], out[i-1][2], out[i-1][3], train_acc, val_acc ,test_acc, Epochs_acc]) csv_writer.writerow(calculate[int((i+1)/4-1)]) csv_file.close() file.close() def findAllFile(base): for root, ds, fs in os.walk(base): for f in fs: if f.endswith('.txt'): fullname = os.path.join(root, f) yield fullname def main(): # base = 'Planetoid_node_classification/results/'SBMs_node_classification base = 'SBMs_node_classification/results/' # for path in findAllFile(base): # print(path) np.set_printoptions(precision=4) write2csv(base) if __name__ == '__main__': main()
4,018
45.732558
189
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/out/log2csv-sbm.py
import os import re import numpy as np import csv def write2csv(path): # path='Planetoid_node_classification/results/result_GAT_pyg_Citeseer_GPU0_23h12m32s_on_Oct_28_2020.txt' csv_file=open('results.csv','w',encoding='gbk',newline='') csv_writer=csv.writer(csv_file) csv_writer.writerow(['data','model','L','params','train','val','test','epoch']) totals = [] for path in findAllFile(path): print(path) file=open(path) iterf=iter(file) for line in iterf: a = line.find('Dataset:') b = line.find('net_params={\'L\':') c=line.find('Model:') d=line.find('Total Parameters:') e=line.find('TEST ACCURACY') h = line.find('val ACCURACY') f=line.find('TRAIN ACCURACY') g=line.find(' Convergence Time (Epochs):') # h=line.find('params={\'seed\':') # print(g) if a == 0: dataset = line[line.index(':') + 2:line.index(',')] if b == 0: net = line[line.index(':') + 2:line.index(',')] if c == 0: model = line[line.index(':')+2:line.index('_')] if d == 0: Parameters = line[line.index(':')+2:line.index('\n')] if e == 0: TEST = line[line.index(':')+2:line.index('\n')] if h == 0: val = line[line.index(':')+2:line.index('\n')] if f == 0: TRAIN = line[line.index(':') + 2:line.index('\n')] # if h == 0: # seed = line[line.index(':') + 2:line.index(',')] if g == 0: Epochs = line[line.index(':') + 2:line.index('\n')] totals.append([dataset, model, net, Parameters, TRAIN, val,TEST, Epochs]) # csv_writer.writerow([dataset, model, net, Parameters, TRAIN, TEST, Epochs]) break totals.sort(key=lambda x: ((x[0]), (x[1]), int(x[2])), reverse=False) out = [] calculate = [] for i in range(totals.__len__()): out.append(totals[i]) csv_writer.writerow(out[i]) if (i+1)%4 == 0: avg_train_acc = np.array(totals[i-3:i+1])[:,4] avg_val_acc = np.array(totals[i-3:i+1])[:,5] avg_test_acc = np.array(totals[i-3:i+1])[:,6] # avg_test_acc [totals[i-4:i][0][4], totals[:4][1][4], totals[:4][2][4], totals[:4][3][4]] avg_epoch = np.array(totals[i-3:i+1])[:,7] train_acc=str(np.around(np.mean(np.array(avg_train_acc, dtype=np.float32)),decimals=4))+'±'+str(np.around(np.std(np.array(avg_train_acc, dtype=np.float32),ddof = 1),decimals=4)) val_acc = str(np.around(np.mean(np.array(avg_val_acc, dtype=np.float32)),decimals=4)) + '±' + str(np.around(np.std(np.array(avg_val_acc, dtype=np.float32), ddof=1),decimals=4)) test_acc= str(np.around(np.mean(np.array(avg_test_acc, dtype=np.float32)),decimals=4))+'±'+str(np.around(np.std(np.array(avg_test_acc, dtype=np.float32),ddof = 1),decimals=4)) Epochs_acc = str(np.around(np.mean(np.array(avg_epoch, dtype=np.float32)),decimals=4))+'±'+str(np.around(np.std(np.array(avg_epoch, dtype=np.float32),ddof = 1),decimals=4)) calculate.append([out[i-1][0], out[i-1][1], out[i-1][2], out[i-1][3], train_acc, val_acc ,test_acc, Epochs_acc]) csv_writer.writerow(calculate[int((i+1)/4-1)]) csv_file.close() file.close() def findAllFile(base): for root, ds, fs in os.walk(base): for f in fs: if f.endswith('.txt'): fullname = os.path.join(root, f) yield fullname def main(): # base = 'Planetoid_node_classification/results/'SBMs_node_classification base = 'SBMs_node_classification/results/' # for path in findAllFile(base): # print(path) np.set_printoptions(precision=4) write2csv(base) if __name__ == '__main__': main()
3,972
45.197674
189
py
benchmarking-gnns-pyg
benchmarking-gnns-pyg-master/out/log2csv-ogb.py
import os import re import numpy as np import csv def write2csv(path): # path='Planetoid_node_classification/results/result_GAT_pyg_Citeseer_GPU0_23h12m32s_on_Oct_28_2020.txt' csv_file=open('results.csv','w',encoding='gbk',newline='') csv_writer=csv.writer(csv_file) csv_writer.writerow(['data','model','L','params','train','val','test','epoch']) totals = [] for path in findAllFile(path): print(path) file=open(path) iterf=iter(file) for line in iterf: a = line.find('Dataset:') b = line.find('net_params={\'L\':') c=line.find('Model:') d=line.find('Total Parameters:') e=line.find('TEST ACCURACY:') h=line.find('val ACCURACY:') f=line.find('TRAIN ACCURACY:') g=line.find(' Convergence Time (Epochs):') # h=line.find('params={\'seed\':') # print(g) if a == 0: dataset = line[line.index(':') + 2:line.index(',')] if b == 0: net = line[line.index(':') + 2:line.index(',')] if c == 0: model = line[line.index(':')+2:line.index('_')] if d == 0: Parameters = line[line.index(':')+2:line.index('\n')] if e == 0: TEST = line[line.index(':')+2:line.index('\n')] if h == 0: val = line[line.index(':')+2:line.index('\n')] if f == 0: TRAIN = line[line.index(':') + 2:line.index('\n')] # if h == 0: # seed = line[line.index(':') + 2:line.index(',')] if g == 0: Epochs = line[line.index(':') + 2:line.index('\n')] totals.append([dataset, model, net, Parameters, TRAIN, val, TEST, Epochs]) # csv_writer.writerow([dataset, model, net, Parameters, TRAIN, TEST, Epochs]) break totals.sort(key=lambda x: ((x[0]), (x[1]), int(x[2])), reverse=False) out = [] calculate = [] for i in range(totals.__len__()): out.append(totals[i]) csv_writer.writerow(out[i]) if (i+1)%4 == 0: avg_train_acc = np.array(totals[i-3:i+1])[:,4] avg_val_acc = np.array(totals[i-3:i+1])[:,5] avg_test_acc = np.array(totals[i - 3:i + 1])[:, 6] # avg_test_acc [totals[i-4:i][0][4], totals[:4][1][4], totals[:4][2][4], totals[:4][3][4]] avg_epoch = np.array(totals[i-3:i+1])[:,7] # train_acc=str(np.around((np.mean(np.array(avg_train_acc, dtype=np.float32)))*100,decimals=4))+'±'+str(np.around((np.std(np.array(avg_train_acc, dtype=np.float32),ddof = 1))*100,decimals=4)) # val_acc = str(np.around((np.mean(np.array(avg_val_acc, dtype=np.float32)))*100,decimals=4)) + '±' + str(np.around((np.std(np.array(avg_val_acc, dtype=np.float32), ddof=1))*100,decimals=4)) # test_acc= str(np.around((np.mean(np.array(avg_test_acc, dtype=np.float32)))*100,decimals=4))+'±'+str(np.around((np.std(np.array(avg_test_acc, dtype=np.float32),ddof = 1))*100,decimals=4)) train_acc = str( np.around(np.mean(np.array(avg_train_acc, dtype=np.float32)), decimals=4)) + '±' + str( np.around(np.std(np.array(avg_train_acc, dtype=np.float32), ddof=1), decimals=4)) val_acc = str(np.around(np.mean(np.array(avg_val_acc, dtype=np.float32)), decimals=4)) + '±' + str( np.around(np.std(np.array(avg_val_acc, dtype=np.float32), ddof=1), decimals=4)) test_acc = str( np.around(np.mean(np.array(avg_test_acc, dtype=np.float32)), decimals=4)) + '±' + str( np.around(np.std(np.array(avg_test_acc, dtype=np.float32), ddof=1), decimals=4)) Epochs_acc = str(np.around(np.mean(np.array(avg_epoch, dtype=np.float32)),decimals=4))+'±'+str(np.around(np.std(np.array(avg_epoch, dtype=np.float32),ddof = 1),decimals=4)) calculate.append([out[i-1][0], out[i-1][1], out[i-1][2], out[i-1][3], train_acc, val_acc,test_acc , Epochs_acc]) csv_writer.writerow(calculate[int((i+1)/4-1)]) csv_file.close() file.close() def findAllFile(base): for root, ds, fs in os.walk(base): for f in fs: if f.endswith('.txt'): fullname = os.path.join(root, f) yield fullname def main(): # base = 'ogb_node_classification/results/' base = 'out-input/posenc/results/' # / home / karl / works / syncbenchmarking - gnns / out / out - input / node_embedding / results # for path in findAllFile(base): # print(path) np.set_printoptions(precision=4) write2csv(base) if __name__ == '__main__': main()
4,751
49.021053
203
py
SleePyCo
SleePyCo-main/train_mtcl.py
import os import json import argparse import warnings import torch import torch.nn as nn import torch.optim as optim from torch.utils.data import DataLoader from utils import * from loader import EEGDataLoader from models.main_model import MainModel class OneFoldTrainer: def __init__(self, args, fold, config): self.args = args self.fold = fold self.cfg = config self.ds_cfg = config['dataset'] self.fp_cfg = config['feature_pyramid'] self.tp_cfg = config['training_params'] self.es_cfg = self.tp_cfg['early_stopping'] self.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') print('[INFO] Config name: {}'.format(config['name'])) self.train_iter = 0 self.model = self.build_model() self.loader_dict = self.build_dataloader() self.criterion = nn.CrossEntropyLoss() self.activate_train_mode() self.optimizer = optim.Adam([p for p in self.model.parameters() if p.requires_grad], lr=self.tp_cfg['lr'], weight_decay=self.tp_cfg['weight_decay']) self.ckpt_path = os.path.join('checkpoints', config['name']) self.ckpt_name = 'ckpt_fold-{0:02d}.pth'.format(self.fold) self.early_stopping = EarlyStopping(patience=self.es_cfg['patience'], verbose=True, ckpt_path=self.ckpt_path, ckpt_name=self.ckpt_name, mode=self.es_cfg['mode']) def build_model(self): model = MainModel(self.cfg) print('[INFO] Number of params of model: ', sum(p.numel() for p in model.parameters() if p.requires_grad)) model = torch.nn.DataParallel(model, device_ids=list(range(len(self.args.gpu.split(","))))) if self.tp_cfg['mode'] != 'scratch': print('[INFO] Model loaded for finetune') load_name = self.cfg['name'].replace('SL-{:02d}'.format(self.ds_cfg['seq_len']), 'SL-01') load_name = load_name.replace('numScales-{}'.format(self.fp_cfg['num_scales']), 'numScales-1') load_name = load_name.replace(self.tp_cfg['mode'], 'pretrain') load_path = os.path.join('checkpoints', load_name, 'ckpt_fold-{0:02d}.pth'.format(self.fold)) model.load_state_dict(torch.load(load_path), strict=False) model.to(self.device) print('[INFO] Model prepared, Device used: {} GPU:{}'.format(self.device, self.args.gpu)) return model def build_dataloader(self): train_dataset = EEGDataLoader(self.cfg, self.fold, set='train') train_loader = DataLoader(dataset=train_dataset, batch_size=self.tp_cfg['batch_size'], shuffle=True, num_workers=4*len(self.args.gpu.split(",")), pin_memory=True) val_dataset = EEGDataLoader(self.cfg, self.fold, set='val') val_loader = DataLoader(dataset=val_dataset, batch_size=self.tp_cfg['batch_size'], shuffle=False, num_workers=4*len(self.args.gpu.split(",")), pin_memory=True) test_dataset = EEGDataLoader(self.cfg, self.fold, set='test') test_loader = DataLoader(dataset=test_dataset, batch_size=self.tp_cfg['batch_size'], shuffle=False, num_workers=4*len(self.args.gpu.split(",")), pin_memory=True) print('[INFO] Dataloader prepared') return {'train': train_loader, 'val': val_loader, 'test': test_loader} def activate_train_mode(self): self.model.train() if self.tp_cfg['mode'] == 'freezefinetune': print('[INFO] Freeze backone') self.model.module.feature.train(False) for p in self.model.module.feature.parameters(): p.requires_grad = False print('[INFO] Unfreeze conv_c5') self.model.module.feature.conv_c5.train(True) for p in self.model.module.feature.conv_c5.parameters(): p.requires_grad = True if self.fp_cfg['num_scales'] > 1: print('[INFO] Unfreeze conv_c4') self.model.module.feature.conv_c4.train(True) for p in self.model.module.feature.conv_c4.parameters(): p.requires_grad = True if self.fp_cfg['num_scales'] > 2: print('[INFO] Unfreeze conv_c3') self.model.module.feature.conv_c3.train(True) for p in self.model.module.feature.conv_c3.parameters(): p.requires_grad = True def train_one_epoch(self, epoch): correct, total, train_loss = 0, 0, 0 for i, (inputs, labels) in enumerate(self.loader_dict['train']): loss = 0 total += labels.size(0) inputs = inputs.to(self.device) labels = labels.view(-1).to(self.device) outputs = self.model(inputs) outputs_sum = torch.zeros_like(outputs[0]) for j in range(len(outputs)): loss += self.criterion(outputs[j], labels) outputs_sum += outputs[j] self.optimizer.zero_grad() loss.backward() self.optimizer.step() train_loss += loss.item() predicted = torch.argmax(outputs_sum, 1) correct += predicted.eq(labels).sum().item() self.train_iter += 1 progress_bar(i, len(self.loader_dict['train']), 'Loss: %.3f | Acc: %.3f%% (%d/%d)' % (train_loss / (i + 1), 100. * correct / total, correct, total)) if self.train_iter % self.tp_cfg['val_period'] == 0: print('') val_acc, val_loss = self.evaluate(mode='val') self.early_stopping(val_acc, val_loss, self.model) self.activate_train_mode() if self.early_stopping.early_stop: break @torch.no_grad() def evaluate(self, mode): self.model.eval() correct, total, eval_loss = 0, 0, 0 y_true = np.zeros(0) y_pred = np.zeros((0, self.cfg['classifier']['num_classes'])) for i, (inputs, labels) in enumerate(self.loader_dict[mode]): loss = 0 total += labels.size(0) inputs = inputs.to(self.device) labels = labels.view(-1).to(self.device) outputs = self.model(inputs) outputs_sum = torch.zeros_like(outputs[0]) for j in range(len(outputs)): loss += self.criterion(outputs[j], labels) outputs_sum += outputs[j] eval_loss += loss.item() predicted = torch.argmax(outputs_sum, 1) correct += predicted.eq(labels).sum().item() y_true = np.concatenate([y_true, labels.cpu().numpy()]) y_pred = np.concatenate([y_pred, outputs_sum.cpu().numpy()]) progress_bar(i, len(self.loader_dict[mode]), 'Loss: %.3f | Acc: %.3f%% (%d/%d)' % (eval_loss / (i + 1), 100. * correct / total, correct, total)) if mode == 'val': return 100. * correct / total, eval_loss elif mode == 'test': return y_true, y_pred else: raise NotImplementedError def run(self): for epoch in range(self.tp_cfg['max_epochs']): print('\n[INFO] Fold: {}, Epoch: {}'.format(self.fold, epoch)) self.train_one_epoch(epoch) if self.early_stopping.early_stop: break self.model.load_state_dict(torch.load(os.path.join(self.ckpt_path, self.ckpt_name))) y_true, y_pred = self.evaluate(mode='test') print('') return y_true, y_pred def main(): warnings.filterwarnings("ignore", category=DeprecationWarning) warnings.filterwarnings("ignore", category=UserWarning) parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter) parser.add_argument('--seed', type=int, default=42, help='random seed') parser.add_argument('--gpu', type=str, default="0", help='gpu id') parser.add_argument('--config', type=str, help='config file path') args = parser.parse_args() os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID" os.environ["CUDA_VISIBLE_DEVICES"] = args.gpu # For reproducibility set_random_seed(args.seed, use_cuda=True) with open(args.config) as config_file: config = json.load(config_file) config['name'] = os.path.basename(args.config).replace('.json', '') Y_true = np.zeros(0) Y_pred = np.zeros((0, config['classifier']['num_classes'])) for fold in range(1, config['dataset']['num_splits'] + 1): trainer = OneFoldTrainer(args, fold, config) y_true, y_pred = trainer.run() Y_true = np.concatenate([Y_true, y_true]) Y_pred = np.concatenate([Y_pred, y_pred]) summarize_result(config, fold, Y_true, Y_pred) if __name__ == "__main__": main()
8,872
40.853774
170
py
SleePyCo
SleePyCo-main/test.py
import os import json import argparse import warnings import torch import torch.nn as nn from torch.utils.data import DataLoader from utils import * from loader import EEGDataLoader from train_mtcl import OneFoldTrainer from models.main_model import MainModel class OneFoldEvaluator(OneFoldTrainer): def __init__(self, args, fold, config): self.args = args self.fold = fold self.cfg = config self.ds_cfg = config['dataset'] self.tp_cfg = config['training_params'] self.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') print('[INFO] Config name: {}'.format(config['name'])) self.model = self.build_model() self.loader_dict = self.build_dataloader() self.criterion = nn.CrossEntropyLoss() self.ckpt_path = os.path.join('checkpoints', config['name']) self.ckpt_name = 'ckpt_fold-{0:02d}.pth'.format(self.fold) def build_model(self): model = MainModel(self.cfg) print('[INFO] Number of params of model: ', sum(p.numel() for p in model.parameters() if p.requires_grad)) model = torch.nn.DataParallel(model, device_ids=list(range(len(self.args.gpu.split(","))))) model.to(self.device) print('[INFO] Model prepared, Device used: {} GPU:{}'.format(self.device, self.args.gpu)) return model def build_dataloader(self): test_dataset = EEGDataLoader(self.cfg, self.fold, set='test') test_loader = DataLoader(dataset=test_dataset, batch_size=self.tp_cfg['batch_size'], shuffle=False, num_workers=4*len(self.args.gpu.split(",")), pin_memory=True) print('[INFO] Dataloader prepared') return {'test': test_loader} def run(self): print('\n[INFO] Fold: {}'.format(self.fold)) self.model.load_state_dict(torch.load(os.path.join(self.ckpt_path, self.ckpt_name))) y_true, y_pred = self.evaluate(mode='test') print('') return y_true, y_pred def main(): warnings.filterwarnings("ignore", category=DeprecationWarning) warnings.filterwarnings("ignore", category=UserWarning) parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter) parser.add_argument('--seed', type=int, default=42, help='random seed') parser.add_argument('--gpu', type=str, default="0", help='gpu id') parser.add_argument('--config', type=str, help='config file path') args = parser.parse_args() os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID" os.environ["CUDA_VISIBLE_DEVICES"] = args.gpu with open(args.config) as config_file: config = json.load(config_file) config['name'] = os.path.basename(args.config).replace('.json', '') Y_true = np.zeros(0) Y_pred = np.zeros((0, config['classifier']['num_classes'])) for fold in range(1, config['dataset']['num_splits'] + 1): evaluator = OneFoldEvaluator(args, fold, config) y_true, y_pred = evaluator.run() Y_true = np.concatenate([Y_true, y_true]) Y_pred = np.concatenate([Y_pred, y_pred]) summarize_result(config, fold, Y_true, Y_pred) if __name__ == "__main__": main()
3,233
34.933333
169
py
SleePyCo
SleePyCo-main/train_crl.py
import os import json import argparse import warnings import torch import torch.optim as optim from torch.utils.data import DataLoader from utils import * from loss import SupConLoss from loader import EEGDataLoader from models.main_model import MainModel class OneFoldTrainer: def __init__(self, args, fold, config): self.args = args self.fold = fold self.cfg = config self.tp_cfg = config['training_params'] self.es_cfg = self.tp_cfg['early_stopping'] self.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') print('[INFO] Config name: {}'.format(config['name'])) self.train_iter = 0 self.model = self.build_model() self.loader_dict = self.build_dataloader() self.criterion = SupConLoss(temperature=self.tp_cfg['temperature']) self.optimizer = optim.Adam(self.model.parameters(), lr=self.tp_cfg['lr'], weight_decay=self.tp_cfg['weight_decay']) self.ckpt_path = os.path.join('checkpoints', config['name']) self.ckpt_name = 'ckpt_fold-{0:02d}.pth'.format(self.fold) self.early_stopping = EarlyStopping(patience=self.es_cfg['patience'], verbose=True, ckpt_path=self.ckpt_path, ckpt_name=self.ckpt_name, mode=self.es_cfg['mode']) def build_model(self): model = MainModel(self.cfg) print('[INFO] Number of params of model: ', sum(p.numel() for p in model.parameters() if p.requires_grad)) model = torch.nn.DataParallel(model, device_ids=list(range(len(self.args.gpu.split(","))))) model.to(self.device) print('[INFO] Model prepared, Device used: {} GPU:{}'.format(self.device, self.args.gpu)) return model def build_dataloader(self): dataloader_args = {'batch_size': self.tp_cfg['batch_size'], 'shuffle': True, 'num_workers': 4*len(self.args.gpu.split(",")), 'pin_memory': True} train_dataset = EEGDataLoader(self.cfg, self.fold, set='train') train_loader = DataLoader(dataset=train_dataset, **dataloader_args) val_dataset = EEGDataLoader(self.cfg, self.fold, set='val') val_loader = DataLoader(dataset=val_dataset, **dataloader_args) print('[INFO] Dataloader prepared') return {'train': train_loader, 'val': val_loader} def train_one_epoch(self): self.model.train() train_loss = 0 for i, (inputs, labels) in enumerate(self.loader_dict['train']): loss = 0 labels = labels.view(-1).to(self.device) inputs = torch.cat([inputs[0], inputs[1]], dim=0).to(self.device) outputs = self.model(inputs)[0] f1, f2 = torch.split(outputs, [labels.size(0), labels.size(0)], dim=0) features = torch.cat([f1.unsqueeze(1), f2.unsqueeze(1)], dim=1) loss += self.criterion(features, labels) self.optimizer.zero_grad() loss.backward() self.optimizer.step() train_loss += loss.item() self.train_iter += 1 progress_bar(i, len(self.loader_dict['train']), 'Lr: %.4e | Loss: %.3f' %(get_lr(self.optimizer), train_loss / (i + 1))) if self.train_iter % self.tp_cfg['val_period'] == 0: print('') val_loss = self.evaluate(mode='val') self.early_stopping(None, val_loss, self.model) self.model.train() if self.early_stopping.early_stop: break @torch.no_grad() def evaluate(self, mode): self.model.eval() eval_loss = 0 for i, (inputs, labels) in enumerate(self.loader_dict[mode]): loss = 0 inputs = inputs.to(self.device) labels = labels.view(-1).to(self.device) outputs = self.model(inputs)[0] features = outputs.unsqueeze(1).repeat(1, 2, 1) loss += self.criterion(features, labels) eval_loss += loss.item() progress_bar(i, len(self.loader_dict[mode]), 'Lr: %.4e | Loss: %.3f' %(get_lr(self.optimizer), eval_loss / (i + 1))) return eval_loss def run(self): for epoch in range(self.tp_cfg['max_epochs']): print('\n[INFO] Fold: {}, Epoch: {}'.format(self.fold, epoch)) self.train_one_epoch() if self.early_stopping.early_stop: break def main(): warnings.filterwarnings("ignore", category=DeprecationWarning) warnings.filterwarnings("ignore", category=UserWarning) parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter) parser.add_argument('--seed', type=int, default=42, help='random seed') parser.add_argument('--gpu', type=str, default="0", help='gpu id') parser.add_argument('--config', type=str, help='config file path') args = parser.parse_args() os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID" os.environ["CUDA_VISIBLE_DEVICES"] = args.gpu # For reproducibility set_random_seed(args.seed, use_cuda=True) with open(args.config) as config_file: config = json.load(config_file) config['name'] = os.path.basename(args.config).replace('.json', '') for fold in range(1, config['dataset']['num_splits'] + 1): trainer = OneFoldTrainer(args, fold, config) trainer.run() if __name__ == "__main__": main()
5,454
36.62069
169
py
SleePyCo
SleePyCo-main/transform.py
import torch import random import numpy as np from scipy import signal from scipy.ndimage.interpolation import shift class TwoTransform: def __init__(self, transform): self.transform = transform def __call__(self, x): return [self.transform(x), self.transform(x)] class Compose: def __init__(self, transforms, mode='full'): self.transforms = transforms self.mode = mode def __call__(self, x): if self.mode == 'random': index = random.randint(0, len(self.transforms) - 1) x = self.transforms[index](x) elif self.mode == 'full': for t in self.transforms: x = t(x) elif self.mode == 'shuffle': transforms = np.random.choice(self.transforms, len(self.transforms), replace=False) for t in transforms: x = t(x) else: raise NotImplementedError return x def __repr__(self): format_string = self.__class__.__name__ + '(' for t in self.transforms: format_string += '\n' format_string += ' {0}'.format(t) format_string += '\n)' return format_string class RandomAmplitudeScale: def __init__(self, range=(0.5, 2.0), p=0.5): self.range = range self.p = p def __call__(self, x): if torch.rand(1) < self.p: scale = random.uniform(self.range[0], self.range[1]) return x * scale return x def __repr__(self): return self.__class__.__name__ + '()' class RandomDCShift: def __init__(self, range=(-10.0, 10.0), p=0.5): self.range = range self.p = p def __call__(self, x): if torch.rand(1) < self.p: shift = random.uniform(self.range[0], self.range[1]) return x + shift return x def __repr__(self): return self.__class__.__name__ + '()' class RandomTimeShift: def __init__(self, range=(-300, 300), mode='constant', cval=0.0, p=0.5): self.range = range self.mode = mode self.cval = cval self.p = p def __call__(self, x): if torch.rand(1) < self.p: t_shift = random.randint(self.range[0], self.range[1]) if len(x.shape) == 2: x = x[0] x = shift(input=x, shift=t_shift, mode=self.mode, cval=self.cval) x = np.expand_dims(x, axis=0) return x def __repr__(self): return self.__class__.__name__ + '()' class RandomZeroMasking: def __init__(self, range=(0, 300), p=0.5): self.range = range self.p = p def __call__(self, x): if torch.rand(1) < self.p: mask_len = random.randint(self.range[0], self.range[1]) random_pos = random.randint(0, x.shape[1] - mask_len) mask = np.concatenate([np.ones((1, random_pos)), np.zeros((1, mask_len)), np.ones((1, x.shape[1] - mask_len - random_pos))], axis=1) return x * mask return x def __repr__(self): return self.__class__.__name__ + '()' class RandomAdditiveGaussianNoise: def __init__(self, range=(0.0, 0.2), p=0.5): self.range = range self.p = p def __call__(self, x): if torch.rand(1) < self.p: sigma = random.uniform(self.range[0], self.range[1]) return x + np.random.normal(0, sigma, x.shape) return x def __repr__(self): return self.__class__.__name__ + '()' class RandomBandStopFilter: def __init__(self, range=(0.5, 30.0), band_width=2.0, sampling_rate=100.0, p=0.5): self.range = range self.band_width = band_width self.sampling_rate = sampling_rate self.p = p def __call__(self, x): if torch.rand(1) < self.p: low_freq = random.uniform(self.range[0], self.range[1]) center_freq = low_freq + self.band_width / 2.0 b, a = signal.iirnotch(center_freq, center_freq / self.band_width, fs=self.sampling_rate) x = signal.lfilter(b, a, x) return x def __repr__(self): return self.__class__.__name__ + '()'
4,204
26.48366
144
py
SleePyCo
SleePyCo-main/loss.py
import torch import torch.nn as nn class SupConLoss(nn.Module): """Supervised Contrastive Learning: https://arxiv.org/pdf/2004.11362.pdf. It also supports the unsupervised contrastive loss in SimCLR""" def __init__(self, temperature=0.07, contrast_mode='all', base_temperature=0.07): super(SupConLoss, self).__init__() self.temperature = temperature self.contrast_mode = contrast_mode self.base_temperature = base_temperature def forward(self, features, labels=None, mask=None): """Compute loss for model. If both `labels` and `mask` are None, it degenerates to SimCLR unsupervised loss: https://arxiv.org/pdf/2002.05709.pdf Args: features: hidden vector of shape [bsz, n_views, ...]. labels: ground truth of shape [bsz]. mask: contrastive mask of shape [bsz, bsz], mask_{i,j}=1 if sample j has the same class as sample i. Can be asymmetric. Returns: A loss scalar. """ device = (torch.device('cuda') if features.is_cuda else torch.device('cpu')) if len(features.shape) < 3: raise ValueError('`features` needs to be [bsz, n_views, ...],' 'at least 3 dimensions are required') if len(features.shape) > 3: features = features.view(features.shape[0], features.shape[1], -1) batch_size = features.shape[0] if labels is not None and mask is not None: raise ValueError('Cannot define both `labels` and `mask`') elif labels is None and mask is None: mask = torch.eye(batch_size, dtype=torch.float32).to(device) elif labels is not None: labels = labels.contiguous().view(-1, 1) if labels.shape[0] != batch_size: raise ValueError('Num of labels does not match num of features') mask = torch.eq(labels, labels.T).float().to(device) else: mask = mask.float().to(device) contrast_count = features.shape[1] contrast_feature = torch.cat(torch.unbind(features, dim=1), dim=0) if self.contrast_mode == 'one': anchor_feature = features[:, 0] anchor_count = 1 elif self.contrast_mode == 'all': anchor_feature = contrast_feature anchor_count = contrast_count else: raise ValueError('Unknown mode: {}'.format(self.contrast_mode)) # compute logits anchor_dot_contrast = torch.div( torch.matmul(anchor_feature, contrast_feature.T), self.temperature) # for numerical stability logits_max, _ = torch.max(anchor_dot_contrast, dim=1, keepdim=True) logits = anchor_dot_contrast - logits_max.detach() # tile mask mask = mask.repeat(anchor_count, contrast_count) # mask-out self-contrast cases logits_mask = torch.scatter( torch.ones_like(mask), 1, torch.arange(batch_size * anchor_count).view(-1, 1).to(device), 0 ) mask = mask * logits_mask # compute log_prob exp_logits = torch.exp(logits) * logits_mask log_prob = logits - torch.log(exp_logits.sum(1, keepdim=True)) # compute mean of log-likelihood over positive mean_log_prob_pos = (mask * log_prob).sum(1) / mask.sum(1) # loss loss = - (self.temperature / self.base_temperature) * mean_log_prob_pos loss = loss.view(anchor_count, batch_size).mean() return loss
3,650
38.684783
80
py
SleePyCo
SleePyCo-main/utils.py
import os import sys import math import time import torch import random import numpy as np import sklearn.metrics as skmet from terminaltables import SingleTable from termcolor import colored _, term_width = os.popen('stty size', 'r').read().split() term_width = int(term_width) TOTAL_BAR_LENGTH = 25. last_time = time.time() begin_time = last_time def progress_bar(current, total, msg=None): global last_time, begin_time if current == 0: begin_time = time.time() # Reset for new bar. cur_len = int(TOTAL_BAR_LENGTH*current/total) rest_len = int(TOTAL_BAR_LENGTH - cur_len) - 1 sys.stdout.write(' [') for i in range(cur_len): sys.stdout.write('=') sys.stdout.write('>') for i in range(rest_len): sys.stdout.write('.') sys.stdout.write(']') cur_time = time.time() step_time = cur_time - last_time last_time = cur_time tot_time = cur_time - begin_time L = [] L.append(' Step: %s' % format_time(step_time)) L.append(' | Tot: %s' % format_time(tot_time)) if msg: L.append(' | ' + msg) msg = ''.join(L) sys.stdout.write(msg) for i in range(term_width-int(TOTAL_BAR_LENGTH)-len(msg)-3): sys.stdout.write(' ') # Go back to the center of the bar. for i in range(term_width-int(TOTAL_BAR_LENGTH/2)+2): sys.stdout.write('\b') sys.stdout.write(' %d/%d ' % (current+1, total)) if current < total-1: sys.stdout.write('\r') else: sys.stdout.write('\n') sys.stdout.flush() def format_time(seconds): days = int(seconds / 3600/24) seconds = seconds - days*3600*24 hours = int(seconds / 3600) seconds = seconds - hours*3600 minutes = int(seconds / 60) seconds = seconds - minutes*60 secondsf = int(seconds) seconds = seconds - secondsf millis = int(seconds*1000) f = '' i = 1 if days > 0: f += str(days) + 'D' i += 1 if hours > 0 and i <= 2: f += str(hours) + 'h' i += 1 if minutes > 0 and i <= 2: f += str(minutes) + 'm' i += 1 if secondsf > 0 and i <= 2: f += str(secondsf) + 's' i += 1 if millis > 0 and i <= 2: f += str(millis) + 'ms' i += 1 if f == '': f = '0ms' return f def get_lr(optimizer): for param_group in optimizer.param_groups: return param_group['lr'] class EarlyStopping: """Early stops the training if validation loss doesn't improve after a given patience.""" def __init__(self, patience=7, verbose=False, delta=0, ckpt_path='./checkpoints', ckpt_name='checkpoint.pth', mode='min'): """ Args: patience (int): How long to wait after last time validation loss improved. Default: 7 verbose (bool): If True, prints a message for each validation loss improvement. Default: False delta (float): Minimum change in the monitored quantity to qualify as an improvement. Default: 0 path (str): Path for the checkpoint to be saved to. Default: 'checkpoint.pt' """ self.patience = patience self.verbose = verbose self.counter = 0 self.best_score = None self.early_stop = False self.mode = mode if mode == 'max': self.init_metric = 0 elif mode == 'min': self.init_metric = -np.inf else: raise NotImplementedError self.delta = delta self.ckpt_path = ckpt_path self.ckpt_name = ckpt_name if '.pth' in ckpt_name else ckpt_name + '.pth' os.makedirs(self.ckpt_path, exist_ok=True) def __call__(self, val_acc, val_loss, model): if self.mode == 'max': score = val_acc val_metric = val_acc elif self.mode == 'min': score = -val_loss val_metric = val_loss else: raise NotImplementedError if self.best_score is None: self.best_score = score self.save_checkpoint(val_metric, model) elif score < self.best_score + self.delta: self.counter += 1 print(f'EarlyStopping counter: {self.counter} out of {self.patience}\n') if self.counter >= self.patience: self.early_stop = True else: self.best_score = score self.save_checkpoint(val_metric, model) self.counter = 0 def save_checkpoint(self, val_metric, model): '''Saves model when validation loss decrease.''' if self.verbose: if self.mode == 'max': print(f'[INFO] Validation accuracy increased ({self.init_metric:.6f} --> {val_metric:.6f}). Saving model ...\n') elif self.mode == 'min': print(f'[INFO] Validation loss decreased ({self.init_metric:.6f} --> {val_metric:.6f}). Saving model ...\n') else: raise NotImplementedError torch.save(model.state_dict(), os.path.join(self.ckpt_path, self.ckpt_name)) self.init_metric = val_metric def summarize_result(config, fold, y_true, y_pred, save=True): os.makedirs('results', exist_ok=True) y_pred_argmax = np.argmax(y_pred, 1) result_dict = skmet.classification_report(y_true, y_pred_argmax, digits=3, output_dict=True) cm = skmet.confusion_matrix(y_true, y_pred_argmax) accuracy = round(result_dict['accuracy']*100, 1) macro_f1 = round(result_dict['macro avg']['f1-score']*100, 1) kappa = round(skmet.cohen_kappa_score(y_true, y_pred_argmax), 3) wpr = round(result_dict['0.0']['precision']*100, 1) wre = round(result_dict['0.0']['recall']*100, 1) wf1 = round(result_dict['0.0']['f1-score']*100, 1) n1pr = round(result_dict['1.0']['precision']*100, 1) n1re = round(result_dict['1.0']['recall']*100, 1) n1f1 = round(result_dict['1.0']['f1-score']*100, 1) n2pr = round(result_dict['2.0']['precision']*100, 1) n2re = round(result_dict['2.0']['recall']*100, 1) n2f1 = round(result_dict['2.0']['f1-score']*100, 1) n3pr = round(result_dict['3.0']['precision']*100, 1) n3re = round(result_dict['3.0']['recall']*100, 1) n3f1 = round(result_dict['3.0']['f1-score']*100, 1) rpr = round(result_dict['4.0']['precision']*100, 1) rre = round(result_dict['4.0']['recall']*100, 1) rf1 = round(result_dict['4.0']['f1-score']*100, 1) overall_data = [ ['ACC', 'MF1', '\u03BA'], [accuracy, macro_f1, kappa], ] perclass_data = [ [colored('A', 'cyan') + '\\' + colored('P', 'green'), 'W', 'N1', 'N2', 'N3', 'R', 'PR', 'RE', 'F1'], ['W', cm[0][0], cm[0][1], cm[0][2], cm[0][3], cm[0][4], wpr, wre, wf1], ['N1', cm[1][0], cm[1][1], cm[1][2], cm[1][3], cm[1][4], n1pr, n1re, n1f1], ['N2', cm[2][0], cm[2][1], cm[2][2], cm[2][3], cm[2][4], n2pr, n2re, n2f1], ['N3', cm[3][0], cm[3][1], cm[3][2], cm[3][3], cm[3][4], n3pr, n3re, n3f1], ['R', cm[4][0], cm[4][1], cm[4][2], cm[4][3], cm[4][4], rpr, rre, rf1], ] overall_dt = SingleTable(overall_data, colored('OVERALL RESULT', 'red')) perclass_dt = SingleTable(perclass_data, colored('PER-CLASS RESULT', 'red')) print('\n[INFO] Evaluation result from fold 1 to {}'.format(fold)) print('\n' + overall_dt.table) print('\n' + perclass_dt.table) print(colored(' A', 'cyan') + ': Actual Class, ' + colored('P', 'green') + ': Predicted Class' + '\n\n') if save: with open(os.path.join('results', config['name'] + '.txt'), 'w') as f: f.write( str(fold) + ' ' + str(round(result_dict['accuracy']*100, 1)) + ' ' + str(round(result_dict['macro avg']['f1-score']*100, 1)) + ' ' + str(round(kappa, 3)) + ' ' + str(round(result_dict['0.0']['f1-score']*100, 1)) + ' ' + str(round(result_dict['1.0']['f1-score']*100, 1)) + ' ' + str(round(result_dict['2.0']['f1-score']*100, 1)) + ' ' + str(round(result_dict['3.0']['f1-score']*100, 1)) + ' ' + str(round(result_dict['4.0']['f1-score']*100, 1)) + ' ' ) def set_random_seed(seed_value, use_cuda=True): np.random.seed(seed_value) # cpu vars torch.manual_seed(seed_value) # cpu vars random.seed(seed_value) # Python os.environ['PYTHONHASHSEED'] = str(seed_value) # Python hash buildin if use_cuda: torch.cuda.manual_seed(seed_value) torch.cuda.manual_seed_all(seed_value) # gpu vars torch.backends.cudnn.deterministic = True #needed torch.backends.cudnn.benchmark = False
8,844
34.239044
129
py
SleePyCo
SleePyCo-main/loader.py
import os import glob import torch import numpy as np from transform import * from torch.utils.data import Dataset class EEGDataLoader(Dataset): def __init__(self, config, fold, set='train'): self.set = set self.fold = fold self.sr = 100 self.dset_cfg = config['dataset'] self.root_dir = self.dset_cfg['root_dir'] self.dset_name = self.dset_cfg['name'] self.num_splits = self.dset_cfg['num_splits'] self.eeg_channel = self.dset_cfg['eeg_channel'] self.seq_len = self.dset_cfg['seq_len'] self.target_idx = self.dset_cfg['target_idx'] self.training_mode = config['training_params']['mode'] self.dataset_path = os.path.join(self.root_dir, 'dset', self.dset_name, 'npz') self.inputs, self.labels, self.epochs = self.split_dataset() if self.training_mode == 'pretrain': self.transform = Compose( transforms=[ RandomAmplitudeScale(), RandomTimeShift(), RandomDCShift(), RandomZeroMasking(), RandomAdditiveGaussianNoise(), RandomBandStopFilter(), ] ) self.two_transform = TwoTransform(self.transform) def __len__(self): return len(self.epochs) def __getitem__(self, idx): n_sample = 30 * self.sr * self.seq_len file_idx, idx, seq_len = self.epochs[idx] inputs = self.inputs[file_idx][idx:idx+seq_len] if self.set == 'train': if self.training_mode == 'pretrain': assert seq_len == 1 input_a, input_b = self.two_transform(inputs) input_a = torch.from_numpy(input_a).float() input_b = torch.from_numpy(input_b).float() inputs = [input_a, input_b] elif self.training_mode in ['scratch', 'fullyfinetune', 'freezefinetune']: inputs = inputs.reshape(1, n_sample) inputs = torch.from_numpy(inputs).float() else: raise NotImplementedError else: if not self.training_mode == 'pretrain': inputs = inputs.reshape(1, n_sample) inputs = torch.from_numpy(inputs).float() labels = self.labels[file_idx][idx:idx+seq_len] labels = torch.from_numpy(labels).long() labels = labels[self.target_idx] return inputs, labels def split_dataset(self): file_idx = 0 inputs, labels, epochs = [], [], [] data_root = os.path.join(self.dataset_path, self.eeg_channel) data_fname_list = [os.path.basename(x) for x in sorted(glob.glob(os.path.join(data_root, '*.npz')))] data_fname_dict = {'train': [], 'test': [], 'val': []} split_idx_list = np.load(os.path.join('./split_idx', 'idx_{}.npy'.format(self.dset_name)), allow_pickle=True) assert len(split_idx_list) == self.num_splits if self.dset_name == 'Sleep-EDF-2013': for i in range(len(data_fname_list)): subject_idx = int(data_fname_list[i][3:5]) if subject_idx == self.fold - 1: data_fname_dict['test'].append(data_fname_list[i]) elif subject_idx in split_idx_list[self.fold - 1]: data_fname_dict['val'].append(data_fname_list[i]) else: data_fname_dict['train'].append(data_fname_list[i]) elif self.dset_name == 'Sleep-EDF-2018': for i in range(len(data_fname_list)): subject_idx = int(data_fname_list[i][3:5]) if subject_idx in split_idx_list[self.fold - 1][self.set]: data_fname_dict[self.set].append(data_fname_list[i]) elif self.dset_name == 'MASS' or self.dset_name == 'Physio2018' or self.dset_name == 'SHHS': for i in range(len(data_fname_list)): if i in split_idx_list[self.fold - 1][self.set]: data_fname_dict[self.set].append(data_fname_list[i]) else: raise NameError("dataset '{}' cannot be found.".format(self.dataset)) for data_fname in data_fname_dict[self.set]: npz_file = np.load(os.path.join(data_root, data_fname)) inputs.append(npz_file['x']) labels.append(npz_file['y']) seq_len = self.seq_len if self.dset_name== 'MASS' and ('-02-' in data_fname or '-04-' in data_fname or '-05-' in data_fname): seq_len = int(self.seq_len * 1.5) for i in range(len(npz_file['y']) - seq_len + 1): epochs.append([file_idx, i, seq_len]) file_idx += 1 return inputs, labels, epochs
4,913
39.278689
117
py
SleePyCo
SleePyCo-main/models/iitnet.py
import torch.nn as nn def conv3(in_planes, out_planes, stride=1): return nn.Conv1d(in_planes, out_planes, kernel_size=3, stride=stride, padding=1, bias=False) class Bottleneck(nn.Module): expansion = 4 def __init__(self, inplanes, planes, stride=1, downsample=None): super(Bottleneck, self).__init__() self.conv1 = nn.Conv1d(inplanes, planes, kernel_size=1, bias=False) self.bn1 = nn.BatchNorm1d(planes) self.conv2 = nn.Conv1d(planes, planes, kernel_size=3, stride=stride, padding=1, bias=False) self.bn2 = nn.BatchNorm1d(planes) self.conv3 = nn.Conv1d(planes, planes * self.expansion, kernel_size=1, bias=False) self.bn3 = nn.BatchNorm1d(planes * self.expansion) self.relu = nn.ReLU(inplace=True) self.downsample = downsample self.stride = stride def forward(self, x): residual = x out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) out = self.relu(out) out = self.conv3(out) out = self.bn3(out) if self.downsample is not None: residual = self.downsample(x) out += residual out = self.relu(out) return out class IITNetBackbone(nn.Module): def __init__(self, config): super(IITNetBackbone, self).__init__() block = Bottleneck self.training_mode = config['training_params']['mode'] self.inplanes = 16 self.layers = [3, 4, 6, 3] self.initial_layer = nn.Sequential( nn.Conv1d(1, 16, 7, 2, 3, bias=False), nn.BatchNorm1d(16), nn.ReLU(), nn.MaxPool1d(3, 2, 1)) self.layer1 = self._make_layer(block, 16, self.layers[0], stride=1, first=True) self.layer2 = self._make_layer(block, 16, self.layers[1], stride=2) self.layer3 = self._make_layer(block, 32, self.layers[2], stride=2) self.layer4 = self._make_layer(block, 32, self.layers[3], stride=2) self.maxpool = nn.MaxPool1d(3, 2, 1) if self.training_mode == 'freezefinetune': self.fp_dim = config['feature_pyramid']['dim'] self.num_scales = config['feature_pyramid']['num_scales'] self.conv_c5 = nn.Conv1d(128, self.fp_dim, 1, 1, 0) assert self.num_scales == 1 if config['backbone']['init_weights']: self._initialize_weights() def _initialize_weights(self): for m in self.modules(): if isinstance(m, nn.Conv1d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') if m.bias is not None: nn.init.constant_(m.bias, 0) elif isinstance(m, nn.BatchNorm1d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) def _make_layer(self, block, planes, blocks, stride=1, first=False): downsample = None if (stride != 1 and first is False) or self.inplanes != planes * block.expansion: downsample = nn.Sequential( nn.Conv1d(self.inplanes, planes * block.expansion, 1, stride, bias=False), nn.BatchNorm1d(planes * block.expansion) ) layers = [] layers.append(block(self.inplanes, planes, stride, downsample)) self.inplanes = planes * block.expansion for _ in range(1, blocks): layers.append(block(self.inplanes, planes)) return nn.Sequential(*layers) def forward(self, x): out = [] c1 = self.initial_layer(x) c2 = self.layer1(c1) c3 = self.layer2(c2) c4 = self.layer3(self.maxpool(c3)) c5 = self.layer4(c4) if self.training_mode == 'pretrain': out.append(c5) elif self.training_mode in ['scratch', 'fullyfinetune', 'freezefinetune']: p5 = self.conv_c5(c5) out.append(p5) return out
4,111
31.896
96
py
SleePyCo
SleePyCo-main/models/xsleepnet.py
import torch.nn as nn class XSleepNetFeature(nn.Module): def __init__(self, config): super(XSleepNetFeature, self).__init__() self.training_mode = config['training_params']['mode'] # architecture self.conv1 = self.make_layers(1, 16) self.conv2 = self.make_layers(16, 16) self.conv3 = self.make_layers(16, 32) self.conv4 = self.make_layers(32, 32) self.conv5 = self.make_layers(32, 64) self.conv6 = self.make_layers(64, 64) self.conv7 = self.make_layers(64, 128) self.conv8 = self.make_layers(128, 128) self.conv9 = self.make_layers(128, 256) if self.training_mode == 'freezefinetune': self.fp_dim = config['feature_pyramid']['dim'] self.num_scales = config['feature_pyramid']['num_scales'] self.conv_c5 = nn.Conv1d(256, self.fp_dim, 1, 1, 0) if self.num_scales > 1: self.conv_c4 = nn.Conv1d(128, self.fp_dim, 1, 1, 0) if self.num_scales > 2: self.conv_c3 = nn.Conv1d(128, self.fp_dim, 1, 1, 0) if config['backbone']['init_weights']: self._initialize_weights() def _initialize_weights(self): for m in self.modules(): if isinstance(m, nn.Conv1d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') if m.bias is not None: nn.init.constant_(m.bias, 0) elif isinstance(m, nn.BatchNorm1d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) def make_layers(self, in_channels, out_channels): layer = [nn.Conv1d(in_channels, out_channels, 31, 2, 15)] layer.append(nn.BatchNorm1d(out_channels)) layer.append(nn.PReLU()) return nn.Sequential(*layer) def forward(self, x): out = [] x = self.conv1(x) x = self.conv2(x) x = self.conv3(x) x = self.conv4(x) c1 = self.conv5(x) c2 = self.conv6(c1) c3 = self.conv7(c2) c4 = self.conv8(c3) c5 = self.conv9(c4) if self.training_mode == 'pretrain': out.append(c5) elif self.training_mode in ['scratch', 'fullyfinetune', 'freezefinetune']: p5 = self.conv_c5(c5) out.append(p5) if self.num_scales > 1: p4 = self.conv_c4(c4) out.append(p4) if self.num_scales > 2: p3 = self.conv_c3(c3) out.append(p3) return out
2,631
32.74359
86
py
SleePyCo
SleePyCo-main/models/utils.py
import torch.utils.data from torch.nn import functional as F import math import torch import torch.nn as nn from torch.nn.parameter import Parameter from torch.nn.functional import pad from torch.nn.modules import Module from torch.nn.modules.utils import _single, _pair, _triple class _ConvNd(Module): def __init__(self, in_channels, out_channels, kernel_size, stride, padding, dilation, transposed, output_padding, groups, bias, weight=None): super(_ConvNd, self).__init__() if in_channels % groups != 0: raise ValueError('in_channels must be divisible by groups') if out_channels % groups != 0: raise ValueError('out_channels must be divisible by groups') self.in_channels = in_channels self.out_channels = out_channels self.kernel_size = kernel_size self.stride = stride self.padding = padding self.dilation = dilation self.transposed = transposed self.output_padding = output_padding self.groups = groups if transposed: self.weight = Parameter(torch.Tensor( in_channels, out_channels // groups, *kernel_size)) else: self.weight = Parameter(torch.Tensor( out_channels, in_channels // groups, *kernel_size)) if bias: self.bias = Parameter(torch.Tensor(out_channels)) else: self.register_parameter('bias', None) self.reset_parameters(weight) def reset_parameters(self, weight): if weight == None: n = self.in_channels for k in self.kernel_size: n *= k stdv = 1. / math.sqrt(n) self.weight.data.uniform_(-stdv, stdv) else: self.weight.data = torch.FloatTensor(weight) if self.bias is not None: self.bias.data.uniform_(-stdv, stdv) def __repr__(self): s = ('{name}({in_channels}, {out_channels}, kernel_size={kernel_size}' ', stride={stride}') if self.padding != (0,) * len(self.padding): s += ', padding={padding}' if self.dilation != (1,) * len(self.dilation): s += ', dilation={dilation}' if self.output_padding != (0,) * len(self.output_padding): s += ', output_padding={output_padding}' if self.groups != 1: s += ', groups={groups}' if self.bias is None: s += ', bias=False' s += ')' return s.format(name=self.__class__.__name__, **self.__dict__) class Conv1d(_ConvNd): def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding='VALID', dilation=1, groups=1, bias=True, weight=None): kernel_size = _single(kernel_size) stride = _single(stride) padding = _single(padding) dilation = _single(dilation) super(Conv1d, self).__init__( in_channels, out_channels, kernel_size, stride, padding, dilation, False, _pair(0), groups, bias, weight) def forward(self, input): return conv1d_same_padding(input, self.weight, self.bias, self.stride, self.padding, self.dilation, self.groups) # custom con2d, because pytorch don't have "padding='same'" option. def conv1d_same_padding(input, weight, bias=None, stride=1, padding='VALID', dilation=1, groups=1): def check_format(*argv): argv_format = [] for i in range(len(argv)): if type(argv[i]) is int: argv_format.append((argv[i], argv[i])) elif hasattr(argv[i], "__getitem__"): argv_format.append(tuple(argv[i])) else: raise TypeError('all input should be int or list-type, now is {}'.format(argv[i])) return argv_format stride, dilation = check_format(stride, dilation) if padding == 'SAME': padding = 0 input_rows = input.size(2) filter_rows = weight.size(2) out_rows = (input_rows + stride[0] - 1) // stride[0] padding_rows = max(0, (out_rows - 1) * stride[0] + (filter_rows - 1) * dilation[0] + 1 - input_rows) rows_odd = padding_rows % 2 # input_cols = input.size(3) # filter_cols = weight.size(3) # out_cols = (input_cols + stride[1] - 1) // stride[1] # padding_cols = max(0, (out_cols - 1) * stride[1] + # (filter_cols - 1) * dilation[1] + 1 - input_cols) # cols_odd = padding_cols % 2 input = pad(input, [padding_rows // 2, padding_rows // 2 + int(rows_odd)]) elif padding == 'VALID': padding = 0 elif type(padding) != int: raise ValueError('Padding should be SAME, VALID or specific integer, but not {}.'.format(padding)) return F.conv1d(input, weight, bias, stride=stride, padding=padding, dilation=dilation, groups=groups) class _MaxPoolNd(Module): def __init__(self, kernel_size, stride=None, padding='VALID', dilation=1, return_indices=False, ceil_mode=False): super(_MaxPoolNd, self).__init__() self.kernel_size = kernel_size self.stride = stride or kernel_size self.padding = padding self.dilation = dilation self.return_indices = return_indices self.ceil_mode = ceil_mode def extra_repr(self): return 'kernel_size={kernel_size}, stride={stride}, padding={padding}' \ ', dilation={dilation}, ceil_mode={ceil_mode}'.format(**self.__dict__) class MaxPool1d(_MaxPoolNd): def forward(self, input): return maxpool1d_same_padding(input, self.kernel_size, self.stride, self.padding, self.dilation, self.return_indices, self.ceil_mode) def extra_repr(self): return 'kernel_size={kernel_size}, stride={stride}, padding={padding}' \ ', dilation={dilation}, ceil_mode={ceil_mode}'.format(**self.__dict__) def maxpool1d_same_padding(input, kernel_size, stride=None, padding='VALID', dilation=1, return_indices=False, ceil_mode=False): if stride is None: _stride, dilation = [kernel_size], [dilation] else: _stride, dilation = [stride], [dilation] if padding == 'SAME': padding = 0 input_rows = input.size(2) filter_rows = kernel_size out_rows = (input_rows + _stride[0] - 1) // _stride[0] padding_rows = max(0, (out_rows - 1) * _stride[0] + (filter_rows - 1) * dilation[0] + 1 - input_rows) rows_odd = padding_rows % 2 # input_cols = input.size(3) # filter_cols = weight.size(3) # out_cols = (input_cols + _stride[1] - 1) // _stride[1] # padding_cols = max(0, (out_cols - 1) * _stride[1] + # (filter_cols - 1) * dilation[1] + 1 - input_cols) # cols_odd = padding_cols % 2 input = pad(input, [padding_rows // 2, padding_rows // 2 + int(rows_odd)]) elif padding == 'VALID': padding = 0 elif type(padding) != int: raise ValueError('Padding should be SAME, VALID or specific integer, but not {}.'.format(padding)) return F.max_pool1d(input, kernel_size, stride, padding, dilation, return_indices, ceil_mode)
7,340
37.434555
128
py
SleePyCo
SleePyCo-main/models/main_model.py
import torch.nn as nn import torch.nn.functional as F from .sleepyco import SleePyCoBackbone from .xsleepnet import XSleepNetFeature from .iitnet import IITNetBackbone from .utime import UTimeEncoder from .deepsleepnet import DeepSleepNetFeature from .classifiers import get_classifier last_chn_dict = { 'SleePyCo': 256, 'XSleepNet': 256, 'IITNet': 128, 'UTime': 256, 'DeepSleepNet': 128 } class MainModel(nn.Module): def __init__(self, config): super(MainModel, self).__init__() self.cfg = config self.bb_cfg = config['backbone'] self.training_mode = config['training_params']['mode'] if self.bb_cfg['name'] == 'SleePyCo': self.feature = SleePyCoBackbone(self.cfg) elif self.bb_cfg['name'] == 'XSleepNet': self.feature = XSleepNetFeature(self.cfg) elif self.bb_cfg['name'] == 'UTime': self.feature = UTimeEncoder(self.cfg) elif self.bb_cfg['name'] == 'IITNet': self.feature = IITNetBackbone(self.cfg) elif self.bb_cfg['name'] == 'DeepSleepNet': self.feature = DeepSleepNetFeature(self.cfg) else: raise NotImplementedError('backbone not supported: {}'.format(config['backbone']['name'])) if self.bb_cfg['dropout']: self.dropout = nn.Dropout(p=0.5) if self.training_mode == 'pretrain': proj_dim = self.cfg['proj_head']['dim'] if config['proj_head']['name'] == 'Linear': self.head = nn.Sequential( nn.AdaptiveAvgPool1d(1), nn.Flatten(), nn.Linear(last_chn_dict[config['backbone']['name']], proj_dim) ) elif config['proj_head']['name'] == 'MLP': self.head = nn.Sequential( nn.AdaptiveAvgPool1d(1), nn.Flatten(), nn.Linear(last_chn_dict[config['backbone']['name']], proj_dim), nn.ReLU(inplace=True), nn.Linear(proj_dim, proj_dim) ) else: raise NotImplementedError('head not supported: {}'.format(config['proj_head']['name'])) print('[INFO] Number of params of backbone: ', sum(p.numel() for p in self.feature.parameters() if p.requires_grad)) print('[INFO] Number of params of proj_head: ', sum(p.numel() for p in self.head.parameters() if p.requires_grad)) elif self.training_mode in ['scratch', 'fullfinetune', 'freezefinetune']: self.classifier = get_classifier(config) print('[INFO] Number of params of backbone: ', sum(p.numel() for p in self.feature.parameters() if p.requires_grad)) print('[INFO] Number of params of classifier: ', sum(p.numel() for p in self.classifier.parameters() if p.requires_grad)) else: raise NotImplementedError('head not supported: {}'.format(config['training_params']['mode'])) def get_max_len(self, features): len_list = [] for feature in features: len_list.append(feature.shape[1]) return max(len_list) def forward(self, x): outputs = [] features = self.feature(x) for feature in features: if self.bb_cfg['dropout']: feature = self.dropout(feature) if self.training_mode == 'pretrain': outputs.append(F.normalize(self.head(feature))) elif self.training_mode in ['scratch', 'fullfinetune', 'freezefinetune']: feature = feature.transpose(1, 2) output = self.classifier(feature) outputs.append(output) # (B, L, H) else: raise NotImplementedError return outputs
3,898
36.490385
133
py
SleePyCo
SleePyCo-main/models/classifiers.py
import math import torch import torch.nn as nn feature_len_dict = { 'SleePyCo': [ [5, 24, 120], [10, 48, 240], [15, 72, 360], [20, 96, 480], [24, 120, 600], [29, 144, 720], [34, 168, 840], [39, 192, 960], [44, 216, 1080], [48, 240, 1200]], } class PlainRNN(nn.Module): def __init__(self, config): super(PlainRNN, self).__init__() self.cfg = config['classifier'] self.num_classes = config['num_classes'] self.input_dim = config['comp_chn'] self.hidden_dim = config['hidden_dim'] self.num_layers = config['num_rnn_layers'] self.bidirectional = config['bidirectional'] # architecture self.rnn = nn.RNN( input_size=self.input_dim, hidden_size=self.hidden_dim, num_layers=self.num_layers, batch_first=True, bidirectional=self.bidirectional ) self.fc = nn.Linear(self.hidden_dim * 2 if config['bidirectional'] else self.hidden_dim, self.num_classes) def init_hidden(self, x): h0 = torch.zeros((self.num_layers * (2 if self.bidirectional else 1), x.size(0), self.hidden_dim)).cuda() return h0 def forward(self, x): hidden = self.init_hidden(x) rnn_output, hidden = self.rnn(x, hidden) if self.bidirectional: output_f = rnn_output[:, -1, :self.hidden_dim] output_b = rnn_output[:, 0, self.hidden_dim:] output = torch.cat((output_f, output_b), dim=1) else: output = rnn_output[:, -1, :] output = self.fc(output) return output class PlainGRU(PlainRNN): def __init__(self, config): super(PlainGRU, self).__init__(config) self.rnn = nn.GRU( input_size=self.input_dim, hidden_size=self.hidden_dim, num_layers=self.num_layers, batch_first=True, bidirectional=self.bidirectional ) class PlainLSTM(PlainRNN): def __init__(self, config): super(PlainLSTM, self).__init__(config) self.rnn = nn.LSTM( input_size=self.input_dim, hidden_size=self.hidden_dim, num_layers=self.num_layers, batch_first=True, bidirectional=self.bidirectional ) def init_hidden(self, x): h0 = torch.zeros((self.num_layers * (2 if self.bidirectional else 1), x.size(0), self.hidden_dim)).cuda() c0 = torch.zeros((self.num_layers * (2 if self.bidirectional else 1), x.size(0), self.hidden_dim)).cuda() return h0, c0 class AttRNN(PlainRNN): def __init__(self, config): super(AttRNN, self).__init__(config) # architecture self.fc = nn.Linear(self.hidden_dim, self.num_classes) self.w_ha = nn.Linear(self.hidden_dim * 2 if config['bidirectional'] else self.hidden_dim, self.hidden_dim, bias=True) self.w_att = nn.Linear(self.hidden_dim, 1, bias=False) def forward(self, x): hidden = self.init_hidden(x) rnn_output, hidden = self.rnn(x, hidden) a_states = self.w_ha(rnn_output) alpha = torch.softmax(self.w_att(a_states), dim=1).view(x.size(0), 1, x.size(1)) weighted_sum = torch.bmm(alpha, a_states) output = weighted_sum.view(x.size(0), -1) output = self.fc(output) return output class AttGRU(AttRNN): def __init__(self, config): super(AttGRU, self).__init__(config) self.rnn = nn.GRU( input_size=self.input_dim, hidden_size=self.hidden_dim, num_layers=self.num_layers, batch_first=True, bidirectional=self.bidirectional ) class AttLSTM(AttRNN): def __init__(self, config): super(AttLSTM, self).__init__(config) self.rnn = nn.LSTM( input_size=self.input_dim, hidden_size=self.hidden_dim, num_layers=self.num_layers, batch_first=True, bidirectional=self.bidirectional ) def init_hidden(self, x): h0 = torch.zeros((self.num_layers * (2 if self.bidirectional else 1), x.size(0), self.hidden_dim)).cuda() c0 = torch.zeros((self.num_layers * (2 if self.bidirectional else 1), x.size(0), self.hidden_dim)).cuda() return h0, c0 class PositionalEncoding(nn.Module): def __init__(self, config, in_features, out_features, dropout=0.1): super(PositionalEncoding, self).__init__() self.cfg = config['classifier']['pos_enc'] if self.cfg['dropout']: self.dropout = nn.Dropout(p=dropout) self.fc = nn.Linear(in_features=in_features, out_features=out_features) self.act_fn = nn.PReLU() self.max_len = feature_len_dict[config['backbone']['name']][config['dataset']['seq_len'] - 1][config['feature_pyramid']['num_scales'] - 1] print('[INFO] Maximum length of pos_enc: {}'.format(self.max_len)) pe = torch.zeros(self.max_len, out_features) position = torch.arange(0, self.max_len, dtype=torch.float).unsqueeze(1) div_term = torch.exp(torch.arange(0, out_features, 2).float() * (-math.log(10000.0) / out_features)) pe[:, 0::2] = torch.sin(position * div_term) pe[:, 1::2] = torch.cos(position * div_term) pe = pe.unsqueeze(0).transpose(0, 1) self.register_buffer('pe', pe) def forward(self, x): x = self.act_fn(self.fc(x)) hop = self.max_len // x.size(0) pe = self.pe[hop//2::hop, :] if pe.shape[0] != x.size(0): pe = pe[:x.size(0), :] x = x + pe if self.cfg['dropout']: x = self.dropout(x) return x class Transformer(nn.Module): def __init__(self, config, nheads, num_encoder_layers, pool='mean'): super(Transformer, self).__init__() self.cfg = config['classifier'] self.model_dim = self.cfg['model_dim'] self.feedforward_dim = self.cfg['feedforward_dim'] self.in_features = config['feature_pyramid']['dim'] self.out_features = self.cfg['model_dim'] self.pos_encoding = PositionalEncoding(config, self.in_features, self.out_features) self.transformer_layer = nn.TransformerEncoderLayer( d_model=self.model_dim, nhead=nheads, dim_feedforward=self.feedforward_dim, dropout=0.1 if self.cfg['dropout'] else 0.0 ) self.transformer = nn.TransformerEncoder(self.transformer_layer, num_layers=num_encoder_layers) self.pool = pool if self.cfg['dropout']: self.dropout = nn.Dropout(p=0.5) if pool == 'attn': self.w_ha = nn.Linear(self.model_dim, self.model_dim, bias=True) self.w_at = nn.Linear(self.model_dim, 1, bias=False) self.fc = nn.Linear(self.model_dim, self.cfg['num_classes']) def forward(self, x): x = x.transpose(0, 1) x = self.pos_encoding(x) x = self.transformer(x) x = x.transpose(0, 1) if self.pool == 'mean': x = x.mean(dim=1) elif self.pool == 'last': x = x[:, -1] elif self.pool == 'attn': a_states = torch.tanh(self.w_ha(x)) alpha = torch.softmax(self.w_at(a_states), dim=1).view(x.size(0), 1, x.size(1)) x = torch.bmm(alpha, a_states).view(x.size(0), -1) elif self.pool == None: x = x else: raise NotImplementedError if self.cfg['dropout']: x = self.dropout(x) out = self.fc(x) return out def get_classifier(config): classifier_name = config['classifier']['name'] if classifier_name == 'PlainRNN': classifier = PlainRNN(config) elif classifier_name == 'AttentionRNN': classifier = AttRNN(config) if classifier_name == 'PlainLSTM': classifier = PlainLSTM(config) elif classifier_name == 'AttentionLSTM': classifier = AttLSTM(config) elif classifier_name == 'PlainGRU': classifier = PlainGRU(config) elif classifier_name == 'AttentionGRU': classifier = AttGRU(config) elif classifier_name == 'Transformer': classifier = Transformer(config, nheads=8, num_encoder_layers=6, pool=config['classifier']['pool']) return classifier
8,588
30.811111
146
py
SleePyCo
SleePyCo-main/models/sleepyco.py
import torch import torch.nn as nn import torch.nn.functional as F class SleePyCoBackbone(nn.Module): def __init__(self, config): super(SleePyCoBackbone, self).__init__() self.training_mode = config['training_params']['mode'] # architecture self.init_layer = self.make_layers(in_channels=1, out_channels=64, n_layers=2, maxpool_size=None, first=True) self.layer1 = self.make_layers(in_channels=64, out_channels=128, n_layers=2, maxpool_size=5) self.layer2 = self.make_layers(in_channels=128, out_channels=192, n_layers=3, maxpool_size=5) self.layer3 = self.make_layers(in_channels=192, out_channels=256, n_layers=3, maxpool_size=5) self.layer4 = self.make_layers(in_channels=256, out_channels=256, n_layers=3, maxpool_size=5) if self.training_mode == 'freezefinetune': self.fp_dim = config['feature_pyramid']['dim'] self.num_scales = config['feature_pyramid']['num_scales'] self.conv_c5 = nn.Conv1d(256, self.fp_dim, 1, 1, 0) if self.num_scales > 1: self.conv_c4 = nn.Conv1d(256, self.fp_dim, 1, 1, 0) if self.num_scales > 2: self.conv_c3 = nn.Conv1d(192, self.fp_dim, 1, 1, 0) if config['backbone']['init_weights']: self._initialize_weights() def _initialize_weights(self): for m in self.modules(): if isinstance(m, nn.Conv1d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') if m.bias is not None: nn.init.constant_(m.bias, 0) elif isinstance(m, nn.BatchNorm1d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) def make_layers(self, in_channels, out_channels, n_layers, maxpool_size, first=False): layers = [] layers = layers + [MaxPool1d(maxpool_size)] if not first else layers for i in range(n_layers): conv1d = nn.Conv1d(in_channels, out_channels, kernel_size=3, padding=1) layers += [conv1d, nn.BatchNorm1d(out_channels)] if i == n_layers - 1: layers += [ChannelGate(in_channels)] layers += [nn.PReLU()] in_channels = out_channels return nn.Sequential(*layers) def forward(self, x): out = [] c1 = self.init_layer(x) c2 = self.layer1(c1) c3 = self.layer2(c2) c4 = self.layer3(c3) c5 = self.layer4(c4) if self.training_mode == 'pretrain': out.append(c5) elif self.training_mode in ['scratch', 'fullyfinetune', 'freezefinetune']: p5 = self.conv_c5(c5) out.append(p5) if self.num_scales > 1: p4 = self.conv_c4(c4) out.append(p4) if self.num_scales > 2: p3 = self.conv_c3(c3) out.append(p3) return out class MaxPool1d(nn.Module): def __init__(self, maxpool_size): super(MaxPool1d, self).__init__() self.maxpool_size = maxpool_size self.maxpool = nn.MaxPool1d(kernel_size=maxpool_size, stride=maxpool_size) def forward(self, x): _, _, n_samples = x.size() if n_samples % self.maxpool_size != 0: pad_size = self.maxpool_size - (n_samples % self.maxpool_size) if pad_size % 2 != 0: left_pad = pad_size // 2 right_pad = pad_size // 2 + 1 else: left_pad = pad_size // 2 right_pad = pad_size // 2 x = F.pad(x, (left_pad, right_pad), mode='constant') x = self.maxpool(x) return x class BasicConv(nn.Module): def __init__(self, in_planes, out_planes, kernel_size, stride=1, padding=0, dilation=1, groups=1, relu=True, bn=True, bias=False): super(BasicConv, self).__init__() self.out_channels = out_planes self.conv = nn.Conv1d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, groups=groups, bias=bias) self.bn = nn.BatchNorm1d(out_planes,eps=1e-5, momentum=0.01, affine=True) if bn else None self.relu = nn.ReLU() if relu else None def forward(self, x): x = self.conv(x) if self.bn is not None: x = self.bn(x) if self.relu is not None: x = self.relu(x) return x class ChannelGate(nn.Module): def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg']): super(ChannelGate, self).__init__() self.gate_channels = gate_channels self.mlp = nn.Sequential( nn.Flatten(), nn.Linear(gate_channels, gate_channels // reduction_ratio), nn.ReLU(), nn.Linear(gate_channels // reduction_ratio, gate_channels) ) self.pool_types = pool_types def forward(self, x): channel_att_sum = None for pool_type in self.pool_types: if pool_type=='avg': avg_pool = F.avg_pool1d(x, x.size(2), stride=x.size(2)) channel_att_raw = self.mlp(avg_pool) elif pool_type=='max': max_pool = F.max_pool1d(x, x.size(2), stride=x.size(2)) channel_att_raw = self.mlp( max_pool ) elif pool_type=='lp': lp_pool = F.lp_pool2d( x, 2, (x.size(2), x.size(3)), stride=(x.size(2), x.size(3))) channel_att_raw = self.mlp( lp_pool ) elif pool_type=='lse': # LSE pool only lse_pool = logsumexp_2d(x) channel_att_raw = self.mlp( lse_pool ) if channel_att_sum is None: channel_att_sum = channel_att_raw else: channel_att_sum = channel_att_sum + channel_att_raw scale = F.sigmoid(channel_att_sum).unsqueeze(2).expand_as(x) return x * scale def logsumexp_2d(tensor): tensor_flatten = tensor.view(tensor.size(0), tensor.size(1), -1) s, _ = torch.max(tensor_flatten, dim=2, keepdim=True) outputs = s + (tensor_flatten - s).exp().sum(dim=2, keepdim=True).log() return outputs
6,271
37.012121
154
py
SleePyCo
SleePyCo-main/models/deepsleepnet.py
import torch import torch.nn as nn from .utils import Conv1d, MaxPool1d class DeepSleepNetFeature(nn.Module): def __init__(self, config): super(DeepSleepNetFeature, self).__init__() self.chn = 64 # architecture self.dropout = nn.Dropout(p=0.5) self.path1 = nn.Sequential(Conv1d(1, self.chn, 50, 6, padding='SAME', bias=False), nn.BatchNorm1d(self.chn), nn.ReLU(inplace=True), MaxPool1d(8, padding='SAME'), nn.Dropout(), Conv1d(self.chn, self.chn*2, 8, 1, padding='SAME', bias=False), nn.BatchNorm1d(self.chn*2), nn.ReLU(inplace=True), Conv1d(self.chn*2, self.chn*2, 8, 1, padding='SAME', bias=False), nn.BatchNorm1d(self.chn * 2), nn.ReLU(inplace=True), Conv1d(self.chn*2, self.chn*2, 8, 1, padding='SAME', bias=False), nn.BatchNorm1d(self.chn*2), nn.ReLU(inplace=True), MaxPool1d(4, padding='SAME') ) self.path2 = nn.Sequential(Conv1d(1, self.chn, 400, 50, padding='SAME', bias=False), nn.BatchNorm1d(self.chn), nn.ReLU(inplace=True), MaxPool1d(4, padding='SAME'), nn.Dropout(), Conv1d(self.chn, self.chn*2, 8, 1, padding='SAME', bias=False), nn.BatchNorm1d(self.chn*2), nn.ReLU(inplace=True), Conv1d(self.chn*2, self.chn*2, 8, 1, padding='SAME', bias=False), nn.BatchNorm1d(self.chn * 2), nn.ReLU(inplace=True), Conv1d(self.chn*2, self.chn*2, 8, 1, padding='SAME', bias=False), nn.BatchNorm1d(self.chn*2), nn.ReLU(inplace=True), MaxPool1d(2, padding='SAME')) self.compress = nn.Conv1d(self.chn*4, 128, 1, 1, 0) self.smooth = nn.Conv1d(128, 128, 3, 1, 1) if self.training_mode == 'freezefinetune': self.fp_dim = config['feature_pyramid']['dim'] self.num_scales = config['feature_pyramid']['num_scales'] self.conv_c5 = nn.Conv1d(128, self.fp_dim, 1, 1, 0) assert self.num_scales == 1 if config['backbone']['init_weights']: self._initialize_weights() def _initialize_weights(self): for m in self.modules(): if isinstance(m, nn.Conv1d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') if m.bias is not None: nn.init.constant_(m.bias, 0) elif isinstance(m, nn.BatchNorm1d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) def forward(self, x): out = [] x1 = self.path1(x) # path 1 x2 = self.path2(x) # path 2 x2 = torch.nn.functional.interpolate(x2, x1.size(2)) c5 = self.smooth(self.compress(torch.cat([x1, x2], dim=1))) if self.training_mode == 'pretrain': out.append(c5) elif self.training_mode in ['scratch', 'fullyfinetune', 'freezefinetune']: p5 = self.conv_c5(c5) out.append(p5) return out
3,855
44.904762
100
py
SleePyCo
SleePyCo-main/models/utime.py
import torch import torch.nn as nn from .utils import Conv1d class ConvUnit(nn.Module): def __init__(self, in_channels, out_channels, kernel_size, stride, padding, dilation): super(ConvUnit, self).__init__() self.conv = Conv1d( in_channels=in_channels, out_channels=out_channels, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation ) self.bn = nn.BatchNorm1d(num_features=out_channels) self.relu = nn.ReLU() def forward(self, x): x = self.conv(x) x = self.bn(x) x = self.relu(x) return x class UTimeEncoder(nn.Module): def __init__(self, config): super(UTimeEncoder, self).__init__() torch.backends.cudnn.deterministic = False self.training_mode = config['training_params']['mode'] self.conv1_1 = ConvUnit(in_channels=1, out_channels=16, kernel_size=5, stride=1, padding='SAME', dilation=2) self.conv1_2 = ConvUnit(in_channels=16, out_channels=16, kernel_size=5, stride=1, padding='SAME', dilation=2) self.mp1 = nn.MaxPool1d(kernel_size=8) self.conv2_1 = ConvUnit(in_channels=16, out_channels=32, kernel_size=5, stride=1, padding='SAME', dilation=2) self.conv2_2 = ConvUnit(in_channels=32, out_channels=32, kernel_size=5, stride=1, padding='SAME', dilation=2) self.mp2 = nn.MaxPool1d(kernel_size=6) self.conv3_1 = ConvUnit(in_channels=32, out_channels=64, kernel_size=5, stride=1, padding='SAME', dilation=2) self.conv3_2 = ConvUnit(in_channels=64, out_channels=64, kernel_size=5, stride=1, padding='SAME', dilation=2) self.mp3 = nn.MaxPool1d(kernel_size=4) self.conv4_1 = ConvUnit(in_channels=64, out_channels=128, kernel_size=5, stride=1, padding='SAME', dilation=2) self.conv4_2 = ConvUnit(in_channels=128, out_channels=128, kernel_size=5, stride=1, padding='SAME', dilation=2) self.mp4 = nn.MaxPool1d(kernel_size=2) self.conv5_1 = ConvUnit(in_channels=128, out_channels=256, kernel_size=5, stride=1, padding='SAME', dilation=2) self.conv5_2 = ConvUnit(in_channels=256, out_channels=256, kernel_size=5, stride=1, padding='SAME', dilation=2) if self.training_mode == 'freezefinetune': self.fp_dim = config['feature_pyramid']['dim'] self.num_scales = config['feature_pyramid']['num_scales'] self.conv_c5 = nn.Conv1d(256, self.fp_dim, 1, 1, 0) if self.n_anchor > 1: self.conv_c4 = nn.Conv1d(128, self.fp_dim, 1, 1, 0) if self.n_anchor > 2: self.conv_c3 = nn.Conv1d(64, self.fp_dim, 1, 1, 0) if config['backbone']['init_weights']: self._initialize_weights() def _initialize_weights(self): for m in self.modules(): if isinstance(m, nn.Conv1d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') if m.bias is not None: nn.init.constant_(m.bias, 0) elif isinstance(m, nn.BatchNorm1d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) def forward(self, x): out = [] x = self.conv1_1(x) c1 = self.conv1_2(x) x = self.mp1(c1) x = self.conv2_1(x) c2 = self.conv2_2(x) x = self.mp2(c2) x = self.conv3_1(x) c3 = self.conv3_2(x) x = self.mp3(c3) x = self.conv4_1(x) c4 = self.conv4_2(x) x = self.mp4(c4) x = self.conv5_1(x) c5 = self.conv5_2(x) if self.training_mode == 'pretrain': out.append(c5) elif self.training_mode in ['scratch', 'fullyfinetune', 'freezefinetune']: p5 = self.conv_c5(c5) out.append(p5) if self.num_scales > 1: p4 = self.conv_c4(c4) out.append(p4) if self.num_scales > 2: p3 = self.conv_c3(c3) out.append(p3) return out
4,199
37.181818
119
py
SleePyCo
SleePyCo-main/dset/Sleep-EDF-2013/download_sleep-edf-2013.py
import os os.makedirs('./edf', exist_ok=True) os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4001E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4001EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4002E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4002EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4011E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4011EH-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4012E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4012EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4021E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4021EH-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4022E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4022EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4031E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4031EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4032E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4032EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4041E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4041EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4042E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4042EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4051E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4051EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4052E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4052EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4061E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4061EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4062E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4062EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4071E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4071EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4072E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4072EH-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4081E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4081EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4082E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4082EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4091E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4091EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4092E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4092EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4101E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4101EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4102E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4102EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4111E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4111EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4112E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4112EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4121E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4121EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4122E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4122EV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4131E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4131EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4141E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4141EU-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4142E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4142EU-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4151E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4151EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4152E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4152EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4161E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4161EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4162E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4162EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4171E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4171EU-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4172E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4172EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4181E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4181EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4182E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4182EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4191E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4191EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4192E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4192EV-Hypnogram.edf -P ./edf')
9,329
112.780488
121
py
SleePyCo
SleePyCo-main/dset/Sleep-EDF-2013/prepare_sleep-edf-2013.py
import os import glob import ntpath import logging import argparse import pyedflib import numpy as np # Label values W = 0 N1 = 1 N2 = 2 N3 = 3 REM = 4 MOVE = 5 UNK = 6 stage_dict = { "W": W, "N1": N1, "N2": N2, "N3": N3, "REM": REM, "MOVE": MOVE, "UNK": UNK } # Have to manually define based on the dataset ann2label = { "Sleep stage W": 0, "Sleep stage 1": 1, "Sleep stage 2": 2, "Sleep stage 3": 3, "Sleep stage 4": 3, # Follow AASM Manual "Sleep stage R": 4, "Sleep stage ?": 6, "Movement time": 5 } def main(): parser = argparse.ArgumentParser() parser.add_argument("--data_dir", type=str, default="./edf", help="File path to the Sleep-EDF dataset.") parser.add_argument("--output_dir", type=str, default="./npz", help="Directory where to save outputs.") parser.add_argument("--select_ch", type=str, default="EEG Fpz-Cz", help="Name of the channel in the dataset.") parser.add_argument("--log_file", type=str, default="info_ch_extract.log", help="Log file.") args = parser.parse_args() # Output dir args.output_dir = os.path.join(args.output_dir, args.select_ch.split(' ')[-1]) os.makedirs(args.output_dir, exist_ok=True) args.log_file = os.path.join(os.path.join(args.output_dir, args.log_file)) # Create logger logger = logging.getLogger() logger.setLevel(logging.INFO) formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s') stream_handler = logging.StreamHandler() stream_handler.setFormatter(formatter) logger.addHandler(stream_handler) file_handler = logging.FileHandler(args.log_file) file_handler.setFormatter(formatter) logger.addHandler(file_handler) # Select channel select_ch = args.select_ch # Read raw and annotation from EDF files psg_fnames = glob.glob(os.path.join(args.data_dir, "*PSG.edf")) ann_fnames = glob.glob(os.path.join(args.data_dir, "*Hypnogram.edf")) psg_fnames.sort() ann_fnames.sort() psg_fnames = np.asarray(psg_fnames) ann_fnames = np.asarray(ann_fnames) for i in range(len(psg_fnames)): logger.info("Loading ...") logger.info("Signal file: {}".format(psg_fnames[i])) logger.info("Annotation file: {}".format(ann_fnames[i])) psg_f = pyedflib.EdfReader(psg_fnames[i]) ann_f = pyedflib.EdfReader(ann_fnames[i]) assert psg_f.getStartdatetime() == ann_f.getStartdatetime() start_datetime = psg_f.getStartdatetime() logger.info("Start datetime: {}".format(str(start_datetime))) file_duration = psg_f.getFileDuration() logger.info("File duration: {} sec".format(file_duration)) epoch_duration = psg_f.datarecord_duration if psg_f.datarecord_duration == 60: # Fix problems of SC4362F0-PSG.edf, SC4362FC-Hypnogram.edf epoch_duration = epoch_duration / 2 logger.info("Epoch duration: {} sec (changed from 60 sec)".format(epoch_duration)) else: logger.info("Epoch duration: {} sec".format(epoch_duration)) # Extract signal from the selected channel ch_names = psg_f.getSignalLabels() ch_samples = psg_f.getNSamples() select_ch_idx = -1 for s in range(psg_f.signals_in_file): if ch_names[s] == select_ch: select_ch_idx = s break if select_ch_idx == -1: raise Exception("Channel not found.") sampling_rate = 100.0 n_epoch_samples = int(epoch_duration * sampling_rate) signals = psg_f.readSignal(select_ch_idx).reshape(-1, n_epoch_samples) logger.info("Select channel: {}".format(select_ch)) logger.info("Select channel samples: {}".format(ch_samples[select_ch_idx])) logger.info("Sample rate: {}".format(sampling_rate)) # Sanity check n_epochs = psg_f.datarecords_in_file if psg_f.datarecord_duration == 60: # Fix problems of SC4362F0-PSG.edf, SC4362FC-Hypnogram.edf n_epochs = n_epochs * 2 assert len(signals) == n_epochs, f"signal: {signals.shape} != {n_epochs}" # Generate labels from onset and duration annotation labels = [] total_duration = 0 ann_onsets, ann_durations, ann_stages = ann_f.readAnnotations() for a in range(len(ann_stages)): onset_sec = int(ann_onsets[a]) duration_sec = int(ann_durations[a]) ann_str = "".join(ann_stages[a]) # Sanity check assert onset_sec == total_duration # Get label value label = ann2label[ann_str] # Compute # of epoch for this stage if duration_sec % epoch_duration != 0: logger.info(f"Something wrong: {duration_sec} {epoch_duration}") raise Exception(f"Something wrong: {duration_sec} {epoch_duration}") duration_epoch = int(duration_sec / epoch_duration) # Generate sleep stage labels label_epoch = np.ones(duration_epoch, dtype=np.int) * label labels.append(label_epoch) total_duration += duration_sec logger.info("Include onset:{}, duration:{}, label:{} ({})".format( onset_sec, duration_sec, label, ann_str )) labels = np.hstack(labels) # Remove annotations that are longer than the recorded signals labels = labels[:len(signals)] # Get epochs and their corresponding labels x = signals.astype(np.float32) y = labels.astype(np.int32) # Select only sleep periods w_edge_mins = 30 nw_idx = np.where(y != stage_dict["W"])[0] start_idx = nw_idx[0] - (w_edge_mins * 2) end_idx = nw_idx[-1] + (w_edge_mins * 2) if start_idx < 0: start_idx = 0 if end_idx >= len(y): end_idx = len(y) - 1 select_idx = np.arange(start_idx, end_idx+1) logger.info("Data before selection: {}, {}".format(x.shape, y.shape)) x = x[select_idx] y = y[select_idx] logger.info("Data after selection: {}, {}".format(x.shape, y.shape)) # Remove movement and unknown move_idx = np.where(y == stage_dict["MOVE"])[0] unk_idx = np.where(y == stage_dict["UNK"])[0] if len(move_idx) > 0 or len(unk_idx) > 0: remove_idx = np.union1d(move_idx, unk_idx) logger.info("Remove irrelavant stages") logger.info(" Movement: ({}) {}".format(len(move_idx), move_idx)) logger.info(" Unknown: ({}) {}".format(len(unk_idx), unk_idx)) logger.info(" Remove: ({}) {}".format(len(remove_idx), remove_idx)) logger.info(" Data before removal: {}, {}".format(x.shape, y.shape)) select_idx = np.setdiff1d(np.arange(len(x)), remove_idx) x = x[select_idx] y = y[select_idx] logger.info(" Data after removal: {}, {}".format(x.shape, y.shape)) # Save filename = ntpath.basename(psg_fnames[i]).replace("-PSG.edf", ".npz") save_dict = { "x": x, "y": y, "fs": sampling_rate, "ch_label": select_ch, "start_datetime": start_datetime, "file_duration": file_duration, "epoch_duration": epoch_duration, "n_all_epochs": n_epochs, "n_epochs": len(x), } np.savez(os.path.join(args.output_dir, filename), **save_dict) logger.info("\n=======================================\n") if __name__ == "__main__": main()
7,734
35.314554
102
py
SleePyCo
SleePyCo-main/dset/Sleep-EDF-2018/prepare_sleep-edf-2018.py
import os import glob import ntpath import logging import argparse import pyedflib import numpy as np # Label values W = 0 N1 = 1 N2 = 2 N3 = 3 REM = 4 MOVE = 5 UNK = 6 stage_dict = { "W": W, "N1": N1, "N2": N2, "N3": N3, "REM": REM, "MOVE": MOVE, "UNK": UNK } # Have to manually define based on the dataset ann2label = { "Sleep stage W": 0, "Sleep stage 1": 1, "Sleep stage 2": 2, "Sleep stage 3": 3, "Sleep stage 4": 3, # Follow AASM Manual "Sleep stage R": 4, "Sleep stage ?": 6, "Movement time": 5 } def main(): parser = argparse.ArgumentParser() parser.add_argument("--data_dir", type=str, default="./edf", help="File path to the Sleep-EDF dataset.") parser.add_argument("--output_dir", type=str, default="./npz", help="Directory where to save outputs.") parser.add_argument("--select_ch", type=str, default="EEG Fpz-Cz", help="Name of the channel in the dataset.") parser.add_argument("--log_file", type=str, default="info_ch_extract.log", help="Log file.") args = parser.parse_args() # Output dir args.output_dir = os.path.join(args.output_dir, args.select_ch.split(' ')[-1]) os.makedirs(args.output_dir, exist_ok=True) args.log_file = os.path.join(os.path.join(args.output_dir, args.log_file)) # Create logger logger = logging.getLogger() logger.setLevel(logging.INFO) formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s') stream_handler = logging.StreamHandler() stream_handler.setFormatter(formatter) logger.addHandler(stream_handler) file_handler = logging.FileHandler(args.log_file) file_handler.setFormatter(formatter) logger.addHandler(file_handler) # Select channel select_ch = args.select_ch # Read raw and annotation from EDF files psg_fnames = glob.glob(os.path.join(args.data_dir, "*PSG.edf")) ann_fnames = glob.glob(os.path.join(args.data_dir, "*Hypnogram.edf")) psg_fnames.sort() ann_fnames.sort() psg_fnames = np.asarray(psg_fnames) ann_fnames = np.asarray(ann_fnames) for i in range(len(psg_fnames)): logger.info("Loading ...") logger.info("Signal file: {}".format(psg_fnames[i])) logger.info("Annotation file: {}".format(ann_fnames[i])) psg_f = pyedflib.EdfReader(psg_fnames[i]) ann_f = pyedflib.EdfReader(ann_fnames[i]) assert psg_f.getStartdatetime() == ann_f.getStartdatetime() start_datetime = psg_f.getStartdatetime() logger.info("Start datetime: {}".format(str(start_datetime))) file_duration = psg_f.getFileDuration() logger.info("File duration: {} sec".format(file_duration)) epoch_duration = psg_f.datarecord_duration if psg_f.datarecord_duration == 60: # Fix problems of SC4362F0-PSG.edf, SC4362FC-Hypnogram.edf epoch_duration = epoch_duration / 2 logger.info("Epoch duration: {} sec (changed from 60 sec)".format(epoch_duration)) else: logger.info("Epoch duration: {} sec".format(epoch_duration)) # Extract signal from the selected channel ch_names = psg_f.getSignalLabels() ch_samples = psg_f.getNSamples() select_ch_idx = -1 for s in range(psg_f.signals_in_file): if ch_names[s] == select_ch: select_ch_idx = s break if select_ch_idx == -1: raise Exception("Channel not found.") sampling_rate = 100.0 n_epoch_samples = int(epoch_duration * sampling_rate) signals = psg_f.readSignal(select_ch_idx).reshape(-1, n_epoch_samples) logger.info("Select channel: {}".format(select_ch)) logger.info("Select channel samples: {}".format(ch_samples[select_ch_idx])) logger.info("Sample rate: {}".format(sampling_rate)) # Sanity check n_epochs = psg_f.datarecords_in_file if psg_f.datarecord_duration == 60: # Fix problems of SC4362F0-PSG.edf, SC4362FC-Hypnogram.edf n_epochs = n_epochs * 2 assert len(signals) == n_epochs, f"signal: {signals.shape} != {n_epochs}" # Generate labels from onset and duration annotation labels = [] total_duration = 0 ann_onsets, ann_durations, ann_stages = ann_f.readAnnotations() for a in range(len(ann_stages)): onset_sec = int(ann_onsets[a]) duration_sec = int(ann_durations[a]) ann_str = "".join(ann_stages[a]) # Sanity check assert onset_sec == total_duration # Get label value label = ann2label[ann_str] # Compute # of epoch for this stage if duration_sec % epoch_duration != 0: logger.info(f"Something wrong: {duration_sec} {epoch_duration}") raise Exception(f"Something wrong: {duration_sec} {epoch_duration}") duration_epoch = int(duration_sec / epoch_duration) # Generate sleep stage labels label_epoch = np.ones(duration_epoch, dtype=np.int) * label labels.append(label_epoch) total_duration += duration_sec logger.info("Include onset:{}, duration:{}, label:{} ({})".format( onset_sec, duration_sec, label, ann_str )) labels = np.hstack(labels) # Remove annotations that are longer than the recorded signals labels = labels[:len(signals)] # Get epochs and their corresponding labels x = signals.astype(np.float32) y = labels.astype(np.int32) # Select only sleep periods w_edge_mins = 30 nw_idx = np.where(y != stage_dict["W"])[0] start_idx = nw_idx[0] - (w_edge_mins * 2) end_idx = nw_idx[-1] + (w_edge_mins * 2) if start_idx < 0: start_idx = 0 if end_idx >= len(y): end_idx = len(y) - 1 select_idx = np.arange(start_idx, end_idx+1) logger.info("Data before selection: {}, {}".format(x.shape, y.shape)) x = x[select_idx] y = y[select_idx] logger.info("Data after selection: {}, {}".format(x.shape, y.shape)) # Remove movement and unknown move_idx = np.where(y == stage_dict["MOVE"])[0] unk_idx = np.where(y == stage_dict["UNK"])[0] if len(move_idx) > 0 or len(unk_idx) > 0: remove_idx = np.union1d(move_idx, unk_idx) logger.info("Remove irrelavant stages") logger.info(" Movement: ({}) {}".format(len(move_idx), move_idx)) logger.info(" Unknown: ({}) {}".format(len(unk_idx), unk_idx)) logger.info(" Remove: ({}) {}".format(len(remove_idx), remove_idx)) logger.info(" Data before removal: {}, {}".format(x.shape, y.shape)) select_idx = np.setdiff1d(np.arange(len(x)), remove_idx) x = x[select_idx] y = y[select_idx] logger.info(" Data after removal: {}, {}".format(x.shape, y.shape)) # Save filename = ntpath.basename(psg_fnames[i]).replace("-PSG.edf", ".npz") save_dict = { "x": x, "y": y, "fs": sampling_rate, "ch_label": select_ch, "start_datetime": start_datetime, "file_duration": file_duration, "epoch_duration": epoch_duration, "n_all_epochs": n_epochs, "n_epochs": len(x), } np.savez(os.path.join(args.output_dir, filename), **save_dict) logger.info("\n=======================================\n") if __name__ == "__main__": main()
7,734
35.314554
102
py
SleePyCo
SleePyCo-main/dset/Sleep-EDF-2018/download_sleep-edf-2018.py
import os os.makedirs('./edf', exist_ok=True) os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4001E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4001EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4002E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4002EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4011E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4011EH-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4012E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4012EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4021E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4021EH-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4022E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4022EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4031E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4031EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4032E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4032EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4041E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4041EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4042E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4042EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4051E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4051EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4052E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4052EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4061E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4061EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4062E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4062EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4071E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4071EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4072E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4072EH-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4081E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4081EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4082E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4082EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4091E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4091EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4092E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4092EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4101E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4101EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4102E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4102EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4111E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4111EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4112E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4112EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4121E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4121EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4122E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4122EV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4131E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4131EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4141E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4141EU-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4142E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4142EU-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4151E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4151EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4152E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4152EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4161E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4161EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4162E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4162EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4171E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4171EU-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4172E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4172EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4181E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4181EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4182E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4182EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4191E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4191EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4192E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4192EV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4201E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4201EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4202E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4202EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4211E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4211EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4212E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4212EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4221E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4221EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4222E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4222EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4231E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4231EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4232E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4232EV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4241E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4241EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4242E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4242EA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4251E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4251EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4252E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4252EU-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4261F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4261FM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4262F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4262FC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4271F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4271FC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4272F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4272FM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4281G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4281GC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4282G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4282GC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4291G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4291GA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4292G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4292GC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4301E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4301EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4302E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4302EV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4311E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4311EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4312E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4312EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4321E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4321EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4322E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4322EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4331F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4331FV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4332F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4332FC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4341F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4341FA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4342F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4342FA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4351F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4351FA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4352F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4352FV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4362F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4362FC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4371F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4371FA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4372F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4372FC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4381F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4381FC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4382F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4382FW-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4401E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4401EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4402E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4402EW-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4411E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4411EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4412E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4412EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4421E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4421EA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4422E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4422EA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4431E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4431EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4432E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4432EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4441E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4441EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4442E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4442EV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4451F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4451FY-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4452F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4452FW-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4461F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4461FA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4462F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4462FJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4471F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4471FA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4472F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4472FA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4481F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4481FV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4482F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4482FJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4491G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4491GJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4492G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4492GJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4501E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4501EW-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4502E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4502EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4511E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4511EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4512E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4512EW-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4522E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4522EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4531E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4531EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4532E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4532EV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4541F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4541FA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4542F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4542FW-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4551F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4551FC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4552F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4552FW-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4561F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4561FJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4562F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4562FJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4571F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4571FV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4572F0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4572FC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4581G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4581GM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4582G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4582GP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4591G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4591GY-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4592G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4592GY-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4601E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4601EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4602E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4602EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4611E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4611EG-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4612E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4612EA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4621E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4621EV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4622E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4622EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4631E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4631EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4632E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4632EA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4641E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4641EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4642E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4642EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4651E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4651EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4652E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4652EG-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4661E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4661EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4662E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4662EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4671G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4671GJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4672G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4672GV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4701E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4701EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4702E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4702EA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4711E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4711EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4712E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4712EA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4721E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4721EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4722E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4722EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4731E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4731EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4732E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4732EJ-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4741E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4741EA-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4742E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4742EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4751E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4751EC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4752E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4752EM-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4761E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4761EP-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4762E0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4762EG-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4771G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4771GC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4772G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4772GC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4801G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4801GC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4802G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4802GV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4811G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4811GG-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4812G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4812GV-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4821G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4821GC-Hypnogram.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4822G0-PSG.edf -P ./edf') os.system('wget https://www.physionet.org/physiobank/database/sleep-edfx/sleep-cassette/SC4822GC-Hypnogram.edf -P ./edf')
36,460
116.996764
121
py
SASA
SASA-main/setup.py
import os import subprocess from setuptools import find_packages, setup from torch.utils.cpp_extension import BuildExtension, CUDAExtension def get_git_commit_number(): if not os.path.exists('.git'): return '0000000' cmd_out = subprocess.run(['git', 'rev-parse', 'HEAD'], stdout=subprocess.PIPE) git_commit_number = cmd_out.stdout.decode('utf-8')[:7] return git_commit_number def make_cuda_ext(name, module, sources): cuda_ext = CUDAExtension( name='%s.%s' % (module, name), sources=[os.path.join(*module.split('.'), src) for src in sources] ) return cuda_ext def write_version_to_file(version, target_file): with open(target_file, 'w') as f: print('__version__ = "%s"' % version, file=f) if __name__ == '__main__': version = '0.3.0+%s' % get_git_commit_number() write_version_to_file(version, 'pcdet/version.py') setup( name='pcdet', version=version, description='OpenPCDet is a general codebase for 3D object detection from point cloud', install_requires=[ 'numpy', 'torch>=1.1', 'spconv', 'numba', 'tensorboardX', 'easydict', 'pyyaml' ], author='Shaoshuai Shi', author_email='shaoshuaics@gmail.com', license='Apache License 2.0', packages=find_packages(exclude=['tools', 'data', 'output']), cmdclass={'build_ext': BuildExtension}, ext_modules=[ make_cuda_ext( name='iou3d_nms_cuda', module='pcdet.ops.iou3d_nms', sources=[ 'src/iou3d_cpu.cpp', 'src/iou3d_nms_api.cpp', 'src/iou3d_nms.cpp', 'src/iou3d_nms_kernel.cu', ] ), make_cuda_ext( name='roiaware_pool3d_cuda', module='pcdet.ops.roiaware_pool3d', sources=[ 'src/roiaware_pool3d.cpp', 'src/roiaware_pool3d_kernel.cu', ] ), make_cuda_ext( name='roipoint_pool3d_cuda', module='pcdet.ops.roipoint_pool3d', sources=[ 'src/roipoint_pool3d.cpp', 'src/roipoint_pool3d_kernel.cu', ] ), make_cuda_ext( name='pointnet2_stack_cuda', module='pcdet.ops.pointnet2.pointnet2_stack', sources=[ 'src/pointnet2_api.cpp', 'src/ball_query.cpp', 'src/ball_query_gpu.cu', 'src/group_points.cpp', 'src/group_points_gpu.cu', 'src/sampling.cpp', 'src/sampling_gpu.cu', 'src/interpolate.cpp', 'src/interpolate_gpu.cu', ], ), make_cuda_ext( name='pointnet2_batch_cuda', module='pcdet.ops.pointnet2.pointnet2_batch', sources=[ 'src/pointnet2_api.cpp', 'src/ball_query.cpp', 'src/ball_query_gpu.cu', 'src/group_points.cpp', 'src/group_points_gpu.cu', 'src/interpolate.cpp', 'src/interpolate_gpu.cu', 'src/sampling.cpp', 'src/sampling_gpu.cu', ], ), ], )
3,616
31.294643
95
py
SASA
SASA-main/tools/test.py
import argparse import datetime import glob import os import re import time from pathlib import Path import numpy as np import torch from tensorboardX import SummaryWriter from eval_utils import eval_utils from pcdet.config import cfg, cfg_from_list, cfg_from_yaml_file, log_config_to_file from pcdet.datasets import build_dataloader from pcdet.models import build_network from pcdet.utils import common_utils def parse_config(): parser = argparse.ArgumentParser(description='arg parser') parser.add_argument('--cfg_file', type=str, default=None, help='specify the config for training') parser.add_argument('--batch_size', type=int, default=None, required=False, help='batch size for training') parser.add_argument('--workers', type=int, default=4, help='number of workers for dataloader') parser.add_argument('--extra_tag', type=str, default='default', help='extra tag for this experiment') parser.add_argument('--ckpt', type=str, default=None, help='checkpoint to start from') parser.add_argument('--launcher', choices=['none', 'pytorch', 'slurm'], default='none') parser.add_argument('--tcp_port', type=int, default=18888, help='tcp port for distrbuted training') parser.add_argument('--local_rank', type=int, default=0, help='local rank for distributed training') parser.add_argument('--set', dest='set_cfgs', default=None, nargs=argparse.REMAINDER, help='set extra config keys if needed') parser.add_argument('--max_waiting_mins', type=int, default=30, help='max waiting minutes') parser.add_argument('--start_epoch', type=int, default=0, help='') parser.add_argument('--eval_tag', type=str, default='default', help='eval tag for this experiment') parser.add_argument('--eval_all', action='store_true', default=False, help='whether to evaluate all checkpoints') parser.add_argument('--ckpt_dir', type=str, default=None, help='specify a ckpt directory to be evaluated if needed') parser.add_argument('--save_to_file', action='store_true', default=False, help='') args = parser.parse_args() cfg_from_yaml_file(args.cfg_file, cfg) cfg.TAG = Path(args.cfg_file).stem cfg.EXP_GROUP_PATH = '/'.join(args.cfg_file.split('/')[1:-1]) # remove 'cfgs' and 'xxxx.yaml' np.random.seed(1024) if args.set_cfgs is not None: cfg_from_list(args.set_cfgs, cfg) return args, cfg def eval_single_ckpt(model, test_loader, args, eval_output_dir, logger, epoch_id, dist_test=False): # load checkpoint model.load_params_from_file(filename=args.ckpt, logger=logger, to_cpu=dist_test) model.cuda() # tensorboard log if cfg.LOCAL_RANK == 0: tb_log = SummaryWriter(log_dir=str(eval_output_dir / ('tensorboard_%s' % cfg.DATA_CONFIG.DATA_SPLIT['test']))) # start evaluation tb_dict = eval_utils.eval_one_epoch( cfg, model, test_loader, epoch_id, logger, dist_test=dist_test, result_dir=eval_output_dir, save_to_file=args.save_to_file ) if cfg.LOCAL_RANK == 0: for key, val in tb_dict.items(): if 'curve' in key: for i in range(len(val['recalls'])): tb_log.add_scalars(key, dict(zip(val['precisions'].keys(), [_[i] for _ in val['precisions'].values()])), val['recalls'][i] * 1000) def get_no_evaluated_ckpt(ckpt_dir, ckpt_record_file, args): ckpt_list = glob.glob(os.path.join(ckpt_dir, '*checkpoint_epoch_*.pth')) ckpt_list.sort(key=os.path.getmtime) evaluated_ckpt_list = [float(x.strip()) for x in open(ckpt_record_file, 'r').readlines()] for cur_ckpt in ckpt_list: num_list = re.findall('checkpoint_epoch_(.*).pth', cur_ckpt) if num_list.__len__() == 0: continue epoch_id = num_list[-1] if 'optim' in epoch_id: continue if float(epoch_id) not in evaluated_ckpt_list and int(float(epoch_id)) >= args.start_epoch: return epoch_id, cur_ckpt return -1, None def repeat_eval_ckpt(model, test_loader, args, eval_output_dir, logger, ckpt_dir, dist_test=False): # evaluated ckpt record ckpt_record_file = eval_output_dir / ('eval_list_%s.txt' % cfg.DATA_CONFIG.DATA_SPLIT['test']) with open(ckpt_record_file, 'a'): pass # tensorboard log if cfg.LOCAL_RANK == 0: tb_log = SummaryWriter(log_dir=str(eval_output_dir / ('tensorboard_%s' % cfg.DATA_CONFIG.DATA_SPLIT['test']))) total_time = 0 first_eval = True while True: # check whether there is checkpoint which is not evaluated cur_epoch_id, cur_ckpt = get_no_evaluated_ckpt(ckpt_dir, ckpt_record_file, args) if cur_epoch_id == -1 or int(float(cur_epoch_id)) < args.start_epoch: wait_second = 30 if cfg.LOCAL_RANK == 0: print('Wait %s seconds for next check (progress: %.1f / %d minutes): %s \r' % (wait_second, total_time * 1.0 / 60, args.max_waiting_mins, ckpt_dir), end='', flush=True) time.sleep(wait_second) total_time += 30 if total_time > args.max_waiting_mins * 60 and (first_eval is False): break continue total_time = 0 first_eval = False model.load_params_from_file(filename=cur_ckpt, logger=logger, to_cpu=dist_test) model.cuda() # start evaluation cur_result_dir = eval_output_dir / ('epoch_%s' % cur_epoch_id) / cfg.DATA_CONFIG.DATA_SPLIT['test'] tb_dict = eval_utils.eval_one_epoch( cfg, model, test_loader, cur_epoch_id, logger, dist_test=dist_test, result_dir=cur_result_dir, save_to_file=args.save_to_file ) if cfg.LOCAL_RANK == 0: for key, val in tb_dict.items(): if not 'curve' in key: tb_log.add_scalar(key, val, cur_epoch_id) # record this epoch which has been evaluated with open(ckpt_record_file, 'a') as f: print('%s' % cur_epoch_id, file=f) logger.info('Epoch %s has been evaluated' % cur_epoch_id) # draw precision-recall curves for the last evaluation for key, val in tb_dict.items(): if 'curve' in key: for i in range(len(val['recalls'])): tb_log.add_scalars(key, dict(zip(val['precisions'].keys(), [_[i] for _ in val['precisions'].values()])), val['recalls'][i] * 1000) def main(): args, cfg = parse_config() if args.launcher == 'none': dist_test = False total_gpus = 1 else: total_gpus, cfg.LOCAL_RANK = getattr(common_utils, 'init_dist_%s' % args.launcher)( args.tcp_port, args.local_rank, backend='nccl' ) dist_test = True if args.batch_size is None: args.batch_size = cfg.OPTIMIZATION.BATCH_SIZE_PER_GPU else: assert args.batch_size % total_gpus == 0, 'Batch size should match the number of gpus' args.batch_size = args.batch_size // total_gpus output_dir = cfg.ROOT_DIR / 'output' / cfg.EXP_GROUP_PATH / cfg.TAG / args.extra_tag output_dir.mkdir(parents=True, exist_ok=True) eval_output_dir = output_dir / 'eval' if not args.eval_all: num_list = re.findall(r'\d+', args.ckpt) if args.ckpt is not None else [] epoch_id = num_list[-1] if num_list.__len__() > 0 else 'no_number' eval_output_dir = eval_output_dir / ('epoch_%s' % epoch_id) / cfg.DATA_CONFIG.DATA_SPLIT['test'] else: eval_output_dir = eval_output_dir / 'eval_all_default' if args.eval_tag is not None: eval_output_dir = eval_output_dir / args.eval_tag eval_output_dir.mkdir(parents=True, exist_ok=True) log_file = eval_output_dir / ('log_eval_%s.txt' % datetime.datetime.now().strftime('%Y%m%d-%H%M%S')) logger = common_utils.create_logger(log_file, rank=cfg.LOCAL_RANK) # log to file logger.info('**********************Start logging**********************') gpu_list = os.environ['CUDA_VISIBLE_DEVICES'] if 'CUDA_VISIBLE_DEVICES' in os.environ.keys() else 'ALL' logger.info('CUDA_VISIBLE_DEVICES=%s' % gpu_list) if dist_test: logger.info('total_batch_size: %d' % (total_gpus * args.batch_size)) for key, val in vars(args).items(): logger.info('{:16} {}'.format(key, val)) log_config_to_file(cfg, logger=logger) ckpt_dir = args.ckpt_dir if args.ckpt_dir is not None else output_dir / 'ckpt' test_set, test_loader, sampler = build_dataloader( dataset_cfg=cfg.DATA_CONFIG, class_names=cfg.CLASS_NAMES, batch_size=args.batch_size, dist=dist_test, workers=args.workers, logger=logger, training=False ) model = build_network(model_cfg=cfg.MODEL, num_class=len(cfg.CLASS_NAMES), dataset=test_set) with torch.no_grad(): if args.eval_all: repeat_eval_ckpt(model, test_loader, args, eval_output_dir, logger, ckpt_dir, dist_test=dist_test) else: eval_single_ckpt(model, test_loader, args, eval_output_dir, logger, epoch_id, dist_test=dist_test) if __name__ == '__main__': main()
9,291
40.855856
120
py
SASA
SASA-main/tools/demo.py
import argparse import glob from pathlib import Path import mayavi.mlab as mlab import numpy as np import torch from pcdet.config import cfg, cfg_from_yaml_file from pcdet.datasets import DatasetTemplate from pcdet.models import build_network, load_data_to_gpu from pcdet.utils import common_utils from visual_utils import visualize_utils as V class DemoDataset(DatasetTemplate): def __init__(self, dataset_cfg, class_names, training=True, root_path=None, logger=None, ext='.bin'): """ Args: root_path: dataset_cfg: class_names: training: logger: """ super().__init__( dataset_cfg=dataset_cfg, class_names=class_names, training=training, root_path=root_path, logger=logger ) self.root_path = root_path self.ext = ext data_file_list = glob.glob(str(root_path / f'*{self.ext}')) if self.root_path.is_dir() else [self.root_path] data_file_list.sort() self.sample_file_list = data_file_list def __len__(self): return len(self.sample_file_list) def __getitem__(self, index): if self.ext == '.bin': points = np.fromfile(self.sample_file_list[index], dtype=np.float32).reshape(-1, 4) elif self.ext == '.npy': points = np.load(self.sample_file_list[index]) else: raise NotImplementedError input_dict = { 'points': points, 'frame_id': index, } data_dict = self.prepare_data(data_dict=input_dict) return data_dict def parse_config(): parser = argparse.ArgumentParser(description='arg parser') parser.add_argument('--cfg_file', type=str, default='cfgs/kitti_models/second.yaml', help='specify the config for demo') parser.add_argument('--data_path', type=str, default='demo_data', help='specify the point cloud data file or directory') parser.add_argument('--ckpt', type=str, default=None, help='specify the pretrained model') parser.add_argument('--ext', type=str, default='.bin', help='specify the extension of your point cloud data file') args = parser.parse_args() cfg_from_yaml_file(args.cfg_file, cfg) return args, cfg def main(): args, cfg = parse_config() logger = common_utils.create_logger() logger.info('-----------------Quick Demo of OpenPCDet-------------------------') demo_dataset = DemoDataset( dataset_cfg=cfg.DATA_CONFIG, class_names=cfg.CLASS_NAMES, training=False, root_path=Path(args.data_path), ext=args.ext, logger=logger ) logger.info(f'Total number of samples: \t{len(demo_dataset)}') model = build_network(model_cfg=cfg.MODEL, num_class=len(cfg.CLASS_NAMES), dataset=demo_dataset) model.load_params_from_file(filename=args.ckpt, logger=logger, to_cpu=True) model.cuda() model.eval() with torch.no_grad(): for idx, data_dict in enumerate(demo_dataset): logger.info(f'Visualized sample index: \t{idx + 1}') data_dict = demo_dataset.collate_batch([data_dict]) load_data_to_gpu(data_dict) pred_dicts, _ = model.forward(data_dict) V.draw_scenes( points=data_dict['points'][:, 1:], ref_boxes=pred_dicts[0]['pred_boxes'], ref_scores=pred_dicts[0]['pred_scores'], ref_labels=pred_dicts[0]['pred_labels'] ) mlab.show(stop=True) logger.info('Demo done.') if __name__ == '__main__': main()
3,575
33.384615
118
py
SASA
SASA-main/tools/train.py
import argparse import datetime import glob import os from pathlib import Path from test import repeat_eval_ckpt import torch import torch.distributed as dist import torch.nn as nn from tensorboardX import SummaryWriter from pcdet.config import cfg, cfg_from_list, cfg_from_yaml_file, log_config_to_file from pcdet.datasets import build_dataloader from pcdet.models import build_network, model_fn_decorator from pcdet.utils import common_utils from train_utils.optimization import build_optimizer, build_scheduler from train_utils.train_utils import train_model def parse_config(): parser = argparse.ArgumentParser(description='arg parser') parser.add_argument('--cfg_file', type=str, default=None, help='specify the config for training') parser.add_argument('--batch_size', type=int, default=None, required=False, help='batch size for training') parser.add_argument('--epochs', type=int, default=None, required=False, help='number of epochs to train for') parser.add_argument('--workers', type=int, default=8, help='number of workers for dataloader') parser.add_argument('--extra_tag', type=str, default='default', help='extra tag for this experiment') parser.add_argument('--ckpt', type=str, default=None, help='checkpoint to start from') parser.add_argument('--pretrained_model', type=str, default=None, help='pretrained_model') parser.add_argument('--launcher', choices=['none', 'pytorch', 'slurm'], default='none') parser.add_argument('--tcp_port', type=int, default=18888, help='tcp port for distrbuted training') parser.add_argument('--sync_bn', action='store_true', default=False, help='whether to use sync bn') parser.add_argument('--fix_random_seed', type=int, default=666, help='random seed') parser.add_argument('--ckpt_save_interval', type=int, default=1, help='number of training epochs') parser.add_argument('--local_rank', type=int, default=0, help='local rank for distributed training') parser.add_argument('--max_ckpt_save_num', type=int, default=10, help='max number of saved checkpoint') parser.add_argument('--merge_all_iters_to_one_epoch', action='store_true', default=False, help='') parser.add_argument('--set', dest='set_cfgs', default=None, nargs=argparse.REMAINDER, help='set extra config keys if needed') parser.add_argument('--max_waiting_mins', type=int, default=0, help='max waiting minutes') parser.add_argument('--start_epoch', type=int, default=0, help='') parser.add_argument('--save_to_file', action='store_true', default=False, help='') args = parser.parse_args() cfg_from_yaml_file(args.cfg_file, cfg) cfg.TAG = Path(args.cfg_file).stem cfg.EXP_GROUP_PATH = '/'.join(args.cfg_file.split('/')[1:-1]) # remove 'cfgs' and 'xxxx.yaml' if args.set_cfgs is not None: cfg_from_list(args.set_cfgs, cfg) return args, cfg def main(): args, cfg = parse_config() if args.launcher == 'none': dist_train = False total_gpus = 1 else: total_gpus, cfg.LOCAL_RANK = getattr(common_utils, 'init_dist_%s' % args.launcher)( args.tcp_port, args.local_rank, backend='nccl' ) dist_train = True if args.batch_size is None: args.batch_size = cfg.OPTIMIZATION.BATCH_SIZE_PER_GPU else: assert args.batch_size % total_gpus == 0, 'Batch size should match the number of gpus' args.batch_size = args.batch_size // total_gpus args.epochs = cfg.OPTIMIZATION.NUM_EPOCHS if args.epochs is None else args.epochs if not args.fix_random_seed == 0: common_utils.set_random_seed(args.fix_random_seed) output_dir = cfg.ROOT_DIR / 'output' / cfg.EXP_GROUP_PATH / cfg.TAG / args.extra_tag ckpt_dir = output_dir / 'ckpt' output_dir.mkdir(parents=True, exist_ok=True) ckpt_dir.mkdir(parents=True, exist_ok=True) log_file = output_dir / ('log_train_%s.txt' % datetime.datetime.now().strftime('%Y%m%d-%H%M%S')) logger = common_utils.create_logger(log_file, rank=cfg.LOCAL_RANK) # log to file logger.info('**********************Start logging**********************') gpu_list = os.environ['CUDA_VISIBLE_DEVICES'] if 'CUDA_VISIBLE_DEVICES' in os.environ.keys() else 'ALL' logger.info('CUDA_VISIBLE_DEVICES=%s' % gpu_list) if dist_train: logger.info('total_batch_size: %d' % (total_gpus * args.batch_size)) for key, val in vars(args).items(): logger.info('{:16} {}'.format(key, val)) log_config_to_file(cfg, logger=logger) if cfg.LOCAL_RANK == 0: os.system('cp %s %s' % (args.cfg_file, output_dir)) tb_log = SummaryWriter(log_dir=str(output_dir / 'tensorboard')) if cfg.LOCAL_RANK == 0 else None # -----------------------create dataloader & network & optimizer--------------------------- train_set, train_loader, train_sampler = build_dataloader( dataset_cfg=cfg.DATA_CONFIG, class_names=cfg.CLASS_NAMES, batch_size=args.batch_size, dist=dist_train, workers=args.workers, logger=logger, training=True, merge_all_iters_to_one_epoch=args.merge_all_iters_to_one_epoch, total_epochs=args.epochs ) model = build_network(model_cfg=cfg.MODEL, num_class=len(cfg.CLASS_NAMES), dataset=train_set) if args.sync_bn: model = torch.nn.SyncBatchNorm.convert_sync_batchnorm(model) model.cuda() optimizer = build_optimizer(model, cfg.OPTIMIZATION) # load checkpoint if it is possible start_epoch = it = 0 last_epoch = -1 if args.pretrained_model is not None: model.load_params_from_file(filename=args.pretrained_model, to_cpu=dist, logger=logger) if args.ckpt is not None: it, start_epoch = model.load_params_with_optimizer(args.ckpt, to_cpu=dist, optimizer=optimizer, logger=logger) last_epoch = start_epoch + 1 else: ckpt_list = glob.glob(str(ckpt_dir / '*checkpoint_epoch_*.pth')) if len(ckpt_list) > 0: ckpt_list.sort(key=os.path.getmtime) it, start_epoch = model.load_params_with_optimizer( ckpt_list[-1], to_cpu=dist, optimizer=optimizer, logger=logger ) last_epoch = start_epoch + 1 model.train() # before wrap to DistributedDataParallel to support fixed some parameters if dist_train: model = nn.parallel.DistributedDataParallel(model, device_ids=[cfg.LOCAL_RANK % torch.cuda.device_count()]) logger.info(model) total_iters_each_epoch = len(train_loader) if args.merge_all_iters_to_one_epoch: total_iters_each_epoch = total_iters_each_epoch // args.epochs lr_scheduler, lr_warmup_scheduler = build_scheduler( optimizer, total_iters_each_epoch=total_iters_each_epoch, total_epochs=args.epochs, last_epoch=last_epoch, optim_cfg=cfg.OPTIMIZATION ) # -----------------------start training--------------------------- logger.info('**********************Start training %s/%s(%s)**********************' % (cfg.EXP_GROUP_PATH, cfg.TAG, args.extra_tag)) train_model( model, optimizer, train_loader, model_func=model_fn_decorator(), lr_scheduler=lr_scheduler, optim_cfg=cfg.OPTIMIZATION, start_epoch=start_epoch, total_epochs=args.epochs, start_iter=it, rank=cfg.LOCAL_RANK, tb_log=tb_log, ckpt_save_dir=ckpt_dir, train_sampler=train_sampler, lr_warmup_scheduler=lr_warmup_scheduler, ckpt_save_interval=args.ckpt_save_interval, max_ckpt_save_num=args.max_ckpt_save_num, merge_all_iters_to_one_epoch=args.merge_all_iters_to_one_epoch ) logger.info('**********************End training %s/%s(%s)**********************\n\n\n' % (cfg.EXP_GROUP_PATH, cfg.TAG, args.extra_tag)) logger.info('**********************Start evaluation %s/%s(%s)**********************' % (cfg.EXP_GROUP_PATH, cfg.TAG, args.extra_tag)) test_set, test_loader, sampler = build_dataloader( dataset_cfg=cfg.DATA_CONFIG, class_names=cfg.CLASS_NAMES, batch_size=args.batch_size, dist=dist_train, workers=args.workers, logger=logger, training=False ) eval_output_dir = output_dir / 'eval' / 'eval_with_train' eval_output_dir.mkdir(parents=True, exist_ok=True) args.start_epoch = max(args.epochs - 10, 0) # Only evaluate the last 10 epochs repeat_eval_ckpt( model.module if dist_train else model, test_loader, args, eval_output_dir, logger, ckpt_dir, dist_test=dist_train ) logger.info('**********************End evaluation %s/%s(%s)**********************' % (cfg.EXP_GROUP_PATH, cfg.TAG, args.extra_tag)) if __name__ == '__main__': main()
8,839
42.762376
118
py
SASA
SASA-main/tools/eval_utils/eval_utils.py
import pickle import time import numpy as np import torch import tqdm from pcdet.models import load_data_to_gpu from pcdet.utils import common_utils def statistics_info(cfg, ret_dict, metric, disp_dict): for key in metric.keys(): if key in ret_dict: metric[key] += ret_dict[key] min_thresh = cfg.MODEL.POST_PROCESSING.RECALL_THRESH_LIST[0] disp_dict['recall_%s' % str(min_thresh)] = \ '(%d, %d) / %d' % (metric['recall_roi_%s' % str(min_thresh)], metric['recall_rcnn_%s' % str(min_thresh)], metric['gt_num']) def eval_one_epoch(cfg, model, dataloader, epoch_id, logger, dist_test=False, save_to_file=False, result_dir=None): result_dir.mkdir(parents=True, exist_ok=True) final_output_dir = result_dir / 'final_result' / 'data' if save_to_file: final_output_dir.mkdir(parents=True, exist_ok=True) metric = { 'gt_num': 0, } for cur_thresh in cfg.MODEL.POST_PROCESSING.RECALL_THRESH_LIST: metric['recall_roi_%s' % str(cur_thresh)] = 0 metric['recall_rcnn_%s' % str(cur_thresh)] = 0 # initialize customized statistics if hasattr(model, 'init_recall_record'): model.init_recall_record(metric) dataset = dataloader.dataset class_names = dataset.class_names det_annos = [] logger.info('*************** EPOCH %s EVALUATION *****************' % epoch_id) if dist_test: num_gpus = torch.cuda.device_count() local_rank = cfg.LOCAL_RANK % num_gpus model = torch.nn.parallel.DistributedDataParallel( model, device_ids=[local_rank], broadcast_buffers=False ) model.eval() if cfg.LOCAL_RANK == 0: progress_bar = tqdm.tqdm(total=len(dataloader), leave=True, desc='eval', dynamic_ncols=True) start_time = time.time() for i, batch_dict in enumerate(dataloader): load_data_to_gpu(batch_dict) with torch.no_grad(): pred_dicts, ret_dict = model(batch_dict) disp_dict = {} statistics_info(cfg, ret_dict, metric, disp_dict) annos = dataset.generate_prediction_dicts( batch_dict, pred_dicts, class_names, output_path=final_output_dir if save_to_file else None ) det_annos += annos if cfg.LOCAL_RANK == 0: progress_bar.set_postfix(disp_dict) progress_bar.update() if cfg.LOCAL_RANK == 0: progress_bar.close() if dist_test: rank, world_size = common_utils.get_dist_info() det_annos = common_utils.merge_results_dist(det_annos, len(dataset), tmpdir=result_dir / 'tmpdir') metric = common_utils.merge_results_dist([metric], world_size, tmpdir=result_dir / 'tmpdir') logger.info('*************** Performance of EPOCH %s *****************' % epoch_id) sec_per_example = (time.time() - start_time) / len(dataloader.dataset) logger.info('Generate label finished(sec_per_example: %.4f second).' % sec_per_example) if cfg.LOCAL_RANK != 0: return {} ret_dict = {} if dist_test: for key, val in metric[0].items(): for k in range(1, world_size): metric[0][key] += metric[k][key] metric = metric[0] gt_num_cnt = metric['gt_num'] for cur_thresh in cfg.MODEL.POST_PROCESSING.RECALL_THRESH_LIST: cur_roi_recall = metric['recall_roi_%s' % str(cur_thresh)] / max(gt_num_cnt, 1) cur_rcnn_recall = metric['recall_rcnn_%s' % str(cur_thresh)] / max(gt_num_cnt, 1) logger.info('recall_roi_%s: %f' % (cur_thresh, cur_roi_recall)) logger.info('recall_rcnn_%s: %f' % (cur_thresh, cur_rcnn_recall)) ret_dict['recall/roi_%s' % str(cur_thresh)] = cur_roi_recall ret_dict['recall/rcnn_%s' % str(cur_thresh)] = cur_rcnn_recall # print customized statistics if hasattr(model, 'disp_recall_record'): model.disp_recall_record(metric, logger, sample_num=len(dataloader.dataset)) total_pred_objects = 0 for anno in det_annos: total_pred_objects += anno['name'].__len__() logger.info('Average predicted number of objects(%d samples): %.3f' % (len(det_annos), total_pred_objects / max(1, len(det_annos)))) with open(result_dir / 'result.pkl', 'wb') as f: pickle.dump(det_annos, f) result_str, result_dict = dataset.evaluation( det_annos, class_names, eval_metric=cfg.MODEL.POST_PROCESSING.EVAL_METRIC, output_path=final_output_dir ) logger.info(result_str) ret_dict.update(result_dict) logger.info('Result is save to %s' % result_dir) logger.info('****************Evaluation done.*****************') return ret_dict if __name__ == '__main__': pass
4,772
34.887218
131
py
SASA
SASA-main/tools/train_utils/train_utils.py
import glob import os import torch import tqdm from torch.nn.utils import clip_grad_norm_ def train_one_epoch(model, optimizer, train_loader, model_func, lr_scheduler, accumulated_iter, optim_cfg, rank, tbar, total_it_each_epoch, dataloader_iter, tb_log=None, leave_pbar=False): if total_it_each_epoch == len(train_loader): dataloader_iter = iter(train_loader) if rank == 0: pbar = tqdm.tqdm(total=total_it_each_epoch, leave=leave_pbar, desc='train', dynamic_ncols=True) for cur_it in range(total_it_each_epoch): try: batch = next(dataloader_iter) except StopIteration: dataloader_iter = iter(train_loader) batch = next(dataloader_iter) print('new iters') lr_scheduler.step(accumulated_iter) try: cur_lr = float(optimizer.lr) except: cur_lr = optimizer.param_groups[0]['lr'] if tb_log is not None: tb_log.add_scalar('meta_data/learning_rate', cur_lr, accumulated_iter) model.train() optimizer.zero_grad() loss, tb_dict, disp_dict = model_func(model, batch) loss.backward() clip_grad_norm_(model.parameters(), optim_cfg.GRAD_NORM_CLIP) optimizer.step() accumulated_iter += 1 disp_dict.update({'loss': loss.item(), 'lr': cur_lr}) # log to console and tensorboard if rank == 0: pbar.update() pbar.set_postfix(dict(total_it=accumulated_iter)) tbar.set_postfix(disp_dict) tbar.refresh() if tb_log is not None: tb_log.add_scalar('train/loss', loss, accumulated_iter) tb_log.add_scalar('meta_data/learning_rate', cur_lr, accumulated_iter) for key, val in tb_dict.items(): tb_log.add_scalar('train/' + key, val, accumulated_iter) if rank == 0: pbar.close() return accumulated_iter def train_model(model, optimizer, train_loader, model_func, lr_scheduler, optim_cfg, start_epoch, total_epochs, start_iter, rank, tb_log, ckpt_save_dir, train_sampler=None, lr_warmup_scheduler=None, ckpt_save_interval=1, max_ckpt_save_num=50, merge_all_iters_to_one_epoch=False): accumulated_iter = start_iter with tqdm.trange(start_epoch, total_epochs, desc='epochs', dynamic_ncols=True, leave=(rank == 0)) as tbar: total_it_each_epoch = len(train_loader) if merge_all_iters_to_one_epoch: assert hasattr(train_loader.dataset, 'merge_all_iters_to_one_epoch') train_loader.dataset.merge_all_iters_to_one_epoch(merge=True, epochs=total_epochs) total_it_each_epoch = len(train_loader) // max(total_epochs, 1) dataloader_iter = iter(train_loader) for cur_epoch in tbar: if train_sampler is not None: train_sampler.set_epoch(cur_epoch) # train one epoch if lr_warmup_scheduler is not None and cur_epoch < optim_cfg.WARMUP_EPOCH: cur_scheduler = lr_warmup_scheduler else: cur_scheduler = lr_scheduler accumulated_iter = train_one_epoch( model, optimizer, train_loader, model_func, lr_scheduler=cur_scheduler, accumulated_iter=accumulated_iter, optim_cfg=optim_cfg, rank=rank, tbar=tbar, tb_log=tb_log, leave_pbar=(cur_epoch + 1 == total_epochs), total_it_each_epoch=total_it_each_epoch, dataloader_iter=dataloader_iter ) # save trained model trained_epoch = cur_epoch + 1 if trained_epoch % ckpt_save_interval == 0 and rank == 0: ckpt_list = glob.glob(str(ckpt_save_dir / 'checkpoint_epoch_*.pth')) ckpt_list.sort(key=os.path.getmtime) if ckpt_list.__len__() >= max_ckpt_save_num: for cur_file_idx in range(0, len(ckpt_list) - max_ckpt_save_num + 1): os.remove(ckpt_list[cur_file_idx]) ckpt_name = ckpt_save_dir / ('checkpoint_epoch_%d' % trained_epoch) save_checkpoint( checkpoint_state(model, optimizer, trained_epoch, accumulated_iter), filename=ckpt_name, ) def model_state_to_cpu(model_state): model_state_cpu = type(model_state)() # ordered dict for key, val in model_state.items(): model_state_cpu[key] = val.cpu() return model_state_cpu def checkpoint_state(model=None, optimizer=None, epoch=None, it=None): optim_state = optimizer.state_dict() if optimizer is not None else None if model is not None: if isinstance(model, torch.nn.parallel.DistributedDataParallel): model_state = model_state_to_cpu(model.module.state_dict()) else: model_state = model.state_dict() else: model_state = None try: import pcdet version = 'pcdet+' + pcdet.__version__ except: version = 'none' return {'epoch': epoch, 'it': it, 'model_state': model_state, 'optimizer_state': optim_state, 'version': version} def save_checkpoint(state, filename='checkpoint'): if False and 'optimizer_state' in state: optimizer_state = state['optimizer_state'] state.pop('optimizer_state', None) optimizer_filename = '{}_optim.pth'.format(filename) torch.save({'optimizer_state': optimizer_state}, optimizer_filename) filename = '{}.pth'.format(filename) torch.save(state, filename)
5,667
37.297297
117
py
SASA
SASA-main/tools/train_utils/optimization/fastai_optim.py
# This file is modified from https://github.com/traveller59/second.pytorch from collections import Iterable import torch from torch import nn from torch._utils import _unflatten_dense_tensors from torch.nn.utils import parameters_to_vector bn_types = (nn.BatchNorm1d, nn.BatchNorm2d, nn.BatchNorm3d, nn.SyncBatchNorm) def split_bn_bias(layer_groups): "Split the layers in `layer_groups` into batchnorm (`bn_types`) and non-batchnorm groups." split_groups = [] for l in layer_groups: l1, l2 = [], [] for c in l.children(): if isinstance(c, bn_types): l2.append(c) else: l1.append(c) split_groups += [nn.Sequential(*l1), nn.Sequential(*l2)] return split_groups def get_master(layer_groups, flat_master: bool = False): "Return two lists, one for the model parameters in FP16 and one for the master parameters in FP32." split_groups = split_bn_bias(layer_groups) model_params = [[param for param in lg.parameters() if param.requires_grad] for lg in split_groups] if flat_master: master_params = [] for lg in model_params: if len(lg) != 0: mp = parameters_to_vector([param.data.float() for param in lg]) mp = torch.nn.Parameter(mp, requires_grad=True) if mp.grad is None: mp.grad = mp.new(*mp.size()) master_params.append([mp]) else: master_params.append([]) return model_params, master_params else: master_params = [[param.clone().float().detach() for param in lg] for lg in model_params] for mp in master_params: for param in mp: param.requires_grad = True return model_params, master_params def model_g2master_g(model_params, master_params, flat_master: bool = False) -> None: "Copy the `model_params` gradients to `master_params` for the optimizer step." if flat_master: for model_group, master_group in zip(model_params, master_params): if len(master_group) != 0: master_group[0].grad.data.copy_(parameters_to_vector([p.grad.data.float() for p in model_group])) else: for model_group, master_group in zip(model_params, master_params): for model, master in zip(model_group, master_group): if model.grad is not None: if master.grad is None: master.grad = master.data.new(*master.data.size()) master.grad.data.copy_(model.grad.data) else: master.grad = None def master2model(model_params, master_params, flat_master: bool = False) -> None: "Copy `master_params` to `model_params`." if flat_master: for model_group, master_group in zip(model_params, master_params): if len(model_group) != 0: for model, master in zip(model_group, _unflatten_dense_tensors(master_group[0].data, model_group)): model.data.copy_(master) else: for model_group, master_group in zip(model_params, master_params): for model, master in zip(model_group, master_group): model.data.copy_(master.data) def listify(p=None, q=None): "Make `p` listy and the same length as `q`." if p is None: p = [] elif isinstance(p, str): p = [p] elif not isinstance(p, Iterable): p = [p] n = q if type(q) == int else len(p) if q is None else len(q) if len(p) == 1: p = p * n assert len(p) == n, f'List len mismatch ({len(p)} vs {n})' return list(p) def trainable_params(m: nn.Module): "Return list of trainable params in `m`." res = filter(lambda p: p.requires_grad, m.parameters()) return res def is_tuple(x) -> bool: return isinstance(x, tuple) # copy from fastai. class OptimWrapper(): "Basic wrapper around `opt` to simplify hyper-parameters changes." def __init__(self, opt, wd, true_wd: bool = False, bn_wd: bool = True): self.opt, self.true_wd, self.bn_wd = opt, true_wd, bn_wd self.opt_keys = list(self.opt.param_groups[0].keys()) self.opt_keys.remove('params') self.read_defaults() self.wd = wd @classmethod def create(cls, opt_func, lr, layer_groups, **kwargs): "Create an `optim.Optimizer` from `opt_func` with `lr`. Set lr on `layer_groups`." split_groups = split_bn_bias(layer_groups) opt = opt_func([{'params': trainable_params(l), 'lr': 0} for l in split_groups]) opt = cls(opt, **kwargs) opt.lr, opt.opt_func = listify(lr, layer_groups), opt_func return opt def new(self, layer_groups): "Create a new `OptimWrapper` from `self` with another `layer_groups` but the same hyper-parameters." opt_func = getattr(self, 'opt_func', self.opt.__class__) split_groups = split_bn_bias(layer_groups) opt = opt_func([{'params': trainable_params(l), 'lr': 0} for l in split_groups]) return self.create(opt_func, self.lr, layer_groups, wd=self.wd, true_wd=self.true_wd, bn_wd=self.bn_wd) def __repr__(self) -> str: return f'OptimWrapper over {repr(self.opt)}.\nTrue weight decay: {self.true_wd}' # Pytorch optimizer methods def step(self) -> None: "Set weight decay and step optimizer." # weight decay outside of optimizer step (AdamW) if self.true_wd: for lr, wd, pg1, pg2 in zip(self._lr, self._wd, self.opt.param_groups[::2], self.opt.param_groups[1::2]): for p in pg1['params']: # When some parameters are fixed: Shaoshuai Shi if p.requires_grad is False: continue p.data.mul_(1 - wd * lr) if self.bn_wd: for p in pg2['params']: # When some parameters are fixed: Shaoshuai Shi if p.requires_grad is False: continue p.data.mul_(1 - wd * lr) self.set_val('weight_decay', listify(0, self._wd)) self.opt.step() def zero_grad(self) -> None: "Clear optimizer gradients." self.opt.zero_grad() # Passthrough to the inner opt. def __getattr__(self, k: str): return getattr(self.opt, k, None) def clear(self): "Reset the state of the inner optimizer." sd = self.state_dict() sd['state'] = {} self.load_state_dict(sd) # Hyperparameters as properties @property def lr(self) -> float: return self._lr[-1] @lr.setter def lr(self, val: float) -> None: self._lr = self.set_val('lr', listify(val, self._lr)) @property def mom(self) -> float: return self._mom[-1] @mom.setter def mom(self, val: float) -> None: if 'momentum' in self.opt_keys: self.set_val('momentum', listify(val, self._mom)) elif 'betas' in self.opt_keys: self.set_val('betas', (listify(val, self._mom), self._beta)) self._mom = listify(val, self._mom) @property def beta(self) -> float: return None if self._beta is None else self._beta[-1] @beta.setter def beta(self, val: float) -> None: "Set beta (or alpha as makes sense for given optimizer)." if val is None: return if 'betas' in self.opt_keys: self.set_val('betas', (self._mom, listify(val, self._beta))) elif 'alpha' in self.opt_keys: self.set_val('alpha', listify(val, self._beta)) self._beta = listify(val, self._beta) @property def wd(self) -> float: return self._wd[-1] @wd.setter def wd(self, val: float) -> None: "Set weight decay." if not self.true_wd: self.set_val('weight_decay', listify(val, self._wd), bn_groups=self.bn_wd) self._wd = listify(val, self._wd) # Helper functions def read_defaults(self) -> None: "Read the values inside the optimizer for the hyper-parameters." self._beta = None if 'lr' in self.opt_keys: self._lr = self.read_val('lr') if 'momentum' in self.opt_keys: self._mom = self.read_val('momentum') if 'alpha' in self.opt_keys: self._beta = self.read_val('alpha') if 'betas' in self.opt_keys: self._mom, self._beta = self.read_val('betas') if 'weight_decay' in self.opt_keys: self._wd = self.read_val('weight_decay') def set_val(self, key: str, val, bn_groups: bool = True): "Set `val` inside the optimizer dictionary at `key`." if is_tuple(val): val = [(v1, v2) for v1, v2 in zip(*val)] for v, pg1, pg2 in zip(val, self.opt.param_groups[::2], self.opt.param_groups[1::2]): pg1[key] = v if bn_groups: pg2[key] = v return val def read_val(self, key: str): "Read a hyperparameter `key` in the optimizer dictionary." val = [pg[key] for pg in self.opt.param_groups[::2]] if is_tuple(val[0]): val = [o[0] for o in val], [o[1] for o in val] return val class FastAIMixedOptim(OptimWrapper): @classmethod def create(cls, opt_func, lr, layer_groups, model, flat_master=False, loss_scale=512.0, **kwargs): "Create an `optim.Optimizer` from `opt_func` with `lr`. Set lr on `layer_groups`." opt = OptimWrapper.create(opt_func, lr, layer_groups, **kwargs) opt.model_params, opt.master_params = get_master(layer_groups, flat_master) opt.flat_master = flat_master opt.loss_scale = loss_scale opt.model = model # Changes the optimizer so that the optimization step is done in FP32. # opt = self.learn.opt mom, wd, beta = opt.mom, opt.wd, opt.beta lrs = [lr for lr in opt._lr for _ in range(2)] opt_params = [{'params': mp, 'lr': lr} for mp, lr in zip(opt.master_params, lrs)] opt.opt = opt_func(opt_params) opt.mom, opt.wd, opt.beta = mom, wd, beta return opt def step(self): model_g2master_g(self.model_params, self.master_params, self.flat_master) for group in self.master_params: for param in group: param.grad.div_(self.loss_scale) super(FastAIMixedOptim, self).step() self.model.zero_grad() # Update the params from master to model. master2model(self.model_params, self.master_params, self.flat_master)
10,477
38.992366
117
py
SASA
SASA-main/tools/train_utils/optimization/learning_schedules_fastai.py
# This file is modified from https://github.com/traveller59/second.pytorch import math from functools import partial import numpy as np import torch.optim.lr_scheduler as lr_sched from .fastai_optim import OptimWrapper class LRSchedulerStep(object): def __init__(self, fai_optimizer: OptimWrapper, total_step, lr_phases, mom_phases): # if not isinstance(fai_optimizer, OptimWrapper): # raise TypeError('{} is not a fastai OptimWrapper'.format( # type(fai_optimizer).__name__)) self.optimizer = fai_optimizer self.total_step = total_step self.lr_phases = [] for i, (start, lambda_func) in enumerate(lr_phases): if len(self.lr_phases) != 0: assert self.lr_phases[-1][0] < start if isinstance(lambda_func, str): lambda_func = eval(lambda_func) if i < len(lr_phases) - 1: self.lr_phases.append((int(start * total_step), int(lr_phases[i + 1][0] * total_step), lambda_func)) else: self.lr_phases.append((int(start * total_step), total_step, lambda_func)) assert self.lr_phases[0][0] == 0 self.mom_phases = [] for i, (start, lambda_func) in enumerate(mom_phases): if len(self.mom_phases) != 0: assert self.mom_phases[-1][0] < start if isinstance(lambda_func, str): lambda_func = eval(lambda_func) if i < len(mom_phases) - 1: self.mom_phases.append((int(start * total_step), int(mom_phases[i + 1][0] * total_step), lambda_func)) else: self.mom_phases.append((int(start * total_step), total_step, lambda_func)) assert self.mom_phases[0][0] == 0 def step(self, step): for start, end, func in self.lr_phases: if step >= start: self.optimizer.lr = func((step - start) / (end - start)) for start, end, func in self.mom_phases: if step >= start: self.optimizer.mom = func((step - start) / (end - start)) def annealing_cos(start, end, pct): # print(pct, start, end) "Cosine anneal from `start` to `end` as pct goes from 0.0 to 1.0." cos_out = np.cos(np.pi * pct) + 1 return end + (start - end) / 2 * cos_out class OneCycle(LRSchedulerStep): def __init__(self, fai_optimizer, total_step, lr_max, moms, div_factor, pct_start): self.lr_max = lr_max self.moms = moms self.div_factor = div_factor self.pct_start = pct_start a1 = int(total_step * self.pct_start) a2 = total_step - a1 low_lr = self.lr_max / self.div_factor lr_phases = ((0, partial(annealing_cos, low_lr, self.lr_max)), (self.pct_start, partial(annealing_cos, self.lr_max, low_lr / 1e4))) mom_phases = ((0, partial(annealing_cos, *self.moms)), (self.pct_start, partial(annealing_cos, *self.moms[::-1]))) fai_optimizer.lr, fai_optimizer.mom = low_lr, self.moms[0] super().__init__(fai_optimizer, total_step, lr_phases, mom_phases) class CosineWarmupLR(lr_sched._LRScheduler): def __init__(self, optimizer, T_max, eta_min=0, last_epoch=-1): self.T_max = T_max self.eta_min = eta_min super(CosineWarmupLR, self).__init__(optimizer, last_epoch) def get_lr(self): return [self.eta_min + (base_lr - self.eta_min) * (1 - math.cos(math.pi * self.last_epoch / self.T_max)) / 2 for base_lr in self.base_lrs] class FakeOptim: def __init__(self): self.lr = 0 self.mom = 0 if __name__ == "__main__": import matplotlib.pyplot as plt opt = FakeOptim() # 3e-3, wd=0.4, div_factor=10 schd = OneCycle(opt, 100, 3e-3, (0.95, 0.85), 10.0, 0.1) lrs = [] moms = [] for i in range(100): schd.step(i) lrs.append(opt.lr) moms.append(opt.mom) plt.plot(lrs) # plt.plot(moms) plt.show() plt.plot(moms) plt.show()
4,169
35.26087
118
py
SASA
SASA-main/tools/train_utils/optimization/__init__.py
from functools import partial import torch.nn as nn import torch.optim as optim import torch.optim.lr_scheduler as lr_sched from .fastai_optim import OptimWrapper from .learning_schedules_fastai import CosineWarmupLR, OneCycle def build_optimizer(model, optim_cfg): if optim_cfg.OPTIMIZER == 'sgd': optimizer = optim.SGD( model.parameters(), lr=optim_cfg.LR, weight_decay=optim_cfg.WEIGHT_DECAY, momentum=optim_cfg.MOMENTUM ) elif optim_cfg.OPTIMIZER == 'adam' or optim_cfg.OPTIMIZER == 'adam_onecycle': def children(m: nn.Module): return list(m.children()) def num_children(m: nn.Module) -> int: return len(children(m)) flatten_model = lambda m: sum(map(flatten_model, m.children()), []) if num_children(m) else [m] get_layer_groups = lambda m: [nn.Sequential(*flatten_model(m))] optimizer_func = partial(optim.Adam, betas=(0.9, 0.99)) optimizer = OptimWrapper.create( optimizer_func, 3e-3, get_layer_groups(model), wd=optim_cfg.WEIGHT_DECAY, true_wd=True, bn_wd=True ) else: raise NotImplementedError return optimizer def build_scheduler(optimizer, total_iters_each_epoch, total_epochs, last_epoch, optim_cfg): decay_steps = [x * total_iters_each_epoch for x in optim_cfg.DECAY_STEP_LIST] def lr_lbmd(cur_epoch): cur_decay = 1 for decay_step in decay_steps: if cur_epoch >= decay_step: cur_decay = cur_decay * optim_cfg.LR_DECAY return max(cur_decay, optim_cfg.LR_CLIP / optim_cfg.LR) lr_warmup_scheduler = None total_steps = total_iters_each_epoch * total_epochs if optim_cfg.OPTIMIZER == 'adam_onecycle': lr_scheduler = OneCycle( optimizer, total_steps, optim_cfg.LR, list(optim_cfg.MOMS), optim_cfg.DIV_FACTOR, optim_cfg.PCT_START ) else: lr_scheduler = lr_sched.LambdaLR(optimizer, lr_lbmd, last_epoch=last_epoch) if optim_cfg.LR_WARMUP: lr_warmup_scheduler = CosineWarmupLR( optimizer, T_max=optim_cfg.WARMUP_EPOCH * len(total_iters_each_epoch), eta_min=optim_cfg.LR / optim_cfg.DIV_FACTOR ) return lr_scheduler, lr_warmup_scheduler
2,289
35.935484
113
py
SASA
SASA-main/tools/visual_utils/visualize_utils.py
import mayavi.mlab as mlab import numpy as np import torch box_colormap = [ [1, 1, 1], [0, 1, 0], [0, 1, 1], [1, 1, 0], ] def check_numpy_to_torch(x): if isinstance(x, np.ndarray): return torch.from_numpy(x).float(), True return x, False def rotate_points_along_z(points, angle): """ Args: points: (B, N, 3 + C) angle: (B), angle along z-axis, angle increases x ==> y Returns: """ points, is_numpy = check_numpy_to_torch(points) angle, _ = check_numpy_to_torch(angle) cosa = torch.cos(angle) sina = torch.sin(angle) zeros = angle.new_zeros(points.shape[0]) ones = angle.new_ones(points.shape[0]) rot_matrix = torch.stack(( cosa, sina, zeros, -sina, cosa, zeros, zeros, zeros, ones ), dim=1).view(-1, 3, 3).float() points_rot = torch.matmul(points[:, :, 0:3], rot_matrix) points_rot = torch.cat((points_rot, points[:, :, 3:]), dim=-1) return points_rot.numpy() if is_numpy else points_rot def boxes_to_corners_3d(boxes3d): """ 7 -------- 4 /| /| 6 -------- 5 . | | | | . 3 -------- 0 |/ |/ 2 -------- 1 Args: boxes3d: (N, 7) [x, y, z, dx, dy, dz, heading], (x, y, z) is the box center Returns: """ boxes3d, is_numpy = check_numpy_to_torch(boxes3d) template = boxes3d.new_tensor(( [1, 1, -1], [1, -1, -1], [-1, -1, -1], [-1, 1, -1], [1, 1, 1], [1, -1, 1], [-1, -1, 1], [-1, 1, 1], )) / 2 corners3d = boxes3d[:, None, 3:6].repeat(1, 8, 1) * template[None, :, :] corners3d = rotate_points_along_z(corners3d.view(-1, 8, 3), boxes3d[:, 6]).view(-1, 8, 3) corners3d += boxes3d[:, None, 0:3] return corners3d.numpy() if is_numpy else corners3d def visualize_pts(pts, fig=None, bgcolor=(0, 0, 0), fgcolor=(1.0, 1.0, 1.0), show_intensity=False, size=(800, 600), draw_origin=True): if not isinstance(pts, np.ndarray): pts = pts.cpu().numpy() if fig is None: fig = mlab.figure(figure=None, bgcolor=bgcolor, fgcolor=fgcolor, engine=None, size=size) if show_intensity: G = mlab.points3d(pts[:, 0], pts[:, 1], pts[:, 2], pts[:, 3], mode='point', colormap='gnuplot', scale_factor=1, figure=fig) else: G = mlab.points3d(pts[:, 0], pts[:, 1], pts[:, 2], mode='point', colormap='gnuplot', scale_factor=1, figure=fig) if draw_origin: mlab.points3d(0, 0, 0, color=(1, 1, 1), mode='cube', scale_factor=0.2) mlab.plot3d([0, 3], [0, 0], [0, 0], color=(0, 0, 1), tube_radius=0.1) mlab.plot3d([0, 0], [0, 3], [0, 0], color=(0, 1, 0), tube_radius=0.1) mlab.plot3d([0, 0], [0, 0], [0, 3], color=(1, 0, 0), tube_radius=0.1) return fig def draw_sphere_pts(pts, color=(0, 1, 0), fig=None, bgcolor=(0, 0, 0), scale_factor=0.2): if not isinstance(pts, np.ndarray): pts = pts.cpu().numpy() if fig is None: fig = mlab.figure(figure=None, bgcolor=bgcolor, fgcolor=None, engine=None, size=(600, 600)) if isinstance(color, np.ndarray) and color.shape[0] == 1: color = color[0] color = (color[0] / 255.0, color[1] / 255.0, color[2] / 255.0) if isinstance(color, np.ndarray): pts_color = np.zeros((pts.__len__(), 4), dtype=np.uint8) pts_color[:, 0:3] = color pts_color[:, 3] = 255 G = mlab.points3d(pts[:, 0], pts[:, 1], pts[:, 2], np.arange(0, pts_color.__len__()), mode='sphere', scale_factor=scale_factor, figure=fig) G.glyph.color_mode = 'color_by_scalar' G.glyph.scale_mode = 'scale_by_vector' G.module_manager.scalar_lut_manager.lut.table = pts_color else: mlab.points3d(pts[:, 0], pts[:, 1], pts[:, 2], mode='sphere', color=color, colormap='gnuplot', scale_factor=scale_factor, figure=fig) mlab.points3d(0, 0, 0, color=(1, 1, 1), mode='cube', scale_factor=0.2) mlab.plot3d([0, 3], [0, 0], [0, 0], color=(0, 0, 1), line_width=3, tube_radius=None, figure=fig) mlab.plot3d([0, 0], [0, 3], [0, 0], color=(0, 1, 0), line_width=3, tube_radius=None, figure=fig) mlab.plot3d([0, 0], [0, 0], [0, 3], color=(1, 0, 0), line_width=3, tube_radius=None, figure=fig) return fig def draw_grid(x1, y1, x2, y2, fig, tube_radius=None, color=(0.5, 0.5, 0.5)): mlab.plot3d([x1, x1], [y1, y2], [0, 0], color=color, tube_radius=tube_radius, line_width=1, figure=fig) mlab.plot3d([x2, x2], [y1, y2], [0, 0], color=color, tube_radius=tube_radius, line_width=1, figure=fig) mlab.plot3d([x1, x2], [y1, y1], [0, 0], color=color, tube_radius=tube_radius, line_width=1, figure=fig) mlab.plot3d([x1, x2], [y2, y2], [0, 0], color=color, tube_radius=tube_radius, line_width=1, figure=fig) return fig def draw_multi_grid_range(fig, grid_size=20, bv_range=(-60, -60, 60, 60)): for x in range(bv_range[0], bv_range[2], grid_size): for y in range(bv_range[1], bv_range[3], grid_size): fig = draw_grid(x, y, x + grid_size, y + grid_size, fig) return fig def draw_scenes(points, gt_boxes=None, ref_boxes=None, ref_scores=None, ref_labels=None): if not isinstance(points, np.ndarray): points = points.cpu().numpy() if ref_boxes is not None and not isinstance(ref_boxes, np.ndarray): ref_boxes = ref_boxes.cpu().numpy() if gt_boxes is not None and not isinstance(gt_boxes, np.ndarray): gt_boxes = gt_boxes.cpu().numpy() if ref_scores is not None and not isinstance(ref_scores, np.ndarray): ref_scores = ref_scores.cpu().numpy() if ref_labels is not None and not isinstance(ref_labels, np.ndarray): ref_labels = ref_labels.cpu().numpy() fig = visualize_pts(points) fig = draw_multi_grid_range(fig, bv_range=(0, -40, 80, 40)) if gt_boxes is not None: corners3d = boxes_to_corners_3d(gt_boxes) fig = draw_corners3d(corners3d, fig=fig, color=(0, 0, 1), max_num=100) if ref_boxes is not None and len(ref_boxes) > 0: ref_corners3d = boxes_to_corners_3d(ref_boxes) if ref_labels is None: fig = draw_corners3d(ref_corners3d, fig=fig, color=(0, 1, 0), cls=ref_scores, max_num=100) else: for k in range(ref_labels.min(), ref_labels.max() + 1): cur_color = tuple(box_colormap[k % len(box_colormap)]) mask = (ref_labels == k) fig = draw_corners3d(ref_corners3d[mask], fig=fig, color=cur_color, cls=ref_scores[mask], max_num=100) mlab.view(azimuth=-179, elevation=54.0, distance=104.0, roll=90.0) return fig def draw_corners3d(corners3d, fig, color=(1, 1, 1), line_width=2, cls=None, tag='', max_num=500, tube_radius=None): """ :param corners3d: (N, 8, 3) :param fig: :param color: :param line_width: :param cls: :param tag: :param max_num: :return: """ import mayavi.mlab as mlab num = min(max_num, len(corners3d)) for n in range(num): b = corners3d[n] # (8, 3) if cls is not None: if isinstance(cls, np.ndarray): mlab.text3d(b[6, 0], b[6, 1], b[6, 2], '%.2f' % cls[n], scale=(0.3, 0.3, 0.3), color=color, figure=fig) else: mlab.text3d(b[6, 0], b[6, 1], b[6, 2], '%s' % cls[n], scale=(0.3, 0.3, 0.3), color=color, figure=fig) for k in range(0, 4): i, j = k, (k + 1) % 4 mlab.plot3d([b[i, 0], b[j, 0]], [b[i, 1], b[j, 1]], [b[i, 2], b[j, 2]], color=color, tube_radius=tube_radius, line_width=line_width, figure=fig) i, j = k + 4, (k + 1) % 4 + 4 mlab.plot3d([b[i, 0], b[j, 0]], [b[i, 1], b[j, 1]], [b[i, 2], b[j, 2]], color=color, tube_radius=tube_radius, line_width=line_width, figure=fig) i, j = k, k + 4 mlab.plot3d([b[i, 0], b[j, 0]], [b[i, 1], b[j, 1]], [b[i, 2], b[j, 2]], color=color, tube_radius=tube_radius, line_width=line_width, figure=fig) i, j = 0, 5 mlab.plot3d([b[i, 0], b[j, 0]], [b[i, 1], b[j, 1]], [b[i, 2], b[j, 2]], color=color, tube_radius=tube_radius, line_width=line_width, figure=fig) i, j = 1, 4 mlab.plot3d([b[i, 0], b[j, 0]], [b[i, 1], b[j, 1]], [b[i, 2], b[j, 2]], color=color, tube_radius=tube_radius, line_width=line_width, figure=fig) return fig
8,540
38.541667
121
py
SASA
SASA-main/pcdet/config.py
from pathlib import Path import yaml from easydict import EasyDict def log_config_to_file(cfg, pre='cfg', logger=None): for key, val in cfg.items(): if isinstance(cfg[key], EasyDict): logger.info('\n%s.%s = edict()' % (pre, key)) log_config_to_file(cfg[key], pre=pre + '.' + key, logger=logger) continue logger.info('%s.%s: %s' % (pre, key, val)) def cfg_from_list(cfg_list, config): """Set config keys via list (e.g., from command line).""" from ast import literal_eval assert len(cfg_list) % 2 == 0 for k, v in zip(cfg_list[0::2], cfg_list[1::2]): key_list = k.split('.') d = config for subkey in key_list[:-1]: assert subkey in d, 'NotFoundKey: %s' % subkey d = d[subkey] subkey = key_list[-1] assert subkey in d, 'NotFoundKey: %s' % subkey try: value = literal_eval(v) except: value = v if type(value) != type(d[subkey]) and isinstance(d[subkey], EasyDict): key_val_list = value.split(',') for src in key_val_list: cur_key, cur_val = src.split(':') val_type = type(d[subkey][cur_key]) cur_val = val_type(cur_val) d[subkey][cur_key] = cur_val elif type(value) != type(d[subkey]) and isinstance(d[subkey], list): val_list = value.split(',') for k, x in enumerate(val_list): val_list[k] = type(d[subkey][0])(x) d[subkey] = val_list else: assert type(value) == type(d[subkey]), \ 'type {} does not match original type {}'.format(type(value), type(d[subkey])) d[subkey] = value def merge_new_config(config, new_config): if '_BASE_CONFIG_' in new_config: with open(new_config['_BASE_CONFIG_'], 'r') as f: try: yaml_config = yaml.load(f, Loader=yaml.FullLoader) except: yaml_config = yaml.load(f) config.update(EasyDict(yaml_config)) for key, val in new_config.items(): if not isinstance(val, dict): config[key] = val continue if key not in config: config[key] = EasyDict() merge_new_config(config[key], val) return config def cfg_from_yaml_file(cfg_file, config): with open(cfg_file, 'r') as f: try: new_config = yaml.load(f, Loader=yaml.FullLoader) except: new_config = yaml.load(f) merge_new_config(config=config, new_config=new_config) return config cfg = EasyDict() cfg.ROOT_DIR = (Path(__file__).resolve().parent / '../').resolve() cfg.LOCAL_RANK = 0
2,750
30.988372
94
py
SASA
SASA-main/pcdet/__init__.py
import subprocess from pathlib import Path from .version import __version__ __all__ = [ '__version__' ] def get_git_commit_number(): if not (Path(__file__).parent / '../.git').exists(): return '0000000' cmd_out = subprocess.run(['git', 'rev-parse', 'HEAD'], stdout=subprocess.PIPE) git_commit_number = cmd_out.stdout.decode('utf-8')[:7] return git_commit_number script_version = get_git_commit_number() if script_version not in __version__: __version__ = __version__ + '+py%s' % script_version
535
20.44
82
py
SASA
SASA-main/pcdet/models/__init__.py
from collections import namedtuple import numpy as np import torch from .detectors import build_detector def build_network(model_cfg, num_class, dataset): model = build_detector( model_cfg=model_cfg, num_class=num_class, dataset=dataset ) return model def load_data_to_gpu(batch_dict): for key, val in batch_dict.items(): if not isinstance(val, np.ndarray): continue if key in ['frame_id', 'metadata', 'calib', 'image_shape']: continue batch_dict[key] = torch.from_numpy(val).float().cuda() def model_fn_decorator(): ModelReturn = namedtuple('ModelReturn', ['loss', 'tb_dict', 'disp_dict']) def model_func(model, batch_dict): load_data_to_gpu(batch_dict) ret_dict, tb_dict, disp_dict = model(batch_dict) loss = ret_dict['loss'].mean() if hasattr(model, 'update_global_step'): model.update_global_step() else: model.module.update_global_step() return ModelReturn(loss, tb_dict, disp_dict) return model_func
1,074
25.219512
77
py
SASA
SASA-main/pcdet/models/detectors/point_rcnn.py
from .detector3d_template import Detector3DTemplate class PointRCNN(Detector3DTemplate): def __init__(self, model_cfg, num_class, dataset): super().__init__(model_cfg=model_cfg, num_class=num_class, dataset=dataset) self.module_list = self.build_networks() def forward(self, batch_dict): for cur_module in self.module_list: batch_dict = cur_module(batch_dict) if self.training: loss, tb_dict, disp_dict = self.get_training_loss() ret_dict = { 'loss': loss } return ret_dict, tb_dict, disp_dict else: pred_dicts, recall_dicts = self.post_processing(batch_dict) return pred_dicts, recall_dicts def get_training_loss(self): disp_dict = {} loss_point, tb_dict = self.point_head.get_loss() loss_rcnn, tb_dict = self.roi_head.get_loss(tb_dict) loss = loss_point + loss_rcnn return loss, tb_dict, disp_dict
999
31.258065
83
py
SASA
SASA-main/pcdet/models/detectors/pointpillar.py
from .detector3d_template import Detector3DTemplate class PointPillar(Detector3DTemplate): def __init__(self, model_cfg, num_class, dataset): super().__init__(model_cfg=model_cfg, num_class=num_class, dataset=dataset) self.module_list = self.build_networks() def forward(self, batch_dict): for cur_module in self.module_list: batch_dict = cur_module(batch_dict) if self.training: loss, tb_dict, disp_dict = self.get_training_loss() ret_dict = { 'loss': loss } return ret_dict, tb_dict, disp_dict else: pred_dicts, recall_dicts = self.post_processing(batch_dict) return pred_dicts, recall_dicts def get_training_loss(self): disp_dict = {} loss_rpn, tb_dict = self.dense_head.get_loss() tb_dict = { 'loss_rpn': loss_rpn.item(), **tb_dict } loss = loss_rpn return loss, tb_dict, disp_dict
1,018
28.114286
83
py
SASA
SASA-main/pcdet/models/detectors/second_net.py
from .detector3d_template import Detector3DTemplate class SECONDNet(Detector3DTemplate): def __init__(self, model_cfg, num_class, dataset): super().__init__(model_cfg=model_cfg, num_class=num_class, dataset=dataset) self.module_list = self.build_networks() def forward(self, batch_dict): for cur_module in self.module_list: batch_dict = cur_module(batch_dict) if self.training: loss, tb_dict, disp_dict = self.get_training_loss() ret_dict = { 'loss': loss } return ret_dict, tb_dict, disp_dict else: pred_dicts, recall_dicts = self.post_processing(batch_dict) return pred_dicts, recall_dicts def get_training_loss(self): disp_dict = {} loss_rpn, tb_dict = self.dense_head.get_loss() tb_dict = { 'loss_rpn': loss_rpn.item(), **tb_dict } loss = loss_rpn return loss, tb_dict, disp_dict
1,016
28.057143
83
py
SASA
SASA-main/pcdet/models/detectors/detector3d_template.py
import os import torch import torch.nn as nn from ...ops.iou3d_nms import iou3d_nms_utils from .. import backbones_2d, backbones_3d, dense_heads, roi_heads from ..backbones_2d import map_to_bev from ..backbones_3d import pfe, vfe from ..model_utils import model_nms_utils class Detector3DTemplate(nn.Module): def __init__(self, model_cfg, num_class, dataset): super().__init__() self.model_cfg = model_cfg self.num_class = num_class self.dataset = dataset self.class_names = dataset.class_names self.register_buffer('global_step', torch.LongTensor(1).zero_()) self.module_topology = [ 'vfe', 'backbone_3d', 'map_to_bev_module', 'pfe', 'backbone_2d', 'dense_head', 'point_head', 'roi_head' ] @property def mode(self): return 'TRAIN' if self.training else 'TEST' def update_global_step(self): self.global_step += 1 def build_networks(self): model_info_dict = { 'module_list': [], 'num_rawpoint_features': self.dataset.point_feature_encoder.num_point_features, 'num_point_features': self.dataset.point_feature_encoder.num_point_features, 'grid_size': self.dataset.grid_size, 'point_cloud_range': self.dataset.point_cloud_range, 'voxel_size': self.dataset.voxel_size } for module_name in self.module_topology: module, model_info_dict = getattr(self, 'build_%s' % module_name)( model_info_dict=model_info_dict ) self.add_module(module_name, module) return model_info_dict['module_list'] def build_vfe(self, model_info_dict): if self.model_cfg.get('VFE', None) is None: return None, model_info_dict vfe_module = vfe.__all__[self.model_cfg.VFE.NAME]( model_cfg=self.model_cfg.VFE, num_point_features=model_info_dict['num_rawpoint_features'], point_cloud_range=model_info_dict['point_cloud_range'], voxel_size=model_info_dict['voxel_size'] ) model_info_dict['num_point_features'] = vfe_module.get_output_feature_dim() model_info_dict['module_list'].append(vfe_module) return vfe_module, model_info_dict def build_backbone_3d(self, model_info_dict): if self.model_cfg.get('BACKBONE_3D', None) is None: return None, model_info_dict backbone_3d_module = backbones_3d.__all__[self.model_cfg.BACKBONE_3D.NAME]( model_cfg=self.model_cfg.BACKBONE_3D, input_channels=model_info_dict['num_point_features'], grid_size=model_info_dict['grid_size'], voxel_size=model_info_dict['voxel_size'], point_cloud_range=model_info_dict['point_cloud_range'] ) model_info_dict['module_list'].append(backbone_3d_module) model_info_dict['num_point_features'] = backbone_3d_module.num_point_features return backbone_3d_module, model_info_dict def build_map_to_bev_module(self, model_info_dict): if self.model_cfg.get('MAP_TO_BEV', None) is None: return None, model_info_dict map_to_bev_module = map_to_bev.__all__[self.model_cfg.MAP_TO_BEV.NAME]( model_cfg=self.model_cfg.MAP_TO_BEV, grid_size=model_info_dict['grid_size'] ) model_info_dict['module_list'].append(map_to_bev_module) model_info_dict['num_bev_features'] = map_to_bev_module.num_bev_features return map_to_bev_module, model_info_dict def build_backbone_2d(self, model_info_dict): if self.model_cfg.get('BACKBONE_2D', None) is None: return None, model_info_dict backbone_2d_module = backbones_2d.__all__[self.model_cfg.BACKBONE_2D.NAME]( model_cfg=self.model_cfg.BACKBONE_2D, input_channels=model_info_dict['num_bev_features'] ) model_info_dict['module_list'].append(backbone_2d_module) model_info_dict['num_bev_features'] = backbone_2d_module.num_bev_features return backbone_2d_module, model_info_dict def build_pfe(self, model_info_dict): if self.model_cfg.get('PFE', None) is None: return None, model_info_dict pfe_module = pfe.__all__[self.model_cfg.PFE.NAME]( model_cfg=self.model_cfg.PFE, voxel_size=model_info_dict['voxel_size'], point_cloud_range=model_info_dict['point_cloud_range'], num_bev_features=model_info_dict['num_bev_features'], num_rawpoint_features=model_info_dict['num_rawpoint_features'] ) model_info_dict['module_list'].append(pfe_module) model_info_dict['num_point_features'] = pfe_module.num_point_features model_info_dict['num_point_features_before_fusion'] = pfe_module.num_point_features_before_fusion return pfe_module, model_info_dict def build_dense_head(self, model_info_dict): if self.model_cfg.get('DENSE_HEAD', None) is None: return None, model_info_dict dense_head_module = dense_heads.__all__[self.model_cfg.DENSE_HEAD.NAME]( model_cfg=self.model_cfg.DENSE_HEAD, input_channels=model_info_dict['num_bev_features'], num_class=self.num_class if not self.model_cfg.DENSE_HEAD.CLASS_AGNOSTIC else 1, class_names=self.class_names, grid_size=model_info_dict['grid_size'], point_cloud_range=model_info_dict['point_cloud_range'], predict_boxes_when_training=self.model_cfg.get('ROI_HEAD', False) ) model_info_dict['module_list'].append(dense_head_module) return dense_head_module, model_info_dict def build_point_head(self, model_info_dict): if self.model_cfg.get('POINT_HEAD', None) is None: return None, model_info_dict if self.model_cfg.POINT_HEAD.get('USE_POINT_FEATURES_BEFORE_FUSION', False): num_point_features = model_info_dict['num_point_features_before_fusion'] else: num_point_features = model_info_dict['num_point_features'] point_head_module = dense_heads.__all__[self.model_cfg.POINT_HEAD.NAME]( model_cfg=self.model_cfg.POINT_HEAD, input_channels=num_point_features, num_class=self.num_class if not self.model_cfg.POINT_HEAD.CLASS_AGNOSTIC else 1, predict_boxes_when_training=self.model_cfg.get('ROI_HEAD', False) ) model_info_dict['module_list'].append(point_head_module) return point_head_module, model_info_dict def build_roi_head(self, model_info_dict): if self.model_cfg.get('ROI_HEAD', None) is None: return None, model_info_dict point_head_module = roi_heads.__all__[self.model_cfg.ROI_HEAD.NAME]( model_cfg=self.model_cfg.ROI_HEAD, input_channels=model_info_dict['num_point_features'], num_class=self.num_class if not self.model_cfg.ROI_HEAD.CLASS_AGNOSTIC else 1, ) model_info_dict['module_list'].append(point_head_module) return point_head_module, model_info_dict def forward(self, **kwargs): raise NotImplementedError def post_processing(self, batch_dict): """ Args: batch_dict: batch_size: batch_cls_preds: (B, num_boxes, num_classes | 1) or (N1+N2+..., num_classes | 1) or [(B, num_boxes, num_class1), (B, num_boxes, num_class2) ...] multihead_label_mapping: [(num_class1), (num_class2), ...] batch_box_preds: (B, num_boxes, 7+C) or (N1+N2+..., 7+C) cls_preds_normalized: indicate whether batch_cls_preds is normalized batch_index: optional (N1+N2+...) has_class_labels: True/False roi_labels: (B, num_rois) 1 .. num_classes batch_pred_labels: (B, num_boxes, 1) Returns: """ post_process_cfg = self.model_cfg.POST_PROCESSING batch_size = batch_dict['batch_size'] recall_dict = {} pred_dicts = [] for index in range(batch_size): if batch_dict.get('batch_index', None) is not None: assert batch_dict['batch_box_preds'].shape.__len__() == 2 batch_mask = (batch_dict['batch_index'] == index) else: assert batch_dict['batch_box_preds'].shape.__len__() == 3 batch_mask = index box_preds = batch_dict['batch_box_preds'][batch_mask] src_box_preds = box_preds if not isinstance(batch_dict['batch_cls_preds'], list): cls_preds = batch_dict['batch_cls_preds'][batch_mask] src_cls_preds = cls_preds assert cls_preds.shape[1] in [1, self.num_class] if not batch_dict['cls_preds_normalized']: cls_preds = torch.sigmoid(cls_preds) else: cls_preds = [x[batch_mask] for x in batch_dict['batch_cls_preds']] src_cls_preds = cls_preds if not batch_dict['cls_preds_normalized']: cls_preds = [torch.sigmoid(x) for x in cls_preds] if post_process_cfg.NMS_CONFIG.MULTI_CLASSES_NMS: if not isinstance(cls_preds, list): cls_preds = [cls_preds] multihead_label_mapping = [torch.arange(1, self.num_class, device=cls_preds[0].device)] else: multihead_label_mapping = batch_dict['multihead_label_mapping'] cur_start_idx = 0 pred_scores, pred_labels, pred_boxes = [], [], [] for cur_cls_preds, cur_label_mapping in zip(cls_preds, multihead_label_mapping): assert cur_cls_preds.shape[1] == len(cur_label_mapping) cur_box_preds = box_preds[cur_start_idx: cur_start_idx + cur_cls_preds.shape[0]] cur_pred_scores, cur_pred_labels, cur_pred_boxes = model_nms_utils.multi_classes_nms( cls_scores=cur_cls_preds, box_preds=cur_box_preds, nms_config=post_process_cfg.NMS_CONFIG, score_thresh=post_process_cfg.SCORE_THRESH ) cur_pred_labels = cur_label_mapping[cur_pred_labels] pred_scores.append(cur_pred_scores) pred_labels.append(cur_pred_labels) pred_boxes.append(cur_pred_boxes) cur_start_idx += cur_cls_preds.shape[0] final_scores = torch.cat(pred_scores, dim=0) final_labels = torch.cat(pred_labels, dim=0) final_boxes = torch.cat(pred_boxes, dim=0) else: cls_preds, label_preds = torch.max(cls_preds, dim=-1) if batch_dict.get('has_class_labels', False): label_key = 'roi_labels' if 'roi_labels' in batch_dict else 'batch_pred_labels' label_preds = batch_dict[label_key][index] else: label_preds = label_preds + 1 selected, selected_scores = model_nms_utils.class_agnostic_nms( box_scores=cls_preds, box_preds=box_preds, nms_config=post_process_cfg.NMS_CONFIG, score_thresh=post_process_cfg.SCORE_THRESH ) if post_process_cfg.OUTPUT_RAW_SCORE: max_cls_preds, _ = torch.max(src_cls_preds, dim=-1) selected_scores = max_cls_preds[selected] final_scores = selected_scores final_labels = label_preds[selected] final_boxes = box_preds[selected] recall_dict = self.generate_recall_record( box_preds=final_boxes if 'rois' not in batch_dict else src_box_preds, recall_dict=recall_dict, batch_index=index, data_dict=batch_dict, thresh_list=post_process_cfg.RECALL_THRESH_LIST ) record_dict = { 'pred_boxes': final_boxes, 'pred_scores': final_scores, 'pred_labels': final_labels } pred_dicts.append(record_dict) return pred_dicts, recall_dict def generate_recall_record(self, box_preds, recall_dict, batch_index, data_dict=None, thresh_list=None): if 'gt_boxes' not in data_dict: return recall_dict rois = data_dict['rois'][batch_index] if 'rois' in data_dict else None gt_boxes = data_dict['gt_boxes'][batch_index] if recall_dict.__len__() == 0: recall_dict = {'gt_num': 0} for cur_thresh in thresh_list: recall_dict['roi_%s' % (str(cur_thresh))] = 0 recall_dict['rcnn_%s' % (str(cur_thresh))] = 0 cur_gt = gt_boxes k = cur_gt.__len__() - 1 while k > 0 and cur_gt[k].sum() == 0: k -= 1 cur_gt = cur_gt[:k + 1] if cur_gt.shape[0] > 0: if box_preds.shape[0] > 0: iou3d_rcnn = iou3d_nms_utils.boxes_iou3d_gpu(box_preds[:, 0:7], cur_gt[:, 0:7]) else: iou3d_rcnn = torch.zeros((0, cur_gt.shape[0])) if rois is not None: iou3d_roi = iou3d_nms_utils.boxes_iou3d_gpu(rois[:, 0:7], cur_gt[:, 0:7]) for cur_thresh in thresh_list: if iou3d_rcnn.shape[0] == 0: recall_dict['rcnn_%s' % str(cur_thresh)] += 0 else: rcnn_recalled = (iou3d_rcnn.max(dim=0)[0] > cur_thresh).sum().item() recall_dict['rcnn_%s' % str(cur_thresh)] += rcnn_recalled if rois is not None: roi_recalled = (iou3d_roi.max(dim=0)[0] > cur_thresh).sum().item() recall_dict['roi_%s' % str(cur_thresh)] += roi_recalled recall_dict['gt_num'] += cur_gt.shape[0] else: gt_iou = box_preds.new_zeros(box_preds.shape[0]) return recall_dict def load_params_from_file(self, filename, logger, to_cpu=False): if not os.path.isfile(filename): raise FileNotFoundError logger.info('==> Loading parameters from checkpoint %s to %s' % (filename, 'CPU' if to_cpu else 'GPU')) loc_type = torch.device('cpu') if to_cpu else None checkpoint = torch.load(filename, map_location=loc_type) model_state_disk = checkpoint['model_state'] if 'version' in checkpoint: logger.info('==> Checkpoint trained from version: %s' % checkpoint['version']) update_model_state = {} for key, val in model_state_disk.items(): if key in self.state_dict() and self.state_dict()[key].shape == model_state_disk[key].shape: update_model_state[key] = val # logger.info('Update weight %s: %s' % (key, str(val.shape))) state_dict = self.state_dict() state_dict.update(update_model_state) self.load_state_dict(state_dict) for key in state_dict: if key not in update_model_state: logger.info('Not updated weight %s: %s' % (key, str(state_dict[key].shape))) logger.info('==> Done (loaded %d/%d)' % (len(update_model_state), len(self.state_dict()))) def load_params_with_optimizer(self, filename, to_cpu=False, optimizer=None, logger=None): if not os.path.isfile(filename): raise FileNotFoundError logger.info('==> Loading parameters from checkpoint %s to %s' % (filename, 'CPU' if to_cpu else 'GPU')) loc_type = torch.device('cpu') if to_cpu else None checkpoint = torch.load(filename, map_location=loc_type) epoch = checkpoint.get('epoch', -1) it = checkpoint.get('it', 0.0) self.load_state_dict(checkpoint['model_state']) if optimizer is not None: if 'optimizer_state' in checkpoint and checkpoint['optimizer_state'] is not None: logger.info('==> Loading optimizer parameters from checkpoint %s to %s' % (filename, 'CPU' if to_cpu else 'GPU')) optimizer.load_state_dict(checkpoint['optimizer_state']) else: assert filename[-4] == '.', filename src_file, ext = filename[:-4], filename[-3:] optimizer_filename = '%s_optim.%s' % (src_file, ext) if os.path.exists(optimizer_filename): optimizer_ckpt = torch.load(optimizer_filename, map_location=loc_type) optimizer.load_state_dict(optimizer_ckpt['optimizer_state']) if 'version' in checkpoint: print('==> Checkpoint trained from version: %s' % checkpoint['version']) logger.info('==> Done') return it, epoch
17,009
44.119363
111
py
SASA
SASA-main/pcdet/models/detectors/PartA2_net.py
from .detector3d_template import Detector3DTemplate class PartA2Net(Detector3DTemplate): def __init__(self, model_cfg, num_class, dataset): super().__init__(model_cfg=model_cfg, num_class=num_class, dataset=dataset) self.module_list = self.build_networks() def forward(self, batch_dict): for cur_module in self.module_list: batch_dict = cur_module(batch_dict) if self.training: loss, tb_dict, disp_dict = self.get_training_loss() ret_dict = { 'loss': loss } return ret_dict, tb_dict, disp_dict else: pred_dicts, recall_dicts = self.post_processing(batch_dict) return pred_dicts, recall_dicts def get_training_loss(self): disp_dict = {} loss_rpn, tb_dict = self.dense_head.get_loss() loss_point, tb_dict = self.point_head.get_loss(tb_dict) loss_rcnn, tb_dict = self.roi_head.get_loss(tb_dict) loss = loss_rpn + loss_point + loss_rcnn return loss, tb_dict, disp_dict
1,072
32.53125
83
py
SASA
SASA-main/pcdet/models/detectors/pv_rcnn.py
from .detector3d_template import Detector3DTemplate class PVRCNN(Detector3DTemplate): def __init__(self, model_cfg, num_class, dataset): super().__init__(model_cfg=model_cfg, num_class=num_class, dataset=dataset) self.module_list = self.build_networks() def forward(self, batch_dict): for cur_module in self.module_list: batch_dict = cur_module(batch_dict) if self.training: loss, tb_dict, disp_dict = self.get_training_loss() ret_dict = { 'loss': loss } return ret_dict, tb_dict, disp_dict else: pred_dicts, recall_dicts = self.post_processing(batch_dict) return pred_dicts, recall_dicts def get_training_loss(self): disp_dict = {} loss_rpn, tb_dict = self.dense_head.get_loss() loss_point, tb_dict = self.point_head.get_loss(tb_dict) loss_rcnn, tb_dict = self.roi_head.get_loss(tb_dict) loss = loss_rpn + loss_point + loss_rcnn return loss, tb_dict, disp_dict
1,069
32.4375
83
py
SASA
SASA-main/pcdet/models/detectors/__init__.py
from .detector3d_template import Detector3DTemplate from .PartA2_net import PartA2Net from .point_rcnn import PointRCNN from .pointpillar import PointPillar from .pv_rcnn import PVRCNN from .second_net import SECONDNet from .point_3dssd import Point3DSSD __all__ = { 'Detector3DTemplate': Detector3DTemplate, 'SECONDNet': SECONDNet, 'PartA2Net': PartA2Net, 'PVRCNN': PVRCNN, 'PointPillar': PointPillar, 'PointRCNN': PointRCNN, '3DSSD': Point3DSSD } def build_detector(model_cfg, num_class, dataset): model = __all__[model_cfg.NAME]( model_cfg=model_cfg, num_class=num_class, dataset=dataset ) return model
658
24.346154
65
py
SASA
SASA-main/pcdet/models/detectors/point_3dssd.py
import torch from .detector3d_template import Detector3DTemplate from ...ops.iou3d_nms import iou3d_nms_utils from ...ops.roiaware_pool3d import roiaware_pool3d_utils class Point3DSSD(Detector3DTemplate): def __init__(self, model_cfg, num_class, dataset): super().__init__(model_cfg=model_cfg, num_class=num_class, dataset=dataset) self.module_list = self.build_networks() def forward(self, batch_dict): for cur_module in self.module_list: batch_dict = cur_module(batch_dict) if self.training: loss, tb_dict, disp_dict = self.get_training_loss() ret_dict = { 'loss': loss } return ret_dict, tb_dict, disp_dict else: pred_dicts, recall_dicts = self.post_processing(batch_dict) return pred_dicts, recall_dicts def get_training_loss(self): disp_dict = {} loss_point, tb_dict = self.point_head.get_loss() loss = loss_point return loss, tb_dict, disp_dict def init_recall_record(self, metric, **kwargs): # initialize gt_num for all classes for cur_cls in range(len(self.class_names)): metric['gt_num[%s]' % self.class_names[cur_cls]] = 0 # initialize statistics of all sampling segments npoint_list = self.model_cfg.BACKBONE_3D.SA_CONFIG.NPOINT_LIST for cur_layer in range(len(npoint_list)): for cur_seg in range(len(npoint_list[cur_layer])): metric['positive_point_L%dS%d' % (cur_layer, cur_seg)] = 0 metric['recall_point_L%dS%d' % (cur_layer, cur_seg)] = 0 for cur_cls in range(self.num_class): metric['recall_point_L%dS%d[%s]' \ % (cur_layer, cur_seg, self.class_names[cur_cls])] = 0 # initialize statistics of the vote layer metric['positive_point_candidate'] = 0 metric['recall_point_candidate'] = 0 metric['positive_point_vote'] = 0 metric['recall_point_vote'] = 0 for cur_cls in range(len(self.class_names)): metric['recall_point_candidate[%s]' % self.class_names[cur_cls]] = 0 metric['recall_point_vote[%s]' % self.class_names[cur_cls]] = 0 def generate_recall_record(self, box_preds, recall_dict, batch_index, data_dict=None, thresh_list=None): if 'gt_boxes' not in data_dict: return recall_dict # point_coords format: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] point_list = data_dict['point_coords_list'] # ignore raw point input npoint_list = self.model_cfg.BACKBONE_3D.SA_CONFIG.NPOINT_LIST assert len(point_list) == len(npoint_list) cur_points_list = [] for cur_layer in range(npoint_list.__len__()): cur_points = point_list[cur_layer] bs_idx = cur_points[:, 0] bs_mask = (bs_idx == batch_index) cur_points = cur_points[bs_mask][:, 1:4] cur_points_list.append(cur_points.split(npoint_list[cur_layer], dim=0)) base_points = data_dict['point_candidate_coords'] vote_points = data_dict['point_vote_coords'] bs_idx = base_points[:, 0] bs_mask = (bs_idx == batch_index) base_points = base_points[bs_mask][:, 1:4] vote_points = vote_points[bs_mask][:, 1:4] rois = data_dict['rois'][batch_index] if 'rois' in data_dict else None gt_boxes = data_dict['gt_boxes'][batch_index] # initialize recall_dict if recall_dict.__len__() == 0: recall_dict = {'gt_num': 0} for cur_thresh in thresh_list: recall_dict['recall_roi_%s' % (str(cur_thresh))] = 0 recall_dict['recall_rcnn_%s' % (str(cur_thresh))] = 0 self.init_recall_record(recall_dict) # init customized statistics cur_gt = gt_boxes k = cur_gt.__len__() - 1 while k > 0 and cur_gt[k].sum() == 0: k -= 1 cur_gt = cur_gt[:k + 1] if cur_gt.shape[0] > 0: # backbone for cur_layer in range(len(npoint_list)): for cur_seg in range(len(npoint_list[cur_layer])): box_idxs_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( cur_points_list[cur_layer][cur_seg].unsqueeze(dim=0), cur_gt[None, :, :7].contiguous() ).long().squeeze(dim=0) box_fg_flag = (box_idxs_of_pts >= 0) recall_dict['positive_point_L%dS%d' % (cur_layer, cur_seg)] += box_fg_flag.long().sum().item() box_recalled = box_idxs_of_pts[box_fg_flag].unique() recall_dict['recall_point_L%dS%d' % (cur_layer, cur_seg)] += box_recalled.size(0) box_recalled_cls = cur_gt[box_recalled, -1] for cur_cls in range(self.num_class): recall_dict['recall_point_L%dS%d[%s]' % (cur_layer, cur_seg, self.class_names[cur_cls])] += \ (box_recalled_cls == (cur_cls + 1)).sum().item() # candidate points box_idxs_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( base_points.unsqueeze(dim=0), cur_gt[None, :, :7] ).long().squeeze(dim=0) box_fg_flag = (box_idxs_of_pts >= 0) recall_dict['positive_point_candidate'] += box_fg_flag.long().sum().item() box_recalled = box_idxs_of_pts[box_fg_flag].unique() recall_dict['recall_point_candidate'] += box_recalled.size(0) box_recalled_cls = cur_gt[box_recalled, -1] for cur_cls in range(self.num_class): recall_dict['recall_point_candidate[%s]' % self.class_names[cur_cls]] += \ (box_recalled_cls == (cur_cls + 1)).sum().item() # vote points box_idxs_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( vote_points.unsqueeze(dim=0), cur_gt[None, :, :7] ).long().squeeze(dim=0) box_fg_flag = (box_idxs_of_pts >= 0) recall_dict['positive_point_vote'] += box_fg_flag.long().sum().item() box_recalled = box_idxs_of_pts[box_fg_flag].unique() recall_dict['recall_point_vote'] += box_recalled.size(0) box_recalled_cls = cur_gt[box_recalled, -1] for cur_cls in range(self.num_class): recall_dict['recall_point_vote[%s]' % self.class_names[cur_cls]] += \ (box_recalled_cls == (cur_cls + 1)).sum().item() if box_preds.shape[0] > 0: iou3d_rcnn = iou3d_nms_utils.boxes_iou3d_gpu(box_preds[:, 0:7], cur_gt[:, 0:7]) else: iou3d_rcnn = torch.zeros((0, cur_gt.shape[0])) if rois is not None: iou3d_roi = iou3d_nms_utils.boxes_iou3d_gpu(rois[:, 0:7], cur_gt[:, 0:7]) for cur_thresh in thresh_list: if iou3d_rcnn.shape[0] == 0: recall_dict['recall_rcnn_%s' % str(cur_thresh)] += 0 else: rcnn_recalled = (iou3d_rcnn.max(dim=0)[0] > cur_thresh).sum().item() recall_dict['recall_rcnn_%s' % str(cur_thresh)] += rcnn_recalled if rois is not None: roi_recalled = (iou3d_roi.max(dim=0)[0] > cur_thresh).sum().item() recall_dict['recall_roi_%s' % str(cur_thresh)] += roi_recalled cur_gt_class = cur_gt[:, -1] for cur_cls in range(self.num_class): cur_cls_gt_num = (cur_gt_class == cur_cls + 1).sum().item() recall_dict['gt_num'] += cur_cls_gt_num recall_dict['gt_num[%s]' % self.class_names[cur_cls]] += cur_cls_gt_num return recall_dict def disp_recall_record(self, metric, logger, sample_num, **kwargs): gt_num = metric['gt_num'] gt_num_cls = [metric['gt_num[%s]' % cur_cls] for cur_cls in self.class_names] # backbone for k in metric.keys(): if 'positive_point_' in k: # count the number of positive points cur_positive_point = metric[k] / sample_num logger.info(k + (': %f' % cur_positive_point)) elif 'recall_point_' in k and not any(cur_cls in k for cur_cls in self.class_names): cur_recall_point = metric[k] / max(gt_num, 1) logger.info(k + (': %f' % cur_recall_point)) for cur_cls in range(len(self.class_names)): cur_recall_point_cls = metric[k + '[%s]' % self.class_names[cur_cls]] / max(gt_num_cls[cur_cls], 1) logger.info('\t- ' + self.class_names[cur_cls] + ': %f' % cur_recall_point_cls) # candidate points positive_point_candidate = metric['positive_point_candidate'] / sample_num logger.info('positive_point_candidate: %f' % positive_point_candidate) recall_point_candidate = metric['recall_point_candidate'] / max(gt_num, 1) logger.info('recall_point_candidate: %f' % recall_point_candidate) for cur_cls in range(len(self.class_names)): cur_recall_point_cls = metric['recall_point_candidate' + '[%s]' % self.class_names[cur_cls]] / max(gt_num_cls[cur_cls], 1) logger.info('\t- ' + self.class_names[cur_cls] + ': %f' % cur_recall_point_cls) # vote points positive_point_vote = metric['positive_point_vote'] / sample_num logger.info('positive_point_vote: %f' % positive_point_vote) recall_point_vote = metric['recall_point_vote'] / max(gt_num, 1) logger.info('recall_point_vote: %f' % recall_point_vote) for cur_cls in range(len(self.class_names)): cur_recall_point_cls = metric['recall_point_vote' + '[%s]' % self.class_names[cur_cls]] / max(gt_num_cls[cur_cls], 1) logger.info('\t- ' + self.class_names[cur_cls] + ': %f' % cur_recall_point_cls)
10,022
47.892683
134
py
SASA
SASA-main/pcdet/models/backbones_3d/spconv_unet.py
from functools import partial import spconv import torch import torch.nn as nn from ...utils import common_utils from .spconv_backbone import post_act_block class SparseBasicBlock(spconv.SparseModule): expansion = 1 def __init__(self, inplanes, planes, stride=1, downsample=None, indice_key=None, norm_fn=None): super(SparseBasicBlock, self).__init__() self.conv1 = spconv.SubMConv3d( inplanes, planes, kernel_size=3, stride=stride, padding=1, bias=False, indice_key=indice_key ) self.bn1 = norm_fn(planes) self.relu = nn.ReLU() self.conv2 = spconv.SubMConv3d( planes, planes, kernel_size=3, stride=1, padding=1, bias=False, indice_key=indice_key ) self.bn2 = norm_fn(planes) self.downsample = downsample self.stride = stride def forward(self, x): identity = x.features assert x.features.dim() == 2, 'x.features.dim()=%d' % x.features.dim() out = self.conv1(x) out.features = self.bn1(out.features) out.features = self.relu(out.features) out = self.conv2(out) out.features = self.bn2(out.features) if self.downsample is not None: identity = self.downsample(x) out.features += identity out.features = self.relu(out.features) return out class UNetV2(nn.Module): """ Sparse Convolution based UNet for point-wise feature learning. Reference Paper: https://arxiv.org/abs/1907.03670 (Shaoshuai Shi, et. al) From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Network """ def __init__(self, model_cfg, input_channels, grid_size, voxel_size, point_cloud_range, **kwargs): super().__init__() self.model_cfg = model_cfg self.sparse_shape = grid_size[::-1] + [1, 0, 0] self.voxel_size = voxel_size self.point_cloud_range = point_cloud_range norm_fn = partial(nn.BatchNorm1d, eps=1e-3, momentum=0.01) self.conv_input = spconv.SparseSequential( spconv.SubMConv3d(input_channels, 16, 3, padding=1, bias=False, indice_key='subm1'), norm_fn(16), nn.ReLU(), ) block = post_act_block self.conv1 = spconv.SparseSequential( block(16, 16, 3, norm_fn=norm_fn, padding=1, indice_key='subm1'), ) self.conv2 = spconv.SparseSequential( # [1600, 1408, 41] <- [800, 704, 21] block(16, 32, 3, norm_fn=norm_fn, stride=2, padding=1, indice_key='spconv2', conv_type='spconv'), block(32, 32, 3, norm_fn=norm_fn, padding=1, indice_key='subm2'), block(32, 32, 3, norm_fn=norm_fn, padding=1, indice_key='subm2'), ) self.conv3 = spconv.SparseSequential( # [800, 704, 21] <- [400, 352, 11] block(32, 64, 3, norm_fn=norm_fn, stride=2, padding=1, indice_key='spconv3', conv_type='spconv'), block(64, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm3'), block(64, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm3'), ) self.conv4 = spconv.SparseSequential( # [400, 352, 11] <- [200, 176, 5] block(64, 64, 3, norm_fn=norm_fn, stride=2, padding=(0, 1, 1), indice_key='spconv4', conv_type='spconv'), block(64, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm4'), block(64, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm4'), ) if self.model_cfg.get('RETURN_ENCODED_TENSOR', True): last_pad = self.model_cfg.get('last_pad', 0) self.conv_out = spconv.SparseSequential( # [200, 150, 5] -> [200, 150, 2] spconv.SparseConv3d(64, 128, (3, 1, 1), stride=(2, 1, 1), padding=last_pad, bias=False, indice_key='spconv_down2'), norm_fn(128), nn.ReLU(), ) else: self.conv_out = None # decoder # [400, 352, 11] <- [200, 176, 5] self.conv_up_t4 = SparseBasicBlock(64, 64, indice_key='subm4', norm_fn=norm_fn) self.conv_up_m4 = block(128, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm4') self.inv_conv4 = block(64, 64, 3, norm_fn=norm_fn, indice_key='spconv4', conv_type='inverseconv') # [800, 704, 21] <- [400, 352, 11] self.conv_up_t3 = SparseBasicBlock(64, 64, indice_key='subm3', norm_fn=norm_fn) self.conv_up_m3 = block(128, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm3') self.inv_conv3 = block(64, 32, 3, norm_fn=norm_fn, indice_key='spconv3', conv_type='inverseconv') # [1600, 1408, 41] <- [800, 704, 21] self.conv_up_t2 = SparseBasicBlock(32, 32, indice_key='subm2', norm_fn=norm_fn) self.conv_up_m2 = block(64, 32, 3, norm_fn=norm_fn, indice_key='subm2') self.inv_conv2 = block(32, 16, 3, norm_fn=norm_fn, indice_key='spconv2', conv_type='inverseconv') # [1600, 1408, 41] <- [1600, 1408, 41] self.conv_up_t1 = SparseBasicBlock(16, 16, indice_key='subm1', norm_fn=norm_fn) self.conv_up_m1 = block(32, 16, 3, norm_fn=norm_fn, indice_key='subm1') self.conv5 = spconv.SparseSequential( block(16, 16, 3, norm_fn=norm_fn, padding=1, indice_key='subm1') ) self.num_point_features = 16 def UR_block_forward(self, x_lateral, x_bottom, conv_t, conv_m, conv_inv): x_trans = conv_t(x_lateral) x = x_trans x.features = torch.cat((x_bottom.features, x_trans.features), dim=1) x_m = conv_m(x) x = self.channel_reduction(x, x_m.features.shape[1]) x.features = x_m.features + x.features x = conv_inv(x) return x @staticmethod def channel_reduction(x, out_channels): """ Args: x: x.features (N, C1) out_channels: C2 Returns: """ features = x.features n, in_channels = features.shape assert (in_channels % out_channels == 0) and (in_channels >= out_channels) x.features = features.view(n, out_channels, -1).sum(dim=2) return x def forward(self, batch_dict): """ Args: batch_dict: batch_size: int vfe_features: (num_voxels, C) voxel_coords: (num_voxels, 4), [batch_idx, z_idx, y_idx, x_idx] Returns: batch_dict: encoded_spconv_tensor: sparse tensor point_features: (N, C) """ voxel_features, voxel_coords = batch_dict['voxel_features'], batch_dict['voxel_coords'] batch_size = batch_dict['batch_size'] input_sp_tensor = spconv.SparseConvTensor( features=voxel_features, indices=voxel_coords.int(), spatial_shape=self.sparse_shape, batch_size=batch_size ) x = self.conv_input(input_sp_tensor) x_conv1 = self.conv1(x) x_conv2 = self.conv2(x_conv1) x_conv3 = self.conv3(x_conv2) x_conv4 = self.conv4(x_conv3) if self.conv_out is not None: # for detection head # [200, 176, 5] -> [200, 176, 2] out = self.conv_out(x_conv4) batch_dict['encoded_spconv_tensor'] = out batch_dict['encoded_spconv_tensor_stride'] = 8 # for segmentation head # [400, 352, 11] <- [200, 176, 5] x_up4 = self.UR_block_forward(x_conv4, x_conv4, self.conv_up_t4, self.conv_up_m4, self.inv_conv4) # [800, 704, 21] <- [400, 352, 11] x_up3 = self.UR_block_forward(x_conv3, x_up4, self.conv_up_t3, self.conv_up_m3, self.inv_conv3) # [1600, 1408, 41] <- [800, 704, 21] x_up2 = self.UR_block_forward(x_conv2, x_up3, self.conv_up_t2, self.conv_up_m2, self.inv_conv2) # [1600, 1408, 41] <- [1600, 1408, 41] x_up1 = self.UR_block_forward(x_conv1, x_up2, self.conv_up_t1, self.conv_up_m1, self.conv5) batch_dict['point_features'] = x_up1.features point_coords = common_utils.get_voxel_centers( x_up1.indices[:, 1:], downsample_times=1, voxel_size=self.voxel_size, point_cloud_range=self.point_cloud_range ) batch_dict['point_coords'] = torch.cat((x_up1.indices[:, 0:1].float(), point_coords), dim=1) return batch_dict
8,445
38.839623
117
py
SASA
SASA-main/pcdet/models/backbones_3d/spconv_backbone.py
from functools import partial import spconv import torch.nn as nn def post_act_block(in_channels, out_channels, kernel_size, indice_key=None, stride=1, padding=0, conv_type='subm', norm_fn=None): if conv_type == 'subm': conv = spconv.SubMConv3d(in_channels, out_channels, kernel_size, bias=False, indice_key=indice_key) elif conv_type == 'spconv': conv = spconv.SparseConv3d(in_channels, out_channels, kernel_size, stride=stride, padding=padding, bias=False, indice_key=indice_key) elif conv_type == 'inverseconv': conv = spconv.SparseInverseConv3d(in_channels, out_channels, kernel_size, indice_key=indice_key, bias=False) else: raise NotImplementedError m = spconv.SparseSequential( conv, norm_fn(out_channels), nn.ReLU(), ) return m class SparseBasicBlock(spconv.SparseModule): expansion = 1 def __init__(self, inplanes, planes, stride=1, norm_fn=None, downsample=None, indice_key=None): super(SparseBasicBlock, self).__init__() assert norm_fn is not None bias = norm_fn is not None self.conv1 = spconv.SubMConv3d( inplanes, planes, kernel_size=3, stride=stride, padding=1, bias=bias, indice_key=indice_key ) self.bn1 = norm_fn(planes) self.relu = nn.ReLU() self.conv2 = spconv.SubMConv3d( planes, planes, kernel_size=3, stride=stride, padding=1, bias=bias, indice_key=indice_key ) self.bn2 = norm_fn(planes) self.downsample = downsample self.stride = stride def forward(self, x): identity = x out = self.conv1(x) out.features = self.bn1(out.features) out.features = self.relu(out.features) out = self.conv2(out) out.features = self.bn2(out.features) if self.downsample is not None: identity = self.downsample(x) out.features += identity.features out.features = self.relu(out.features) return out class VoxelBackBone8x(nn.Module): def __init__(self, model_cfg, input_channels, grid_size, **kwargs): super().__init__() self.model_cfg = model_cfg norm_fn = partial(nn.BatchNorm1d, eps=1e-3, momentum=0.01) self.sparse_shape = grid_size[::-1] + [1, 0, 0] self.conv_input = spconv.SparseSequential( spconv.SubMConv3d(input_channels, 16, 3, padding=1, bias=False, indice_key='subm1'), norm_fn(16), nn.ReLU(), ) block = post_act_block self.conv1 = spconv.SparseSequential( block(16, 16, 3, norm_fn=norm_fn, padding=1, indice_key='subm1'), ) self.conv2 = spconv.SparseSequential( # [1600, 1408, 41] <- [800, 704, 21] block(16, 32, 3, norm_fn=norm_fn, stride=2, padding=1, indice_key='spconv2', conv_type='spconv'), block(32, 32, 3, norm_fn=norm_fn, padding=1, indice_key='subm2'), block(32, 32, 3, norm_fn=norm_fn, padding=1, indice_key='subm2'), ) self.conv3 = spconv.SparseSequential( # [800, 704, 21] <- [400, 352, 11] block(32, 64, 3, norm_fn=norm_fn, stride=2, padding=1, indice_key='spconv3', conv_type='spconv'), block(64, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm3'), block(64, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm3'), ) self.conv4 = spconv.SparseSequential( # [400, 352, 11] <- [200, 176, 5] block(64, 64, 3, norm_fn=norm_fn, stride=2, padding=(0, 1, 1), indice_key='spconv4', conv_type='spconv'), block(64, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm4'), block(64, 64, 3, norm_fn=norm_fn, padding=1, indice_key='subm4'), ) last_pad = 0 last_pad = self.model_cfg.get('last_pad', last_pad) self.conv_out = spconv.SparseSequential( # [200, 150, 5] -> [200, 150, 2] spconv.SparseConv3d(64, 128, (3, 1, 1), stride=(2, 1, 1), padding=last_pad, bias=False, indice_key='spconv_down2'), norm_fn(128), nn.ReLU(), ) self.num_point_features = 128 def forward(self, batch_dict): """ Args: batch_dict: batch_size: int vfe_features: (num_voxels, C) voxel_coords: (num_voxels, 4), [batch_idx, z_idx, y_idx, x_idx] Returns: batch_dict: encoded_spconv_tensor: sparse tensor """ voxel_features, voxel_coords = batch_dict['voxel_features'], batch_dict['voxel_coords'] batch_size = batch_dict['batch_size'] input_sp_tensor = spconv.SparseConvTensor( features=voxel_features, indices=voxel_coords.int(), spatial_shape=self.sparse_shape, batch_size=batch_size ) x = self.conv_input(input_sp_tensor) x_conv1 = self.conv1(x) x_conv2 = self.conv2(x_conv1) x_conv3 = self.conv3(x_conv2) x_conv4 = self.conv4(x_conv3) # for detection head # [200, 176, 5] -> [200, 176, 2] out = self.conv_out(x_conv4) batch_dict.update({ 'encoded_spconv_tensor': out, 'encoded_spconv_tensor_stride': 8 }) batch_dict.update({ 'multi_scale_3d_features': { 'x_conv1': x_conv1, 'x_conv2': x_conv2, 'x_conv3': x_conv3, 'x_conv4': x_conv4, } }) return batch_dict class VoxelResBackBone8x(nn.Module): def __init__(self, model_cfg, input_channels, grid_size, **kwargs): super().__init__() self.model_cfg = model_cfg norm_fn = partial(nn.BatchNorm1d, eps=1e-3, momentum=0.01) self.sparse_shape = grid_size[::-1] + [1, 0, 0] self.conv_input = spconv.SparseSequential( spconv.SubMConv3d(input_channels, 16, 3, padding=1, bias=False, indice_key='subm1'), norm_fn(16), nn.ReLU(), ) block = post_act_block self.conv1 = spconv.SparseSequential( SparseBasicBlock(16, 16, norm_fn=norm_fn, indice_key='res1'), SparseBasicBlock(16, 16, norm_fn=norm_fn, indice_key='res1'), ) self.conv2 = spconv.SparseSequential( # [1600, 1408, 41] <- [800, 704, 21] block(16, 32, 3, norm_fn=norm_fn, stride=2, padding=1, indice_key='spconv2', conv_type='spconv'), SparseBasicBlock(32, 32, norm_fn=norm_fn, indice_key='res2'), SparseBasicBlock(32, 32, norm_fn=norm_fn, indice_key='res2'), ) self.conv3 = spconv.SparseSequential( # [800, 704, 21] <- [400, 352, 11] block(32, 64, 3, norm_fn=norm_fn, stride=2, padding=1, indice_key='spconv3', conv_type='spconv'), SparseBasicBlock(64, 64, norm_fn=norm_fn, indice_key='res3'), SparseBasicBlock(64, 64, norm_fn=norm_fn, indice_key='res3'), ) self.conv4 = spconv.SparseSequential( # [400, 352, 11] <- [200, 176, 5] block(64, 128, 3, norm_fn=norm_fn, stride=2, padding=(0, 1, 1), indice_key='spconv4', conv_type='spconv'), SparseBasicBlock(128, 128, norm_fn=norm_fn, indice_key='res4'), SparseBasicBlock(128, 128, norm_fn=norm_fn, indice_key='res4'), ) last_pad = 0 last_pad = self.model_cfg.get('last_pad', last_pad) self.conv_out = spconv.SparseSequential( # [200, 150, 5] -> [200, 150, 2] spconv.SparseConv3d(128, 128, (3, 1, 1), stride=(2, 1, 1), padding=last_pad, bias=False, indice_key='spconv_down2'), norm_fn(128), nn.ReLU(), ) self.num_point_features = 128 def forward(self, batch_dict): """ Args: batch_dict: batch_size: int vfe_features: (num_voxels, C) voxel_coords: (num_voxels, 4), [batch_idx, z_idx, y_idx, x_idx] Returns: batch_dict: encoded_spconv_tensor: sparse tensor """ voxel_features, voxel_coords = batch_dict['voxel_features'], batch_dict['voxel_coords'] batch_size = batch_dict['batch_size'] input_sp_tensor = spconv.SparseConvTensor( features=voxel_features, indices=voxel_coords.int(), spatial_shape=self.sparse_shape, batch_size=batch_size ) x = self.conv_input(input_sp_tensor) x_conv1 = self.conv1(x) x_conv2 = self.conv2(x_conv1) x_conv3 = self.conv3(x_conv2) x_conv4 = self.conv4(x_conv3) # for detection head # [200, 176, 5] -> [200, 176, 2] out = self.conv_out(x_conv4) batch_dict.update({ 'encoded_spconv_tensor': out, 'encoded_spconv_tensor_stride': 8 }) batch_dict.update({ 'multi_scale_3d_features': { 'x_conv1': x_conv1, 'x_conv2': x_conv2, 'x_conv3': x_conv3, 'x_conv4': x_conv4, } }) return batch_dict
9,376
34.790076
118
py
SASA
SASA-main/pcdet/models/backbones_3d/__init__.py
from .pointnet2_backbone import PointNet2Backbone, PointNet2MSG, PointNet2FSMSG from .spconv_backbone import VoxelBackBone8x, VoxelResBackBone8x from .spconv_unet import UNetV2 __all__ = { 'VoxelBackBone8x': VoxelBackBone8x, 'UNetV2': UNetV2, 'PointNet2Backbone': PointNet2Backbone, 'PointNet2MSG': PointNet2MSG, 'PointNet2FSMSG': PointNet2FSMSG, 'VoxelResBackBone8x': VoxelResBackBone8x, }
416
31.076923
79
py
SASA
SASA-main/pcdet/models/backbones_3d/pointnet2_backbone.py
import torch import torch.nn as nn from ...ops.pointnet2.pointnet2_batch import pointnet2_modules from ...ops.pointnet2.pointnet2_stack import pointnet2_modules as pointnet2_modules_stack from ...ops.pointnet2.pointnet2_stack import pointnet2_utils as pointnet2_utils_stack class PointNet2MSG(nn.Module): def __init__(self, model_cfg, input_channels, **kwargs): super().__init__() self.model_cfg = model_cfg self.SA_modules = nn.ModuleList() channel_in = input_channels - 3 self.num_points_each_layer = [] skip_channel_list = [input_channels - 3] for k in range(self.model_cfg.SA_CONFIG.NPOINTS.__len__()): mlps = self.model_cfg.SA_CONFIG.MLPS[k].copy() channel_out = 0 for idx in range(mlps.__len__()): mlps[idx] = [channel_in] + mlps[idx] channel_out += mlps[idx][-1] self.SA_modules.append( pointnet2_modules.PointnetSAModuleMSG( npoint=self.model_cfg.SA_CONFIG.NPOINTS[k], radii=self.model_cfg.SA_CONFIG.RADIUS[k], nsamples=self.model_cfg.SA_CONFIG.NSAMPLE[k], mlps=mlps, use_xyz=self.model_cfg.SA_CONFIG.get('USE_XYZ', True), ) ) skip_channel_list.append(channel_out) channel_in = channel_out self.FP_modules = nn.ModuleList() skip_sa_block = self.model_cfg.SA_CONFIG.NPOINTS.__len__() - self.model_cfg.FP_MLPS.__len__() for k in range(self.model_cfg.FP_MLPS.__len__()): pre_channel = self.model_cfg.FP_MLPS[k + 1][-1] if k + 1 < len(self.model_cfg.FP_MLPS) else channel_out self.FP_modules.append( pointnet2_modules.PointnetFPModule( mlp=[pre_channel + skip_channel_list[k + skip_sa_block]] + self.model_cfg.FP_MLPS[k] ) ) self.num_point_features = self.model_cfg.FP_MLPS[0][-1] def break_up_pc(self, pc): batch_idx = pc[:, 0] xyz = pc[:, 1:4].contiguous() features = (pc[:, 4:].contiguous() if pc.size(-1) > 4 else None) return batch_idx, xyz, features def forward(self, batch_dict): """ Args: batch_dict: batch_size: int vfe_features: (num_voxels, C) points: (num_points, 4 + C), [batch_idx, x, y, z, ...] Returns: batch_dict: encoded_spconv_tensor: sparse tensor point_features: (N, C) """ batch_size = batch_dict['batch_size'] points = batch_dict['points'] batch_idx, xyz, features = self.break_up_pc(points) xyz_batch_cnt = xyz.new_zeros(batch_size).int() for bs_idx in range(batch_size): xyz_batch_cnt[bs_idx] = (batch_idx == bs_idx).sum() assert xyz_batch_cnt.min() == xyz_batch_cnt.max() xyz = xyz.view(batch_size, -1, 3) features = features.view(batch_size, -1, features.shape[-1]).permute(0, 2, 1) if features is not None else None l_xyz, l_features = [xyz], [features] for i in range(len(self.SA_modules)): li_xyz, li_features = self.SA_modules[i](l_xyz[i], l_features[i]) l_xyz.append(li_xyz) l_features.append(li_features) for i in range(-1, -(len(self.FP_modules) + 1), -1): l_features[i - 1] = self.FP_modules[i]( l_xyz[i - 1], l_xyz[i], l_features[i - 1], l_features[i] ) # (B, C, N) point_features = l_features[0].permute(0, 2, 1).contiguous() # (B, N, C) batch_dict['point_features'] = point_features.view(-1, point_features.shape[-1]) batch_dict['point_coords'] = torch.cat((batch_idx[:, None].float(), l_xyz[0].view(-1, 3)), dim=1) return batch_dict class PointNet2FSMSG(nn.Module): def __init__(self, model_cfg, input_channels, **kwargs): super().__init__() self.model_cfg = model_cfg self.SA_modules = nn.ModuleList() channel_in = input_channels - 3 use_xyz = self.model_cfg.SA_CONFIG.get('USE_XYZ', True) dilated_group = self.model_cfg.SA_CONFIG.get('DILATED_RADIUS_GROUP', False) skip_connection = self.model_cfg.SA_CONFIG.get('SKIP_CONNECTION', False) weight_gamma = self.model_cfg.SA_CONFIG.get('WEIGHT_GAMMA', 1.0) self.aggregation_mlps = self.model_cfg.SA_CONFIG.get('AGGREGATION_MLPS', None) self.confidence_mlps = self.model_cfg.SA_CONFIG.get('CONFIDENCE_MLPS', None) self.num_points_each_layer = [] skip_channel_list = [input_channels - 3] for k in range(self.model_cfg.SA_CONFIG.NPOINT_LIST.__len__()): mlps = self.model_cfg.SA_CONFIG.MLPS[k].copy() channel_out = 0 for idx in range(mlps.__len__()): mlps[idx] = [channel_in] + mlps[idx] channel_out += mlps[idx][-1] if skip_connection: channel_out += channel_in if self.aggregation_mlps and self.aggregation_mlps[k]: aggregation_mlp = self.aggregation_mlps[k].copy() if aggregation_mlp.__len__() == 0: aggregation_mlp = None else: channel_out = aggregation_mlp[-1] else: aggregation_mlp = None if self.confidence_mlps and self.confidence_mlps[k]: confidence_mlp = self.confidence_mlps[k].copy() if confidence_mlp.__len__() == 0: confidence_mlp = None else: confidence_mlp = None self.SA_modules.append( pointnet2_modules.PointnetSAModuleFSMSG( npoint_list=self.model_cfg.SA_CONFIG.NPOINT_LIST[k], sample_range_list=self.model_cfg.SA_CONFIG.SAMPLE_RANGE_LIST[k], sample_method_list=self.model_cfg.SA_CONFIG.SAMPLE_METHOD_LIST[k], radii=self.model_cfg.SA_CONFIG.RADIUS[k], nsamples=self.model_cfg.SA_CONFIG.NSAMPLE[k], mlps=mlps, use_xyz=use_xyz, dilated_radius_group=dilated_group, skip_connection=skip_connection, weight_gamma=weight_gamma, aggregation_mlp=aggregation_mlp, confidence_mlp=confidence_mlp ) ) self.num_points_each_layer.append( sum(self.model_cfg.SA_CONFIG.NPOINT_LIST[k])) skip_channel_list.append(channel_out) channel_in = channel_out self.num_point_features = channel_out fp_mlps = self.model_cfg.get('FP_MLPS', None) if fp_mlps is not None: self.FP_modules = nn.ModuleList() l_skipped = self.model_cfg.SA_CONFIG.NPOINT_LIST.__len__() - self.model_cfg.FP_MLPS.__len__() for k in range(fp_mlps.__len__()): pre_channel = fp_mlps[k + 1][-1] if k + 1 < len(fp_mlps) else channel_out self.FP_modules.append( pointnet2_modules.PointnetFPModule( mlp=[pre_channel + skip_channel_list[k + l_skipped]] + fp_mlps[k] ) ) self.num_point_features = fp_mlps[0][-1] else: self.FP_modules = None def break_up_pc(self, pc): batch_idx = pc[:, 0] xyz = pc[:, 1:4].contiguous() features = (pc[:, 4:].contiguous() if pc.size(-1) > 4 else None) return batch_idx, xyz, features def forward(self, batch_dict): """ Args: batch_dict: batch_size: int points: (num_points, 4 + C), [batch_idx, x, y, z, ...] Returns: batch_dict: point_coords: (N, 3) point_features: (N, C) point_confidence_scores: (N, 1) """ batch_size = batch_dict['batch_size'] points = batch_dict['points'] batch_idx, xyz, features = self.break_up_pc(points) xyz_batch_cnt = xyz.new_zeros(batch_size).int() for bs_idx in range(batch_size): xyz_batch_cnt[bs_idx] = (batch_idx == bs_idx).sum() assert xyz_batch_cnt.min() == xyz_batch_cnt.max() xyz = xyz.view(batch_size, -1, 3).contiguous() features = features.view(batch_size, -1, features.shape[-1]) if features is not None else None features = features.permute(0, 2, 1).contiguous() if features is not None else None batch_idx = batch_idx.view(batch_size, -1).float() l_xyz, l_features, l_scores = [xyz], [features], [None] for i in range(len(self.SA_modules)): li_xyz, li_features, li_scores = self.SA_modules[i]( l_xyz[i], l_features[i], scores=l_scores[i]) l_xyz.append(li_xyz) l_features.append(li_features) l_scores.append(li_scores) # prepare for confidence loss l_xyz_flatten, l_scores_flatten = [], [] for i in range(1, len(l_xyz)): l_xyz_flatten.append(torch.cat([ batch_idx[:, :l_xyz[i].size(1)].reshape(-1, 1), l_xyz[i].reshape(-1, 3) ], dim=1)) # (N, 4) for i in range(1, len(l_scores)): if l_scores[i] is None: l_scores_flatten.append(None) else: l_scores_flatten.append(l_scores[i].reshape(-1, 1)) # (N, 1) batch_dict['point_coords_list'] = l_xyz_flatten batch_dict['point_scores_list'] = l_scores_flatten if self.FP_modules is not None: for i in range(-1, -(len(self.FP_modules) + 1), -1): l_features[i - 1] = self.FP_modules[i]( l_xyz[i - 1], l_xyz[i], l_features[i - 1], l_features[i] ) # (B, C, N) else: # take l_xyz[i - 1] and l_features[i - 1] i = 0 point_features = l_features[i - 1].permute(0, 2, 1).contiguous() # (B, N, C) batch_dict['point_features'] = point_features.view(-1, point_features.shape[-1]) batch_dict['point_coords'] = torch.cat(( batch_idx[:, :l_xyz[i - 1].size(1)].reshape(-1, 1).float(), l_xyz[i - 1].view(-1, 3)), dim=1) batch_dict['point_scores'] = l_scores[-1] # (B, N) return batch_dict class PointNet2Backbone(nn.Module): """ DO NOT USE THIS CURRENTLY SINCE IT MAY HAVE POTENTIAL BUGS, 20200723 """ def __init__(self, model_cfg, input_channels, **kwargs): assert False, 'DO NOT USE THIS CURRENTLY SINCE IT MAY HAVE POTENTIAL BUGS, 20200723' super().__init__() self.model_cfg = model_cfg self.SA_modules = nn.ModuleList() channel_in = input_channels - 3 self.num_points_each_layer = [] skip_channel_list = [input_channels] for k in range(self.model_cfg.SA_CONFIG.NPOINTS.__len__()): self.num_points_each_layer.append(self.model_cfg.SA_CONFIG.NPOINTS[k]) mlps = self.model_cfg.SA_CONFIG.MLPS[k].copy() channel_out = 0 for idx in range(mlps.__len__()): mlps[idx] = [channel_in] + mlps[idx] channel_out += mlps[idx][-1] self.SA_modules.append( pointnet2_modules_stack.StackSAModuleMSG( radii=self.model_cfg.SA_CONFIG.RADIUS[k], nsamples=self.model_cfg.SA_CONFIG.NSAMPLE[k], mlps=mlps, use_xyz=self.model_cfg.SA_CONFIG.get('USE_XYZ', True), ) ) skip_channel_list.append(channel_out) channel_in = channel_out self.FP_modules = nn.ModuleList() for k in range(self.model_cfg.FP_MLPS.__len__()): pre_channel = self.model_cfg.FP_MLPS[k + 1][-1] if k + 1 < len(self.model_cfg.FP_MLPS) else channel_out self.FP_modules.append( pointnet2_modules_stack.StackPointnetFPModule( mlp=[pre_channel + skip_channel_list[k]] + self.model_cfg.FP_MLPS[k] ) ) self.num_point_features = self.model_cfg.FP_MLPS[0][-1] def break_up_pc(self, pc): batch_idx = pc[:, 0] xyz = pc[:, 1:4].contiguous() features = (pc[:, 4:].contiguous() if pc.size(-1) > 4 else None) return batch_idx, xyz, features def forward(self, batch_dict): """ Args: batch_dict: batch_size: int vfe_features: (num_voxels, C) points: (num_points, 4 + C), [batch_idx, x, y, z, ...] Returns: batch_dict: encoded_spconv_tensor: sparse tensor point_features: (N, C) """ batch_size = batch_dict['batch_size'] points = batch_dict['points'] batch_idx, xyz, features = self.break_up_pc(points) xyz_batch_cnt = xyz.new_zeros(batch_size).int() for bs_idx in range(batch_size): xyz_batch_cnt[bs_idx] = (batch_idx == bs_idx).sum() l_xyz, l_features, l_batch_cnt = [xyz], [features], [xyz_batch_cnt] for i in range(len(self.SA_modules)): new_xyz_list = [] for k in range(batch_size): if len(l_xyz) == 1: cur_xyz = l_xyz[0][batch_idx == k] else: last_num_points = self.num_points_each_layer[i - 1] cur_xyz = l_xyz[-1][k * last_num_points: (k + 1) * last_num_points] cur_pt_idxs = pointnet2_utils_stack.furthest_point_sample( cur_xyz[None, :, :].contiguous(), self.num_points_each_layer[i] ).long()[0] if cur_xyz.shape[0] < self.num_points_each_layer[i]: empty_num = self.num_points_each_layer[i] - cur_xyz.shape[1] cur_pt_idxs[0, -empty_num:] = cur_pt_idxs[0, :empty_num] new_xyz_list.append(cur_xyz[cur_pt_idxs]) new_xyz = torch.cat(new_xyz_list, dim=0) new_xyz_batch_cnt = xyz.new_zeros(batch_size).int().fill_(self.num_points_each_layer[i]) li_xyz, li_features = self.SA_modules[i]( xyz=l_xyz[i], features=l_features[i], xyz_batch_cnt=l_batch_cnt[i], new_xyz=new_xyz, new_xyz_batch_cnt=new_xyz_batch_cnt ) l_xyz.append(li_xyz) l_features.append(li_features) l_batch_cnt.append(new_xyz_batch_cnt) l_features[0] = points[:, 1:] for i in range(-1, -(len(self.FP_modules) + 1), -1): l_features[i - 1] = self.FP_modules[i]( unknown=l_xyz[i - 1], unknown_batch_cnt=l_batch_cnt[i - 1], known=l_xyz[i], known_batch_cnt=l_batch_cnt[i], unknown_feats=l_features[i - 1], known_feats=l_features[i] ) batch_dict['point_features'] = l_features[0] batch_dict['point_coords'] = torch.cat((batch_idx[:, None].float(), l_xyz[0]), dim=1) return batch_dict
15,273
41.077135
119
py
SASA
SASA-main/pcdet/models/backbones_3d/pfe/__init__.py
from .voxel_set_abstraction import VoxelSetAbstraction __all__ = { 'VoxelSetAbstraction': VoxelSetAbstraction }
117
18.666667
54
py
SASA
SASA-main/pcdet/models/backbones_3d/pfe/voxel_set_abstraction.py
import torch import torch.nn as nn from ....ops.pointnet2.pointnet2_stack import pointnet2_modules as pointnet2_stack_modules from ....ops.pointnet2.pointnet2_stack import pointnet2_utils as pointnet2_stack_utils from ....utils import common_utils def bilinear_interpolate_torch(im, x, y): """ Args: im: (H, W, C) [y, x] x: (N) y: (N) Returns: """ x0 = torch.floor(x).long() x1 = x0 + 1 y0 = torch.floor(y).long() y1 = y0 + 1 x0 = torch.clamp(x0, 0, im.shape[1] - 1) x1 = torch.clamp(x1, 0, im.shape[1] - 1) y0 = torch.clamp(y0, 0, im.shape[0] - 1) y1 = torch.clamp(y1, 0, im.shape[0] - 1) Ia = im[y0, x0] Ib = im[y1, x0] Ic = im[y0, x1] Id = im[y1, x1] wa = (x1.type_as(x) - x) * (y1.type_as(y) - y) wb = (x1.type_as(x) - x) * (y - y0.type_as(y)) wc = (x - x0.type_as(x)) * (y1.type_as(y) - y) wd = (x - x0.type_as(x)) * (y - y0.type_as(y)) ans = torch.t((torch.t(Ia) * wa)) + torch.t(torch.t(Ib) * wb) + torch.t(torch.t(Ic) * wc) + torch.t(torch.t(Id) * wd) return ans class VoxelSetAbstraction(nn.Module): def __init__(self, model_cfg, voxel_size, point_cloud_range, num_bev_features=None, num_rawpoint_features=None, **kwargs): super().__init__() self.model_cfg = model_cfg self.voxel_size = voxel_size self.point_cloud_range = point_cloud_range SA_cfg = self.model_cfg.SA_LAYER self.SA_layers = nn.ModuleList() self.SA_layer_names = [] self.downsample_times_map = {} c_in = 0 for src_name in self.model_cfg.FEATURES_SOURCE: if src_name in ['bev', 'raw_points']: continue self.downsample_times_map[src_name] = SA_cfg[src_name].DOWNSAMPLE_FACTOR mlps = SA_cfg[src_name].MLPS for k in range(len(mlps)): mlps[k] = [mlps[k][0]] + mlps[k] cur_layer = pointnet2_stack_modules.StackSAModuleMSG( radii=SA_cfg[src_name].POOL_RADIUS, nsamples=SA_cfg[src_name].NSAMPLE, mlps=mlps, use_xyz=True, pool_method='max_pool', ) self.SA_layers.append(cur_layer) self.SA_layer_names.append(src_name) c_in += sum([x[-1] for x in mlps]) if 'bev' in self.model_cfg.FEATURES_SOURCE: c_bev = num_bev_features c_in += c_bev if 'raw_points' in self.model_cfg.FEATURES_SOURCE: mlps = SA_cfg['raw_points'].MLPS for k in range(len(mlps)): mlps[k] = [num_rawpoint_features - 3] + mlps[k] self.SA_rawpoints = pointnet2_stack_modules.StackSAModuleMSG( radii=SA_cfg['raw_points'].POOL_RADIUS, nsamples=SA_cfg['raw_points'].NSAMPLE, mlps=mlps, use_xyz=True, pool_method='max_pool' ) c_in += sum([x[-1] for x in mlps]) self.vsa_point_feature_fusion = nn.Sequential( nn.Linear(c_in, self.model_cfg.NUM_OUTPUT_FEATURES, bias=False), nn.BatchNorm1d(self.model_cfg.NUM_OUTPUT_FEATURES), nn.ReLU(), ) self.num_point_features = self.model_cfg.NUM_OUTPUT_FEATURES self.num_point_features_before_fusion = c_in def interpolate_from_bev_features(self, keypoints, bev_features, batch_size, bev_stride): x_idxs = (keypoints[:, :, 0] - self.point_cloud_range[0]) / self.voxel_size[0] y_idxs = (keypoints[:, :, 1] - self.point_cloud_range[1]) / self.voxel_size[1] x_idxs = x_idxs / bev_stride y_idxs = y_idxs / bev_stride point_bev_features_list = [] for k in range(batch_size): cur_x_idxs = x_idxs[k] cur_y_idxs = y_idxs[k] cur_bev_features = bev_features[k].permute(1, 2, 0) # (H, W, C) point_bev_features = bilinear_interpolate_torch(cur_bev_features, cur_x_idxs, cur_y_idxs) point_bev_features_list.append(point_bev_features.unsqueeze(dim=0)) point_bev_features = torch.cat(point_bev_features_list, dim=0) # (B, N, C0) return point_bev_features def get_sampled_points(self, batch_dict): batch_size = batch_dict['batch_size'] if self.model_cfg.POINT_SOURCE == 'raw_points': src_points = batch_dict['points'][:, 1:4] batch_indices = batch_dict['points'][:, 0].long() elif self.model_cfg.POINT_SOURCE == 'voxel_centers': src_points = common_utils.get_voxel_centers( batch_dict['voxel_coords'][:, 1:4], downsample_times=1, voxel_size=self.voxel_size, point_cloud_range=self.point_cloud_range ) batch_indices = batch_dict['voxel_coords'][:, 0].long() else: raise NotImplementedError keypoints_list = [] for bs_idx in range(batch_size): bs_mask = (batch_indices == bs_idx) sampled_points = src_points[bs_mask].unsqueeze(dim=0) # (1, N, 3) if self.model_cfg.SAMPLE_METHOD == 'FPS': cur_pt_idxs = pointnet2_stack_utils.furthest_point_sample( sampled_points[:, :, 0:3].contiguous(), self.model_cfg.NUM_KEYPOINTS ).long() if sampled_points.shape[1] < self.model_cfg.NUM_KEYPOINTS: empty_num = self.model_cfg.NUM_KEYPOINTS - sampled_points.shape[1] cur_pt_idxs[0, -empty_num:] = cur_pt_idxs[0, :empty_num] keypoints = sampled_points[0][cur_pt_idxs[0]].unsqueeze(dim=0) elif self.model_cfg.SAMPLE_METHOD == 'FastFPS': raise NotImplementedError else: raise NotImplementedError keypoints_list.append(keypoints) keypoints = torch.cat(keypoints_list, dim=0) # (B, M, 3) return keypoints def forward(self, batch_dict): """ Args: batch_dict: batch_size: keypoints: (B, num_keypoints, 3) multi_scale_3d_features: { 'x_conv4': ... } points: optional (N, 1 + 3 + C) [bs_idx, x, y, z, ...] spatial_features: optional spatial_features_stride: optional Returns: point_features: (N, C) point_coords: (N, 4) """ keypoints = self.get_sampled_points(batch_dict) point_features_list = [] if 'bev' in self.model_cfg.FEATURES_SOURCE: point_bev_features = self.interpolate_from_bev_features( keypoints, batch_dict['spatial_features'], batch_dict['batch_size'], bev_stride=batch_dict['spatial_features_stride'] ) point_features_list.append(point_bev_features) batch_size, num_keypoints, _ = keypoints.shape new_xyz = keypoints.view(-1, 3) new_xyz_batch_cnt = new_xyz.new_zeros(batch_size).int().fill_(num_keypoints) if 'raw_points' in self.model_cfg.FEATURES_SOURCE: raw_points = batch_dict['points'] xyz = raw_points[:, 1:4] xyz_batch_cnt = xyz.new_zeros(batch_size).int() for bs_idx in range(batch_size): xyz_batch_cnt[bs_idx] = (raw_points[:, 0] == bs_idx).sum() point_features = raw_points[:, 4:].contiguous() if raw_points.shape[1] > 4 else None pooled_points, pooled_features = self.SA_rawpoints( xyz=xyz.contiguous(), xyz_batch_cnt=xyz_batch_cnt, new_xyz=new_xyz, new_xyz_batch_cnt=new_xyz_batch_cnt, features=point_features, ) point_features_list.append(pooled_features.view(batch_size, num_keypoints, -1)) for k, src_name in enumerate(self.SA_layer_names): cur_coords = batch_dict['multi_scale_3d_features'][src_name].indices xyz = common_utils.get_voxel_centers( cur_coords[:, 1:4], downsample_times=self.downsample_times_map[src_name], voxel_size=self.voxel_size, point_cloud_range=self.point_cloud_range ) xyz_batch_cnt = xyz.new_zeros(batch_size).int() for bs_idx in range(batch_size): xyz_batch_cnt[bs_idx] = (cur_coords[:, 0] == bs_idx).sum() pooled_points, pooled_features = self.SA_layers[k]( xyz=xyz.contiguous(), xyz_batch_cnt=xyz_batch_cnt, new_xyz=new_xyz, new_xyz_batch_cnt=new_xyz_batch_cnt, features=batch_dict['multi_scale_3d_features'][src_name].features.contiguous(), ) point_features_list.append(pooled_features.view(batch_size, num_keypoints, -1)) point_features = torch.cat(point_features_list, dim=2) batch_idx = torch.arange(batch_size, device=keypoints.device).view(-1, 1).repeat(1, keypoints.shape[1]).view(-1) point_coords = torch.cat((batch_idx.view(-1, 1).float(), keypoints.view(-1, 3)), dim=1) batch_dict['point_features_before_fusion'] = point_features.view(-1, point_features.shape[-1]) point_features = self.vsa_point_feature_fusion(point_features.view(-1, point_features.shape[-1])) batch_dict['point_features'] = point_features # (BxN, C) batch_dict['point_coords'] = point_coords # (BxN, 4) return batch_dict
9,638
39.1625
121
py
SASA
SASA-main/pcdet/models/backbones_3d/vfe/vfe_template.py
import torch.nn as nn class VFETemplate(nn.Module): def __init__(self, model_cfg, **kwargs): super().__init__() self.model_cfg = model_cfg def get_output_feature_dim(self): raise NotImplementedError def forward(self, **kwargs): """ Args: **kwargs: Returns: batch_dict: ... vfe_features: (num_voxels, C) """ raise NotImplementedError
470
19.478261
45
py
SASA
SASA-main/pcdet/models/backbones_3d/vfe/mean_vfe.py
import torch from .vfe_template import VFETemplate class MeanVFE(VFETemplate): def __init__(self, model_cfg, num_point_features, **kwargs): super().__init__(model_cfg=model_cfg) self.num_point_features = num_point_features def get_output_feature_dim(self): return self.num_point_features def forward(self, batch_dict, **kwargs): """ Args: batch_dict: voxels: (num_voxels, max_points_per_voxel, C) voxel_num_points: optional (num_voxels) **kwargs: Returns: vfe_features: (num_voxels, C) """ voxel_features, voxel_num_points = batch_dict['voxels'], batch_dict['voxel_num_points'] points_mean = voxel_features[:, :, :].sum(dim=1, keepdim=False) normalizer = torch.clamp_min(voxel_num_points.view(-1, 1), min=1.0).type_as(voxel_features) points_mean = points_mean / normalizer batch_dict['voxel_features'] = points_mean.contiguous() return batch_dict
1,038
31.46875
99
py
SASA
SASA-main/pcdet/models/backbones_3d/vfe/pillar_vfe.py
import torch import torch.nn as nn import torch.nn.functional as F from .vfe_template import VFETemplate class PFNLayer(nn.Module): def __init__(self, in_channels, out_channels, use_norm=True, last_layer=False): super().__init__() self.last_vfe = last_layer self.use_norm = use_norm if not self.last_vfe: out_channels = out_channels // 2 if self.use_norm: self.linear = nn.Linear(in_channels, out_channels, bias=False) self.norm = nn.BatchNorm1d(out_channels, eps=1e-3, momentum=0.01) else: self.linear = nn.Linear(in_channels, out_channels, bias=True) self.part = 50000 def forward(self, inputs): if inputs.shape[0] > self.part: # nn.Linear performs randomly when batch size is too large num_parts = inputs.shape[0] // self.part part_linear_out = [self.linear(inputs[num_part*self.part:(num_part+1)*self.part]) for num_part in range(num_parts+1)] x = torch.cat(part_linear_out, dim=0) else: x = self.linear(inputs) torch.backends.cudnn.enabled = False x = self.norm(x.permute(0, 2, 1)).permute(0, 2, 1) if self.use_norm else x torch.backends.cudnn.enabled = True x = F.relu(x) x_max = torch.max(x, dim=1, keepdim=True)[0] if self.last_vfe: return x_max else: x_repeat = x_max.repeat(1, inputs.shape[1], 1) x_concatenated = torch.cat([x, x_repeat], dim=2) return x_concatenated class PillarVFE(VFETemplate): def __init__(self, model_cfg, num_point_features, voxel_size, point_cloud_range): super().__init__(model_cfg=model_cfg) self.use_norm = self.model_cfg.USE_NORM self.with_distance = self.model_cfg.WITH_DISTANCE self.use_absolute_xyz = self.model_cfg.USE_ABSLOTE_XYZ num_point_features += 6 if self.use_absolute_xyz else 3 if self.with_distance: num_point_features += 1 self.num_filters = self.model_cfg.NUM_FILTERS assert len(self.num_filters) > 0 num_filters = [num_point_features] + list(self.num_filters) pfn_layers = [] for i in range(len(num_filters) - 1): in_filters = num_filters[i] out_filters = num_filters[i + 1] pfn_layers.append( PFNLayer(in_filters, out_filters, self.use_norm, last_layer=(i >= len(num_filters) - 2)) ) self.pfn_layers = nn.ModuleList(pfn_layers) self.voxel_x = voxel_size[0] self.voxel_y = voxel_size[1] self.voxel_z = voxel_size[2] self.x_offset = self.voxel_x / 2 + point_cloud_range[0] self.y_offset = self.voxel_y / 2 + point_cloud_range[1] self.z_offset = self.voxel_z / 2 + point_cloud_range[2] def get_output_feature_dim(self): return self.num_filters[-1] def get_paddings_indicator(self, actual_num, max_num, axis=0): actual_num = torch.unsqueeze(actual_num, axis + 1) max_num_shape = [1] * len(actual_num.shape) max_num_shape[axis + 1] = -1 max_num = torch.arange(max_num, dtype=torch.int, device=actual_num.device).view(max_num_shape) paddings_indicator = actual_num.int() > max_num return paddings_indicator def forward(self, batch_dict, **kwargs): voxel_features, voxel_num_points, coords = batch_dict['voxels'], batch_dict['voxel_num_points'], batch_dict['voxel_coords'] points_mean = voxel_features[:, :, :3].sum(dim=1, keepdim=True) / voxel_num_points.type_as(voxel_features).view(-1, 1, 1) f_cluster = voxel_features[:, :, :3] - points_mean f_center = torch.zeros_like(voxel_features[:, :, :3]) f_center[:, :, 0] = voxel_features[:, :, 0] - (coords[:, 3].to(voxel_features.dtype).unsqueeze(1) * self.voxel_x + self.x_offset) f_center[:, :, 1] = voxel_features[:, :, 1] - (coords[:, 2].to(voxel_features.dtype).unsqueeze(1) * self.voxel_y + self.y_offset) f_center[:, :, 2] = voxel_features[:, :, 2] - (coords[:, 1].to(voxel_features.dtype).unsqueeze(1) * self.voxel_z + self.z_offset) if self.use_absolute_xyz: features = [voxel_features, f_cluster, f_center] else: features = [voxel_features[..., 3:], f_cluster, f_center] if self.with_distance: points_dist = torch.norm(voxel_features[:, :, :3], 2, 2, keepdim=True) features.append(points_dist) features = torch.cat(features, dim=-1) voxel_count = features.shape[1] mask = self.get_paddings_indicator(voxel_num_points, voxel_count, axis=0) mask = torch.unsqueeze(mask, -1).type_as(voxel_features) features *= mask for pfn in self.pfn_layers: features = pfn(features) features = features.squeeze() batch_dict['pillar_features'] = features return batch_dict
5,089
40.048387
137
py
SASA
SASA-main/pcdet/models/backbones_3d/vfe/__init__.py
from .mean_vfe import MeanVFE from .pillar_vfe import PillarVFE from .vfe_template import VFETemplate __all__ = { 'VFETemplate': VFETemplate, 'MeanVFE': MeanVFE, 'PillarVFE': PillarVFE }
200
19.1
37
py
SASA
SASA-main/pcdet/models/dense_heads/anchor_head_single.py
import numpy as np import torch.nn as nn from .anchor_head_template import AnchorHeadTemplate class AnchorHeadSingle(AnchorHeadTemplate): def __init__(self, model_cfg, input_channels, num_class, class_names, grid_size, point_cloud_range, predict_boxes_when_training=True, **kwargs): super().__init__( model_cfg=model_cfg, num_class=num_class, class_names=class_names, grid_size=grid_size, point_cloud_range=point_cloud_range, predict_boxes_when_training=predict_boxes_when_training ) self.num_anchors_per_location = sum(self.num_anchors_per_location) self.conv_cls = nn.Conv2d( input_channels, self.num_anchors_per_location * self.num_class, kernel_size=1 ) self.conv_box = nn.Conv2d( input_channels, self.num_anchors_per_location * self.box_coder.code_size, kernel_size=1 ) if self.model_cfg.get('USE_DIRECTION_CLASSIFIER', None) is not None: self.conv_dir_cls = nn.Conv2d( input_channels, self.num_anchors_per_location * self.model_cfg.NUM_DIR_BINS, kernel_size=1 ) else: self.conv_dir_cls = None self.init_weights() def init_weights(self): pi = 0.01 nn.init.constant_(self.conv_cls.bias, -np.log((1 - pi) / pi)) nn.init.normal_(self.conv_box.weight, mean=0, std=0.001) def forward(self, data_dict): spatial_features_2d = data_dict['spatial_features_2d'] cls_preds = self.conv_cls(spatial_features_2d) box_preds = self.conv_box(spatial_features_2d) cls_preds = cls_preds.permute(0, 2, 3, 1).contiguous() # [N, H, W, C] box_preds = box_preds.permute(0, 2, 3, 1).contiguous() # [N, H, W, C] self.forward_ret_dict['cls_preds'] = cls_preds self.forward_ret_dict['box_preds'] = box_preds if self.conv_dir_cls is not None: dir_cls_preds = self.conv_dir_cls(spatial_features_2d) dir_cls_preds = dir_cls_preds.permute(0, 2, 3, 1).contiguous() self.forward_ret_dict['dir_cls_preds'] = dir_cls_preds else: dir_cls_preds = None if self.training: targets_dict = self.assign_targets( gt_boxes=data_dict['gt_boxes'] ) self.forward_ret_dict.update(targets_dict) if not self.training or self.predict_boxes_when_training: batch_cls_preds, batch_box_preds = self.generate_predicted_boxes( batch_size=data_dict['batch_size'], cls_preds=cls_preds, box_preds=box_preds, dir_cls_preds=dir_cls_preds ) data_dict['batch_cls_preds'] = batch_cls_preds data_dict['batch_box_preds'] = batch_box_preds data_dict['cls_preds_normalized'] = False return data_dict
2,928
37.539474
136
py
SASA
SASA-main/pcdet/models/dense_heads/point_head_template.py
import torch import torch.nn as nn import torch.nn.functional as F from ...ops.roiaware_pool3d import roiaware_pool3d_utils from ...utils import common_utils, loss_utils class PointHeadTemplate(nn.Module): def __init__(self, model_cfg, num_class): super().__init__() self.model_cfg = model_cfg self.num_class = num_class self.build_losses(self.model_cfg.LOSS_CONFIG) self.forward_ret_dict = None def build_losses(self, losses_cfg): self.add_module( 'cls_loss_func', loss_utils.SigmoidFocalClassificationLoss(alpha=0.25, gamma=2.0) ) self.add_module( 'reg_loss_func', loss_utils.WeightedSmoothL1Loss( code_weights=losses_cfg.LOSS_WEIGHTS.get('code_weights', None) ) ) @staticmethod def make_fc_layers(fc_cfg, input_channels, output_channels): fc_layers = [] c_in = input_channels for k in range(0, fc_cfg.__len__()): fc_layers.extend([ nn.Linear(c_in, fc_cfg[k], bias=False), nn.BatchNorm1d(fc_cfg[k]), nn.ReLU(), ]) c_in = fc_cfg[k] fc_layers.append(nn.Linear(c_in, output_channels, bias=True)) return nn.Sequential(*fc_layers) def assign_stack_targets(self, points, gt_boxes, extend_gt_boxes=None, ret_box_labels=False, ret_part_labels=False, set_ignore_flag=True, use_ball_constraint=False, central_radius=2.0): """ Args: points: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] gt_boxes: (B, M, 8) extend_gt_boxes: [B, M, 8] ret_box_labels: ret_part_labels: set_ignore_flag: use_ball_constraint: central_radius: Returns: point_cls_labels: (N1 + N2 + N3 + ...), long type, 0:background, -1:ignored point_box_labels: (N1 + N2 + N3 + ..., code_size) """ assert len(points.shape) == 2 and points.shape[1] == 4, 'points.shape=%s' % str(points.shape) assert len(gt_boxes.shape) == 3 and gt_boxes.shape[2] == 8, 'gt_boxes.shape=%s' % str(gt_boxes.shape) assert extend_gt_boxes is None or len(extend_gt_boxes.shape) == 3 and extend_gt_boxes.shape[2] == 8, \ 'extend_gt_boxes.shape=%s' % str(extend_gt_boxes.shape) assert set_ignore_flag != use_ball_constraint, 'Choose one only!' batch_size = gt_boxes.shape[0] bs_idx = points[:, 0] point_cls_labels = points.new_zeros(points.shape[0]).long() point_box_labels = gt_boxes.new_zeros((points.shape[0], 8)) if ret_box_labels else None point_part_labels = gt_boxes.new_zeros((points.shape[0], 3)) if ret_part_labels else None for k in range(batch_size): bs_mask = (bs_idx == k) points_single = points[bs_mask][:, 1:4] point_cls_labels_single = point_cls_labels.new_zeros(bs_mask.sum()) box_idxs_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( points_single.unsqueeze(dim=0), gt_boxes[k:k + 1, :, 0:7].contiguous() ).long().squeeze(dim=0) box_fg_flag = (box_idxs_of_pts >= 0) if set_ignore_flag: extend_box_idxs_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( points_single.unsqueeze(dim=0), extend_gt_boxes[k:k+1, :, 0:7].contiguous() ).long().squeeze(dim=0) fg_flag = box_fg_flag ignore_flag = fg_flag ^ (extend_box_idxs_of_pts >= 0) point_cls_labels_single[ignore_flag] = -1 elif use_ball_constraint: box_centers = gt_boxes[k][box_idxs_of_pts][:, 0:3].clone() box_centers[:, 2] += gt_boxes[k][box_idxs_of_pts][:, 5] / 2 ball_flag = ((box_centers - points_single).norm(dim=1) < central_radius) fg_flag = box_fg_flag & ball_flag else: raise NotImplementedError gt_box_of_fg_points = gt_boxes[k][box_idxs_of_pts[fg_flag]] point_cls_labels_single[fg_flag] = 1 if self.num_class == 1 else gt_box_of_fg_points[:, -1].long() point_cls_labels[bs_mask] = point_cls_labels_single if ret_box_labels and gt_box_of_fg_points.shape[0] > 0: point_box_labels_single = point_box_labels.new_zeros((bs_mask.sum(), 8)) fg_point_box_labels = self.box_coder.encode_torch( gt_boxes=gt_box_of_fg_points[:, :-1], points=points_single[fg_flag], gt_classes=gt_box_of_fg_points[:, -1].long() ) point_box_labels_single[fg_flag] = fg_point_box_labels point_box_labels[bs_mask] = point_box_labels_single if ret_part_labels: point_part_labels_single = point_part_labels.new_zeros((bs_mask.sum(), 3)) transformed_points = points_single[fg_flag] - gt_box_of_fg_points[:, 0:3] transformed_points = common_utils.rotate_points_along_z( transformed_points.view(-1, 1, 3), -gt_box_of_fg_points[:, 6] ).view(-1, 3) offset = torch.tensor([0.5, 0.5, 0.5]).view(1, 3).type_as(transformed_points) point_part_labels_single[fg_flag] = (transformed_points / gt_box_of_fg_points[:, 3:6]) + offset point_part_labels[bs_mask] = point_part_labels_single targets_dict = { 'point_cls_labels': point_cls_labels, 'point_box_labels': point_box_labels, 'point_part_labels': point_part_labels } return targets_dict def get_cls_layer_loss(self, tb_dict=None): point_cls_labels = self.forward_ret_dict['point_cls_labels'].view(-1) point_cls_preds = self.forward_ret_dict['point_cls_preds'].view(-1, self.num_class) positives = (point_cls_labels > 0) negative_cls_weights = (point_cls_labels == 0) * 1.0 cls_weights = (negative_cls_weights + 15.0 * positives).float() pos_normalizer = positives.sum(dim=0).float() cls_weights /= torch.clamp(pos_normalizer, min=1.0) one_hot_targets = point_cls_preds.new_zeros(*list(point_cls_labels.shape), self.num_class + 1) one_hot_targets.scatter_(-1, (point_cls_labels * (point_cls_labels >= 0).long()).unsqueeze(dim=-1).long(), 1.0) one_hot_targets = one_hot_targets[..., 1:] cls_loss_src = self.cls_loss_func(point_cls_preds, one_hot_targets, weights=cls_weights) point_loss_cls = cls_loss_src.sum() loss_weights_dict = self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS point_loss_cls = point_loss_cls * loss_weights_dict['point_cls_weight'] if tb_dict is None: tb_dict = {} tb_dict.update({ 'point_loss_cls': point_loss_cls.item(), 'point_pos_num': pos_normalizer.item() }) return point_loss_cls, tb_dict def get_part_layer_loss(self, tb_dict=None): pos_mask = self.forward_ret_dict['point_cls_labels'] > 0 pos_normalizer = max(1, (pos_mask > 0).sum().item()) point_part_labels = self.forward_ret_dict['point_part_labels'] point_part_preds = self.forward_ret_dict['point_part_preds'] point_loss_part = F.binary_cross_entropy(torch.sigmoid(point_part_preds), point_part_labels, reduction='none') point_loss_part = (point_loss_part.sum(dim=-1) * pos_mask.float()).sum() / (3 * pos_normalizer) loss_weights_dict = self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS point_loss_part = point_loss_part * loss_weights_dict['point_part_weight'] if tb_dict is None: tb_dict = {} tb_dict.update({'point_loss_part': point_loss_part.item()}) return point_loss_part, tb_dict def get_box_layer_loss(self, tb_dict=None): pos_mask = self.forward_ret_dict['point_cls_labels'] > 0 point_box_labels = self.forward_ret_dict['point_box_labels'] point_box_preds = self.forward_ret_dict['point_box_preds'] reg_weights = pos_mask.float() pos_normalizer = pos_mask.sum().float() reg_weights /= torch.clamp(pos_normalizer, min=1.0) point_loss_box_src = self.reg_loss_func( point_box_preds[None, ...], point_box_labels[None, ...], weights=reg_weights[None, ...] ) point_loss_box = point_loss_box_src.sum() loss_weights_dict = self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS point_loss_box = point_loss_box * loss_weights_dict['point_box_weight'] if tb_dict is None: tb_dict = {} tb_dict.update({'point_loss_box': point_loss_box.item()}) return point_loss_box, tb_dict def generate_predicted_boxes(self, points, point_cls_preds, point_box_preds): """ Args: points: (N, 3) point_cls_preds: (N, num_class) point_box_preds: (N, box_code_size) Returns: point_cls_preds: (N, num_class) point_box_preds: (N, box_code_size) """ _, pred_classes = point_cls_preds.max(dim=-1) point_box_preds = self.box_coder.decode_torch(point_box_preds, points, pred_classes + 1) return point_cls_preds, point_box_preds def forward(self, **kwargs): raise NotImplementedError
9,475
45
119
py
SASA
SASA-main/pcdet/models/dense_heads/anchor_head_template.py
import numpy as np import torch import torch.nn as nn from ...utils import box_coder_utils, common_utils, loss_utils from .target_assigner.anchor_generator import AnchorGenerator from .target_assigner.atss_target_assigner import ATSSTargetAssigner from .target_assigner.axis_aligned_target_assigner import AxisAlignedTargetAssigner class AnchorHeadTemplate(nn.Module): def __init__(self, model_cfg, num_class, class_names, grid_size, point_cloud_range, predict_boxes_when_training): super().__init__() self.model_cfg = model_cfg self.num_class = num_class self.class_names = class_names self.predict_boxes_when_training = predict_boxes_when_training self.use_multihead = self.model_cfg.get('USE_MULTIHEAD', False) anchor_target_cfg = self.model_cfg.TARGET_ASSIGNER_CONFIG self.box_coder = getattr(box_coder_utils, anchor_target_cfg.BOX_CODER)( num_dir_bins=anchor_target_cfg.get('NUM_DIR_BINS', 6), **anchor_target_cfg.get('BOX_CODER_CONFIG', {}) ) anchor_generator_cfg = self.model_cfg.ANCHOR_GENERATOR_CONFIG anchors, self.num_anchors_per_location = self.generate_anchors( anchor_generator_cfg, grid_size=grid_size, point_cloud_range=point_cloud_range, anchor_ndim=self.box_coder.code_size ) self.anchors = [x.cuda() for x in anchors] self.target_assigner = self.get_target_assigner(anchor_target_cfg) self.forward_ret_dict = {} self.build_losses(self.model_cfg.LOSS_CONFIG) @staticmethod def generate_anchors(anchor_generator_cfg, grid_size, point_cloud_range, anchor_ndim=7): anchor_generator = AnchorGenerator( anchor_range=point_cloud_range, anchor_generator_config=anchor_generator_cfg ) feature_map_size = [grid_size[:2] // config['feature_map_stride'] for config in anchor_generator_cfg] anchors_list, num_anchors_per_location_list = anchor_generator.generate_anchors(feature_map_size) if anchor_ndim != 7: for idx, anchors in enumerate(anchors_list): pad_zeros = anchors.new_zeros([*anchors.shape[0:-1], anchor_ndim - 7]) new_anchors = torch.cat((anchors, pad_zeros), dim=-1) anchors_list[idx] = new_anchors return anchors_list, num_anchors_per_location_list def get_target_assigner(self, anchor_target_cfg): if anchor_target_cfg.NAME == 'ATSS': target_assigner = ATSSTargetAssigner( topk=anchor_target_cfg.TOPK, box_coder=self.box_coder, use_multihead=self.use_multihead, match_height=anchor_target_cfg.MATCH_HEIGHT ) elif anchor_target_cfg.NAME == 'AxisAlignedTargetAssigner': target_assigner = AxisAlignedTargetAssigner( model_cfg=self.model_cfg, class_names=self.class_names, box_coder=self.box_coder, match_height=anchor_target_cfg.MATCH_HEIGHT ) else: raise NotImplementedError return target_assigner def build_losses(self, losses_cfg): self.add_module( 'cls_loss_func', loss_utils.SigmoidFocalClassificationLoss(alpha=0.25, gamma=2.0) ) reg_loss_name = 'WeightedSmoothL1Loss' if losses_cfg.get('REG_LOSS_TYPE', None) is None \ else losses_cfg.REG_LOSS_TYPE self.add_module( 'reg_loss_func', getattr(loss_utils, reg_loss_name)(code_weights=losses_cfg.LOSS_WEIGHTS['code_weights']) ) self.add_module( 'dir_loss_func', loss_utils.WeightedCrossEntropyLoss() ) def assign_targets(self, gt_boxes): """ Args: gt_boxes: (B, M, 8) Returns: """ targets_dict = self.target_assigner.assign_targets( self.anchors, gt_boxes ) return targets_dict def get_cls_layer_loss(self): cls_preds = self.forward_ret_dict['cls_preds'] box_cls_labels = self.forward_ret_dict['box_cls_labels'] batch_size = int(cls_preds.shape[0]) cared = box_cls_labels >= 0 # [N, num_anchors] positives = box_cls_labels > 0 negatives = box_cls_labels == 0 negative_cls_weights = negatives * 1.0 cls_weights = (negative_cls_weights + 1.0 * positives).float() reg_weights = positives.float() if self.num_class == 1: # class agnostic box_cls_labels[positives] = 1 pos_normalizer = positives.sum(1, keepdim=True).float() reg_weights /= torch.clamp(pos_normalizer, min=1.0) cls_weights /= torch.clamp(pos_normalizer, min=1.0) cls_targets = box_cls_labels * cared.type_as(box_cls_labels) cls_targets = cls_targets.unsqueeze(dim=-1) cls_targets = cls_targets.squeeze(dim=-1) one_hot_targets = torch.zeros( *list(cls_targets.shape), self.num_class + 1, dtype=cls_preds.dtype, device=cls_targets.device ) one_hot_targets.scatter_(-1, cls_targets.unsqueeze(dim=-1).long(), 1.0) cls_preds = cls_preds.view(batch_size, -1, self.num_class) one_hot_targets = one_hot_targets[..., 1:] cls_loss_src = self.cls_loss_func(cls_preds, one_hot_targets, weights=cls_weights) # [N, M] cls_loss = cls_loss_src.sum() / batch_size cls_loss = cls_loss * self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS['cls_weight'] tb_dict = { 'rpn_loss_cls': cls_loss.item() } return cls_loss, tb_dict @staticmethod def add_sin_difference(boxes1, boxes2, dim=6): assert dim != -1 rad_pred_encoding = torch.sin(boxes1[..., dim:dim + 1]) * torch.cos(boxes2[..., dim:dim + 1]) rad_tg_encoding = torch.cos(boxes1[..., dim:dim + 1]) * torch.sin(boxes2[..., dim:dim + 1]) boxes1 = torch.cat([boxes1[..., :dim], rad_pred_encoding, boxes1[..., dim + 1:]], dim=-1) boxes2 = torch.cat([boxes2[..., :dim], rad_tg_encoding, boxes2[..., dim + 1:]], dim=-1) return boxes1, boxes2 @staticmethod def get_direction_target(anchors, reg_targets, one_hot=True, dir_offset=0, num_bins=2): batch_size = reg_targets.shape[0] anchors = anchors.view(batch_size, -1, anchors.shape[-1]) rot_gt = reg_targets[..., 6] + anchors[..., 6] offset_rot = common_utils.limit_period(rot_gt - dir_offset, 0, 2 * np.pi) dir_cls_targets = torch.floor(offset_rot / (2 * np.pi / num_bins)).long() dir_cls_targets = torch.clamp(dir_cls_targets, min=0, max=num_bins - 1) if one_hot: dir_targets = torch.zeros(*list(dir_cls_targets.shape), num_bins, dtype=anchors.dtype, device=dir_cls_targets.device) dir_targets.scatter_(-1, dir_cls_targets.unsqueeze(dim=-1).long(), 1.0) dir_cls_targets = dir_targets return dir_cls_targets def get_box_reg_layer_loss(self): box_preds = self.forward_ret_dict['box_preds'] box_dir_cls_preds = self.forward_ret_dict.get('dir_cls_preds', None) box_reg_targets = self.forward_ret_dict['box_reg_targets'] box_cls_labels = self.forward_ret_dict['box_cls_labels'] batch_size = int(box_preds.shape[0]) positives = box_cls_labels > 0 reg_weights = positives.float() pos_normalizer = positives.sum(1, keepdim=True).float() reg_weights /= torch.clamp(pos_normalizer, min=1.0) if isinstance(self.anchors, list): if self.use_multihead: anchors = torch.cat( [anchor.permute(3, 4, 0, 1, 2, 5).contiguous().view(-1, anchor.shape[-1]) for anchor in self.anchors], dim=0) else: anchors = torch.cat(self.anchors, dim=-3) else: anchors = self.anchors anchors = anchors.view(1, -1, anchors.shape[-1]).repeat(batch_size, 1, 1) box_preds = box_preds.view(batch_size, -1, box_preds.shape[-1] // self.num_anchors_per_location if not self.use_multihead else box_preds.shape[-1]) # sin(a - b) = sinacosb-cosasinb box_preds_sin, reg_targets_sin = self.add_sin_difference(box_preds, box_reg_targets) loc_loss_src = self.reg_loss_func(box_preds_sin, reg_targets_sin, weights=reg_weights) # [N, M] loc_loss = loc_loss_src.sum() / batch_size loc_loss = loc_loss * self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS['loc_weight'] box_loss = loc_loss tb_dict = { 'rpn_loss_loc': loc_loss.item() } if box_dir_cls_preds is not None: dir_targets = self.get_direction_target( anchors, box_reg_targets, dir_offset=self.model_cfg.DIR_OFFSET, num_bins=self.model_cfg.NUM_DIR_BINS ) dir_logits = box_dir_cls_preds.view(batch_size, -1, self.model_cfg.NUM_DIR_BINS) weights = positives.type_as(dir_logits) weights /= torch.clamp(weights.sum(-1, keepdim=True), min=1.0) dir_loss = self.dir_loss_func(dir_logits, dir_targets, weights=weights) dir_loss = dir_loss.sum() / batch_size dir_loss = dir_loss * self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS['dir_weight'] box_loss += dir_loss tb_dict['rpn_loss_dir'] = dir_loss.item() return box_loss, tb_dict def get_loss(self): cls_loss, tb_dict = self.get_cls_layer_loss() box_loss, tb_dict_box = self.get_box_reg_layer_loss() tb_dict.update(tb_dict_box) rpn_loss = cls_loss + box_loss tb_dict['rpn_loss'] = rpn_loss.item() return rpn_loss, tb_dict def generate_predicted_boxes(self, batch_size, cls_preds, box_preds, dir_cls_preds=None): """ Args: batch_size: cls_preds: (N, H, W, C1) box_preds: (N, H, W, C2) dir_cls_preds: (N, H, W, C3) Returns: batch_cls_preds: (B, num_boxes, num_classes) batch_box_preds: (B, num_boxes, 7+C) """ if isinstance(self.anchors, list): if self.use_multihead: anchors = torch.cat([anchor.permute(3, 4, 0, 1, 2, 5).contiguous().view(-1, anchor.shape[-1]) for anchor in self.anchors], dim=0) else: anchors = torch.cat(self.anchors, dim=-3) else: anchors = self.anchors num_anchors = anchors.view(-1, anchors.shape[-1]).shape[0] batch_anchors = anchors.view(1, -1, anchors.shape[-1]).repeat(batch_size, 1, 1) batch_cls_preds = cls_preds.view(batch_size, num_anchors, -1).float() \ if not isinstance(cls_preds, list) else cls_preds batch_box_preds = box_preds.view(batch_size, num_anchors, -1) if not isinstance(box_preds, list) \ else torch.cat(box_preds, dim=1).view(batch_size, num_anchors, -1) batch_box_preds = self.box_coder.decode_torch(batch_box_preds, batch_anchors) if dir_cls_preds is not None: dir_offset = self.model_cfg.DIR_OFFSET dir_limit_offset = self.model_cfg.DIR_LIMIT_OFFSET dir_cls_preds = dir_cls_preds.view(batch_size, num_anchors, -1) if not isinstance(dir_cls_preds, list) \ else torch.cat(dir_cls_preds, dim=1).view(batch_size, num_anchors, -1) dir_labels = torch.max(dir_cls_preds, dim=-1)[1] period = (2 * np.pi / self.model_cfg.NUM_DIR_BINS) dir_rot = common_utils.limit_period( batch_box_preds[..., 6] - dir_offset, dir_limit_offset, period ) batch_box_preds[..., 6] = dir_rot + dir_offset + period * dir_labels.to(batch_box_preds.dtype) if isinstance(self.box_coder, box_coder_utils.PreviousResidualDecoder): batch_box_preds[..., 6] = common_utils.limit_period( -(batch_box_preds[..., 6] + np.pi / 2), offset=0.5, period=np.pi * 2 ) return batch_cls_preds, batch_box_preds def forward(self, **kwargs): raise NotImplementedError
12,364
43.800725
118
py
SASA
SASA-main/pcdet/models/dense_heads/anchor_head_multi.py
import numpy as np import torch import torch.nn as nn from ..backbones_2d import BaseBEVBackbone from .anchor_head_template import AnchorHeadTemplate class SingleHead(BaseBEVBackbone): def __init__(self, model_cfg, input_channels, num_class, num_anchors_per_location, code_size, rpn_head_cfg=None, head_label_indices=None, separate_reg_config=None): super().__init__(rpn_head_cfg, input_channels) self.num_anchors_per_location = num_anchors_per_location self.num_class = num_class self.code_size = code_size self.model_cfg = model_cfg self.separate_reg_config = separate_reg_config self.register_buffer('head_label_indices', head_label_indices) if self.separate_reg_config is not None: code_size_cnt = 0 self.conv_box = nn.ModuleDict() self.conv_box_names = [] num_middle_conv = self.separate_reg_config.NUM_MIDDLE_CONV num_middle_filter = self.separate_reg_config.NUM_MIDDLE_FILTER conv_cls_list = [] c_in = input_channels for k in range(num_middle_conv): conv_cls_list.extend([ nn.Conv2d( c_in, num_middle_filter, kernel_size=3, stride=1, padding=1, bias=False ), nn.BatchNorm2d(num_middle_filter), nn.ReLU() ]) c_in = num_middle_filter conv_cls_list.append(nn.Conv2d( c_in, self.num_anchors_per_location * self.num_class, kernel_size=3, stride=1, padding=1 )) self.conv_cls = nn.Sequential(*conv_cls_list) for reg_config in self.separate_reg_config.REG_LIST: reg_name, reg_channel = reg_config.split(':') reg_channel = int(reg_channel) cur_conv_list = [] c_in = input_channels for k in range(num_middle_conv): cur_conv_list.extend([ nn.Conv2d( c_in, num_middle_filter, kernel_size=3, stride=1, padding=1, bias=False ), nn.BatchNorm2d(num_middle_filter), nn.ReLU() ]) c_in = num_middle_filter cur_conv_list.append(nn.Conv2d( c_in, self.num_anchors_per_location * int(reg_channel), kernel_size=3, stride=1, padding=1, bias=True )) code_size_cnt += reg_channel self.conv_box[f'conv_{reg_name}'] = nn.Sequential(*cur_conv_list) self.conv_box_names.append(f'conv_{reg_name}') for m in self.conv_box.modules(): if isinstance(m, nn.Conv2d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') if m.bias is not None: nn.init.constant_(m.bias, 0) assert code_size_cnt == code_size, f'Code size does not match: {code_size_cnt}:{code_size}' else: self.conv_cls = nn.Conv2d( input_channels, self.num_anchors_per_location * self.num_class, kernel_size=1 ) self.conv_box = nn.Conv2d( input_channels, self.num_anchors_per_location * self.code_size, kernel_size=1 ) if self.model_cfg.get('USE_DIRECTION_CLASSIFIER', None) is not None: self.conv_dir_cls = nn.Conv2d( input_channels, self.num_anchors_per_location * self.model_cfg.NUM_DIR_BINS, kernel_size=1 ) else: self.conv_dir_cls = None self.use_multihead = self.model_cfg.get('USE_MULTIHEAD', False) self.init_weights() def init_weights(self): pi = 0.01 if isinstance(self.conv_cls, nn.Conv2d): nn.init.constant_(self.conv_cls.bias, -np.log((1 - pi) / pi)) else: nn.init.constant_(self.conv_cls[-1].bias, -np.log((1 - pi) / pi)) def forward(self, spatial_features_2d): ret_dict = {} spatial_features_2d = super().forward({'spatial_features': spatial_features_2d})['spatial_features_2d'] cls_preds = self.conv_cls(spatial_features_2d) if self.separate_reg_config is None: box_preds = self.conv_box(spatial_features_2d) else: box_preds_list = [] for reg_name in self.conv_box_names: box_preds_list.append(self.conv_box[reg_name](spatial_features_2d)) box_preds = torch.cat(box_preds_list, dim=1) if not self.use_multihead: box_preds = box_preds.permute(0, 2, 3, 1).contiguous() cls_preds = cls_preds.permute(0, 2, 3, 1).contiguous() else: H, W = box_preds.shape[2:] batch_size = box_preds.shape[0] box_preds = box_preds.view(-1, self.num_anchors_per_location, self.code_size, H, W).permute(0, 1, 3, 4, 2).contiguous() cls_preds = cls_preds.view(-1, self.num_anchors_per_location, self.num_class, H, W).permute(0, 1, 3, 4, 2).contiguous() box_preds = box_preds.view(batch_size, -1, self.code_size) cls_preds = cls_preds.view(batch_size, -1, self.num_class) if self.conv_dir_cls is not None: dir_cls_preds = self.conv_dir_cls(spatial_features_2d) if self.use_multihead: dir_cls_preds = dir_cls_preds.view( -1, self.num_anchors_per_location, self.model_cfg.NUM_DIR_BINS, H, W).permute(0, 1, 3, 4, 2).contiguous() dir_cls_preds = dir_cls_preds.view(batch_size, -1, self.model_cfg.NUM_DIR_BINS) else: dir_cls_preds = dir_cls_preds.permute(0, 2, 3, 1).contiguous() else: dir_cls_preds = None ret_dict['cls_preds'] = cls_preds ret_dict['box_preds'] = box_preds ret_dict['dir_cls_preds'] = dir_cls_preds return ret_dict class AnchorHeadMulti(AnchorHeadTemplate): def __init__(self, model_cfg, input_channels, num_class, class_names, grid_size, point_cloud_range, predict_boxes_when_training=True, **kwargs): super().__init__( model_cfg=model_cfg, num_class=num_class, class_names=class_names, grid_size=grid_size, point_cloud_range=point_cloud_range, predict_boxes_when_training=predict_boxes_when_training ) self.model_cfg = model_cfg self.separate_multihead = self.model_cfg.get('SEPARATE_MULTIHEAD', False) if self.model_cfg.get('SHARED_CONV_NUM_FILTER', None) is not None: shared_conv_num_filter = self.model_cfg.SHARED_CONV_NUM_FILTER self.shared_conv = nn.Sequential( nn.Conv2d(input_channels, shared_conv_num_filter, 3, stride=1, padding=1, bias=False), nn.BatchNorm2d(shared_conv_num_filter, eps=1e-3, momentum=0.01), nn.ReLU(), ) else: self.shared_conv = None shared_conv_num_filter = input_channels self.rpn_heads = None self.make_multihead(shared_conv_num_filter) def make_multihead(self, input_channels): rpn_head_cfgs = self.model_cfg.RPN_HEAD_CFGS rpn_heads = [] class_names = [] for rpn_head_cfg in rpn_head_cfgs: class_names.extend(rpn_head_cfg['HEAD_CLS_NAME']) for rpn_head_cfg in rpn_head_cfgs: num_anchors_per_location = sum([self.num_anchors_per_location[class_names.index(head_cls)] for head_cls in rpn_head_cfg['HEAD_CLS_NAME']]) head_label_indices = torch.from_numpy(np.array([ self.class_names.index(cur_name) + 1 for cur_name in rpn_head_cfg['HEAD_CLS_NAME'] ])) rpn_head = SingleHead( self.model_cfg, input_channels, len(rpn_head_cfg['HEAD_CLS_NAME']) if self.separate_multihead else self.num_class, num_anchors_per_location, self.box_coder.code_size, rpn_head_cfg, head_label_indices=head_label_indices, separate_reg_config=self.model_cfg.get('SEPARATE_REG_CONFIG', None) ) rpn_heads.append(rpn_head) self.rpn_heads = nn.ModuleList(rpn_heads) def forward(self, data_dict): spatial_features_2d = data_dict['spatial_features_2d'] if self.shared_conv is not None: spatial_features_2d = self.shared_conv(spatial_features_2d) ret_dicts = [] for rpn_head in self.rpn_heads: ret_dicts.append(rpn_head(spatial_features_2d)) cls_preds = [ret_dict['cls_preds'] for ret_dict in ret_dicts] box_preds = [ret_dict['box_preds'] for ret_dict in ret_dicts] ret = { 'cls_preds': cls_preds if self.separate_multihead else torch.cat(cls_preds, dim=1), 'box_preds': box_preds if self.separate_multihead else torch.cat(box_preds, dim=1), } if self.model_cfg.get('USE_DIRECTION_CLASSIFIER', False): dir_cls_preds = [ret_dict['dir_cls_preds'] for ret_dict in ret_dicts] ret['dir_cls_preds'] = dir_cls_preds if self.separate_multihead else torch.cat(dir_cls_preds, dim=1) self.forward_ret_dict.update(ret) if self.training: targets_dict = self.assign_targets( gt_boxes=data_dict['gt_boxes'] ) self.forward_ret_dict.update(targets_dict) if not self.training or self.predict_boxes_when_training: batch_cls_preds, batch_box_preds = self.generate_predicted_boxes( batch_size=data_dict['batch_size'], cls_preds=ret['cls_preds'], box_preds=ret['box_preds'], dir_cls_preds=ret.get('dir_cls_preds', None) ) if isinstance(batch_cls_preds, list): multihead_label_mapping = [] for idx in range(len(batch_cls_preds)): multihead_label_mapping.append(self.rpn_heads[idx].head_label_indices) data_dict['multihead_label_mapping'] = multihead_label_mapping data_dict['batch_cls_preds'] = batch_cls_preds data_dict['batch_box_preds'] = batch_box_preds data_dict['cls_preds_normalized'] = False return data_dict def get_cls_layer_loss(self): loss_weights = self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS if 'pos_cls_weight' in loss_weights: pos_cls_weight = loss_weights['pos_cls_weight'] neg_cls_weight = loss_weights['neg_cls_weight'] else: pos_cls_weight = neg_cls_weight = 1.0 cls_preds = self.forward_ret_dict['cls_preds'] box_cls_labels = self.forward_ret_dict['box_cls_labels'] if not isinstance(cls_preds, list): cls_preds = [cls_preds] batch_size = int(cls_preds[0].shape[0]) cared = box_cls_labels >= 0 # [N, num_anchors] positives = box_cls_labels > 0 negatives = box_cls_labels == 0 negative_cls_weights = negatives * 1.0 * neg_cls_weight cls_weights = (negative_cls_weights + pos_cls_weight * positives).float() reg_weights = positives.float() if self.num_class == 1: # class agnostic box_cls_labels[positives] = 1 pos_normalizer = positives.sum(1, keepdim=True).float() reg_weights /= torch.clamp(pos_normalizer, min=1.0) cls_weights /= torch.clamp(pos_normalizer, min=1.0) cls_targets = box_cls_labels * cared.type_as(box_cls_labels) one_hot_targets = torch.zeros( *list(cls_targets.shape), self.num_class + 1, dtype=cls_preds[0].dtype, device=cls_targets.device ) one_hot_targets.scatter_(-1, cls_targets.unsqueeze(dim=-1).long(), 1.0) one_hot_targets = one_hot_targets[..., 1:] start_idx = c_idx = 0 cls_losses = 0 for idx, cls_pred in enumerate(cls_preds): cur_num_class = self.rpn_heads[idx].num_class cls_pred = cls_pred.view(batch_size, -1, cur_num_class) if self.separate_multihead: one_hot_target = one_hot_targets[:, start_idx:start_idx + cls_pred.shape[1], c_idx:c_idx + cur_num_class] c_idx += cur_num_class else: one_hot_target = one_hot_targets[:, start_idx:start_idx + cls_pred.shape[1]] cls_weight = cls_weights[:, start_idx:start_idx + cls_pred.shape[1]] cls_loss_src = self.cls_loss_func(cls_pred, one_hot_target, weights=cls_weight) # [N, M] cls_loss = cls_loss_src.sum() / batch_size cls_loss = cls_loss * loss_weights['cls_weight'] cls_losses += cls_loss start_idx += cls_pred.shape[1] assert start_idx == one_hot_targets.shape[1] tb_dict = { 'rpn_loss_cls': cls_losses.item() } return cls_losses, tb_dict def get_box_reg_layer_loss(self): box_preds = self.forward_ret_dict['box_preds'] box_dir_cls_preds = self.forward_ret_dict.get('dir_cls_preds', None) box_reg_targets = self.forward_ret_dict['box_reg_targets'] box_cls_labels = self.forward_ret_dict['box_cls_labels'] positives = box_cls_labels > 0 reg_weights = positives.float() pos_normalizer = positives.sum(1, keepdim=True).float() reg_weights /= torch.clamp(pos_normalizer, min=1.0) if not isinstance(box_preds, list): box_preds = [box_preds] batch_size = int(box_preds[0].shape[0]) if isinstance(self.anchors, list): if self.use_multihead: anchors = torch.cat( [anchor.permute(3, 4, 0, 1, 2, 5).contiguous().view(-1, anchor.shape[-1]) for anchor in self.anchors], dim=0 ) else: anchors = torch.cat(self.anchors, dim=-3) else: anchors = self.anchors anchors = anchors.view(1, -1, anchors.shape[-1]).repeat(batch_size, 1, 1) start_idx = 0 box_losses = 0 tb_dict = {} for idx, box_pred in enumerate(box_preds): box_pred = box_pred.view( batch_size, -1, box_pred.shape[-1] // self.num_anchors_per_location if not self.use_multihead else box_pred.shape[-1] ) box_reg_target = box_reg_targets[:, start_idx:start_idx + box_pred.shape[1]] reg_weight = reg_weights[:, start_idx:start_idx + box_pred.shape[1]] # sin(a - b) = sinacosb-cosasinb if box_dir_cls_preds is not None: box_pred_sin, reg_target_sin = self.add_sin_difference(box_pred, box_reg_target) loc_loss_src = self.reg_loss_func(box_pred_sin, reg_target_sin, weights=reg_weight) # [N, M] else: loc_loss_src = self.reg_loss_func(box_pred, box_reg_target, weights=reg_weight) # [N, M] loc_loss = loc_loss_src.sum() / batch_size loc_loss = loc_loss * self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS['loc_weight'] box_losses += loc_loss tb_dict['rpn_loss_loc'] = tb_dict.get('rpn_loss_loc', 0) + loc_loss.item() if box_dir_cls_preds is not None: if not isinstance(box_dir_cls_preds, list): box_dir_cls_preds = [box_dir_cls_preds] dir_targets = self.get_direction_target( anchors, box_reg_targets, dir_offset=self.model_cfg.DIR_OFFSET, num_bins=self.model_cfg.NUM_DIR_BINS ) box_dir_cls_pred = box_dir_cls_preds[idx] dir_logit = box_dir_cls_pred.view(batch_size, -1, self.model_cfg.NUM_DIR_BINS) weights = positives.type_as(dir_logit) weights /= torch.clamp(weights.sum(-1, keepdim=True), min=1.0) weight = weights[:, start_idx:start_idx + box_pred.shape[1]] dir_target = dir_targets[:, start_idx:start_idx + box_pred.shape[1]] dir_loss = self.dir_loss_func(dir_logit, dir_target, weights=weight) dir_loss = dir_loss.sum() / batch_size dir_loss = dir_loss * self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS['dir_weight'] box_losses += dir_loss tb_dict['rpn_loss_dir'] = tb_dict.get('rpn_loss_dir', 0) + dir_loss.item() start_idx += box_pred.shape[1] return box_losses, tb_dict
17,041
44.566845
117
py
SASA
SASA-main/pcdet/models/dense_heads/point_head_box.py
import torch from ...utils import box_coder_utils, box_utils from ...utils.loss_utils import PointSASALoss from .point_head_template import PointHeadTemplate class PointHeadBox(PointHeadTemplate): """ A simple point-based segmentation head, which are used for PointRCNN. Reference Paper: https://arxiv.org/abs/1812.04244 PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud """ def __init__(self, num_class, input_channels, model_cfg, predict_boxes_when_training=False, **kwargs): super().__init__(model_cfg=model_cfg, num_class=num_class) self.predict_boxes_when_training = predict_boxes_when_training self.cls_layers = self.make_fc_layers( fc_cfg=self.model_cfg.CLS_FC, input_channels=input_channels, output_channels=num_class ) target_cfg = self.model_cfg.TARGET_CONFIG self.box_coder = getattr(box_coder_utils, target_cfg.BOX_CODER)( **target_cfg.BOX_CODER_CONFIG ) self.box_layers = self.make_fc_layers( fc_cfg=self.model_cfg.REG_FC, input_channels=input_channels, output_channels=self.box_coder.code_size ) # SASA Loss sasa_loss_cfg = model_cfg.LOSS_CONFIG.get('LOSS_SASA_CONFIG', None) if sasa_loss_cfg is not None: self.loss_point_sasa = PointSASALoss(**sasa_loss_cfg) else: self.loss_point_sasa = None def assign_targets(self, input_dict): """ Args: input_dict: point_features: (N1 + N2 + N3 + ..., C) batch_size: point_coords: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] gt_boxes (optional): (B, M, 8) Returns: point_cls_labels: (N1 + N2 + N3 + ...), long type, 0:background, -1:ignored point_part_labels: (N1 + N2 + N3 + ..., 3) """ point_coords = input_dict['point_coords'] gt_boxes = input_dict['gt_boxes'] assert gt_boxes.shape.__len__() == 3, 'gt_boxes.shape=%s' % str(gt_boxes.shape) assert point_coords.shape.__len__() in [2], 'points.shape=%s' % str(point_coords.shape) batch_size = gt_boxes.shape[0] extend_gt_boxes = box_utils.enlarge_box3d( gt_boxes.view(-1, gt_boxes.shape[-1]), extra_width=self.model_cfg.TARGET_CONFIG.GT_EXTRA_WIDTH ).view(batch_size, -1, gt_boxes.shape[-1]) targets_dict = self.assign_stack_targets( points=point_coords, gt_boxes=gt_boxes, extend_gt_boxes=extend_gt_boxes, set_ignore_flag=True, use_ball_constraint=False, ret_part_labels=False, ret_box_labels=True ) return targets_dict def get_sasa_layer_loss(self): point_loss_sasa_list = self.loss_point_sasa.loss_forward( self.forward_ret_dict['point_sasa_preds'], self.forward_ret_dict['point_sasa_labels'] ) point_loss_sasa = 0 tb_dict = dict() for i in range(len(point_loss_sasa_list)): cur_point_loss_sasa = point_loss_sasa_list[i] if cur_point_loss_sasa is None: continue point_loss_sasa = point_loss_sasa + cur_point_loss_sasa tb_dict['point_loss_sasa_layer_%d' % i] = point_loss_sasa_list[i].item() tb_dict['point_loss_sasa'] = point_loss_sasa.item() return point_loss_sasa, tb_dict def get_loss(self, tb_dict=None): tb_dict = {} if tb_dict is None else tb_dict point_loss_cls, tb_dict_1 = self.get_cls_layer_loss() point_loss_box, tb_dict_2 = self.get_box_layer_loss() point_loss = point_loss_cls + point_loss_box tb_dict.update(tb_dict_1) tb_dict.update(tb_dict_2) if self.loss_point_sasa is not None: point_loss_sasa, tb_dict_3 = self.get_sasa_layer_loss() point_loss = point_loss + point_loss_sasa tb_dict.update(tb_dict_3) return point_loss, tb_dict def forward(self, batch_dict): """ Args: batch_dict: batch_size: point_features: (N1 + N2 + N3 + ..., C) or (B, N, C) point_features_before_fusion: (N1 + N2 + N3 + ..., C) point_coords: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] point_labels (optional): (N1 + N2 + N3 + ...) gt_boxes (optional): (B, M, 8) Returns: batch_dict: point_cls_scores: (N1 + N2 + N3 + ..., 1) point_part_offset: (N1 + N2 + N3 + ..., 3) """ if self.model_cfg.get('USE_POINT_FEATURES_BEFORE_FUSION', False): point_features = batch_dict['point_features_before_fusion'] else: point_features = batch_dict['point_features'] point_cls_preds = self.cls_layers(point_features) # (total_points, num_class) point_box_preds = self.box_layers(point_features) # (total_points, box_code_size) point_cls_preds_max, _ = point_cls_preds.max(dim=-1) batch_dict['point_cls_scores'] = torch.sigmoid(point_cls_preds_max) ret_dict = {'point_cls_preds': point_cls_preds, 'point_box_preds': point_box_preds} if self.training: targets_dict = self.assign_targets(batch_dict) ret_dict['point_cls_labels'] = targets_dict['point_cls_labels'] ret_dict['point_box_labels'] = targets_dict['point_box_labels'] if self.loss_point_sasa is not None: point_sasa_labels = self.loss_point_sasa( batch_dict['point_coords_list'], batch_dict['point_scores_list'], batch_dict['gt_boxes'] ) ret_dict.update({ 'point_sasa_preds': batch_dict['point_scores_list'], 'point_sasa_labels': point_sasa_labels }) if not self.training or self.predict_boxes_when_training: point_cls_preds, point_box_preds = self.generate_predicted_boxes( points=batch_dict['point_coords'][:, 1:4], point_cls_preds=point_cls_preds, point_box_preds=point_box_preds ) batch_dict['batch_cls_preds'] = point_cls_preds batch_dict['batch_box_preds'] = point_box_preds batch_dict['batch_index'] = batch_dict['point_coords'][:, 0] batch_dict['cls_preds_normalized'] = False self.forward_ret_dict = ret_dict return batch_dict
6,616
41.146497
106
py
SASA
SASA-main/pcdet/models/dense_heads/point_head_simple.py
import torch from ...utils import box_utils from .point_head_template import PointHeadTemplate class PointHeadSimple(PointHeadTemplate): """ A simple point-based segmentation head, which are used for PV-RCNN keypoint segmentaion. Reference Paper: https://arxiv.org/abs/1912.13192 PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection """ def __init__(self, num_class, input_channels, model_cfg, **kwargs): super().__init__(model_cfg=model_cfg, num_class=num_class) self.cls_layers = self.make_fc_layers( fc_cfg=self.model_cfg.CLS_FC, input_channels=input_channels, output_channels=num_class ) def assign_targets(self, input_dict): """ Args: input_dict: point_features: (N1 + N2 + N3 + ..., C) batch_size: point_coords: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] gt_boxes (optional): (B, M, 8) Returns: point_cls_labels: (N1 + N2 + N3 + ...), long type, 0:background, -1:ignored point_part_labels: (N1 + N2 + N3 + ..., 3) """ point_coords = input_dict['point_coords'] gt_boxes = input_dict['gt_boxes'] assert gt_boxes.shape.__len__() == 3, 'gt_boxes.shape=%s' % str(gt_boxes.shape) assert point_coords.shape.__len__() in [2], 'points.shape=%s' % str(point_coords.shape) batch_size = gt_boxes.shape[0] extend_gt_boxes = box_utils.enlarge_box3d( gt_boxes.view(-1, gt_boxes.shape[-1]), extra_width=self.model_cfg.TARGET_CONFIG.GT_EXTRA_WIDTH ).view(batch_size, -1, gt_boxes.shape[-1]) targets_dict = self.assign_stack_targets( points=point_coords, gt_boxes=gt_boxes, extend_gt_boxes=extend_gt_boxes, set_ignore_flag=True, use_ball_constraint=False, ret_part_labels=False ) return targets_dict def get_loss(self, tb_dict=None): tb_dict = {} if tb_dict is None else tb_dict point_loss_cls, tb_dict_1 = self.get_cls_layer_loss() point_loss = point_loss_cls tb_dict.update(tb_dict_1) return point_loss, tb_dict def forward(self, batch_dict): """ Args: batch_dict: batch_size: point_features: (N1 + N2 + N3 + ..., C) or (B, N, C) point_features_before_fusion: (N1 + N2 + N3 + ..., C) point_coords: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] point_labels (optional): (N1 + N2 + N3 + ...) gt_boxes (optional): (B, M, 8) Returns: batch_dict: point_cls_scores: (N1 + N2 + N3 + ..., 1) point_part_offset: (N1 + N2 + N3 + ..., 3) """ if self.model_cfg.get('USE_POINT_FEATURES_BEFORE_FUSION', False): point_features = batch_dict['point_features_before_fusion'] else: point_features = batch_dict['point_features'] point_cls_preds = self.cls_layers(point_features) # (total_points, num_class) ret_dict = { 'point_cls_preds': point_cls_preds, } point_cls_scores = torch.sigmoid(point_cls_preds) batch_dict['point_cls_scores'], _ = point_cls_scores.max(dim=-1) if self.training: targets_dict = self.assign_targets(batch_dict) ret_dict['point_cls_labels'] = targets_dict['point_cls_labels'] self.forward_ret_dict = ret_dict return batch_dict
3,568
37.793478
106
py
SASA
SASA-main/pcdet/models/dense_heads/__init__.py
from .anchor_head_multi import AnchorHeadMulti from .anchor_head_single import AnchorHeadSingle from .anchor_head_template import AnchorHeadTemplate from .point_head_box import PointHeadBox from .point_head_vote import PointHeadVote from .point_head_simple import PointHeadSimple from .point_intra_part_head import PointIntraPartOffsetHead __all__ = { 'AnchorHeadTemplate': AnchorHeadTemplate, 'AnchorHeadSingle': AnchorHeadSingle, 'PointIntraPartOffsetHead': PointIntraPartOffsetHead, 'PointHeadSimple': PointHeadSimple, 'PointHeadBox': PointHeadBox, 'PointHeadVote': PointHeadVote, 'AnchorHeadMulti': AnchorHeadMulti, }
651
35.222222
59
py
SASA
SASA-main/pcdet/models/dense_heads/point_head_vote.py
import torch import torch.nn as nn import torch.nn.functional as F import numpy as np from ...ops.iou3d_nms import iou3d_nms_utils from ...ops.roiaware_pool3d import roiaware_pool3d_utils from ...ops.pointnet2.pointnet2_batch import pointnet2_modules from ...utils import box_coder_utils, box_utils, common_utils, loss_utils from .point_head_template import PointHeadTemplate class PointHeadVote(PointHeadTemplate): """ A simple vote-based detection head, which is used for 3DSSD. Reference Paper: https://arxiv.org/abs/2002.10187 3DSSD: Point-based 3D Single Stage Object Detector """ def __init__(self, num_class, input_channels, model_cfg, predict_boxes_when_training=False, **kwargs): super().__init__(model_cfg=model_cfg, num_class=num_class) use_bn = self.model_cfg.USE_BN self.predict_boxes_when_training = predict_boxes_when_training self.vote_cfg = self.model_cfg.VOTE_CONFIG self.vote_layers = self.make_fc_layers( input_channels=input_channels, output_channels=3, fc_list=self.vote_cfg.VOTE_FC ) self.sa_cfg = self.model_cfg.SA_CONFIG channel_in, channel_out = input_channels, 0 mlps = self.sa_cfg.MLPS.copy() for idx in range(mlps.__len__()): mlps[idx] = [channel_in] + mlps[idx] channel_out += mlps[idx][-1] self.SA_module = pointnet2_modules.PointnetSAModuleFSMSG( radii=self.sa_cfg.RADIUS, nsamples=self.sa_cfg.NSAMPLE, mlps=mlps, use_xyz=True, bn=use_bn ) channel_in = channel_out shared_fc_list = [] for k in range(0, self.model_cfg.SHARED_FC.__len__()): shared_fc_list.extend([ nn.Conv1d(channel_in, self.model_cfg.SHARED_FC[k], kernel_size=1, bias=False), nn.BatchNorm1d(self.model_cfg.SHARED_FC[k]), nn.ReLU() ]) channel_in = self.model_cfg.SHARED_FC[k] self.shared_fc_layer = nn.Sequential(*shared_fc_list) channel_in = self.model_cfg.SHARED_FC[-1] self.cls_layers = self.make_fc_layers( input_channels=channel_in, output_channels=num_class if not self.model_cfg.LOSS_CONFIG.LOSS_CLS == 'CrossEntropy' else num_class + 1, fc_list=self.model_cfg.CLS_FC ) target_cfg = self.model_cfg.TARGET_CONFIG self.box_coder = getattr(box_coder_utils, target_cfg.BOX_CODER)( **target_cfg.BOX_CODER_CONFIG ) self.reg_layers = self.make_fc_layers( input_channels=channel_in, output_channels=self.box_coder.code_size, fc_list=self.model_cfg.REG_FC ) self.init_weights(weight_init='xavier') def init_weights(self, weight_init='xavier'): if weight_init == 'kaiming': init_func = nn.init.kaiming_normal_ elif weight_init == 'xavier': init_func = nn.init.xavier_normal_ elif weight_init == 'normal': init_func = nn.init.normal_ else: raise NotImplementedError for m in self.modules(): if isinstance(m, nn.Conv2d) or isinstance(m, nn.Conv1d): if weight_init == 'normal': init_func(m.weight, mean=0, std=0.001) else: init_func(m.weight) if m.bias is not None: nn.init.constant_(m.bias, 0) def build_losses(self, losses_cfg): # classification loss if losses_cfg.LOSS_CLS.startswith('WeightedBinaryCrossEntropy'): self.add_module( 'cls_loss_func', loss_utils.WeightedBinaryCrossEntropyLoss() ) elif losses_cfg.LOSS_CLS == 'WeightedCrossEntropy': self.add_module( 'cls_loss_func', loss_utils.WeightedCrossEntropyLoss() ) elif losses_cfg.LOSS_CLS == 'FocalLoss': self.add_module( 'cls_loss_func', loss_utils.SigmoidFocalClassificationLoss( **losses_cfg.get('LOSS_CLS_CONFIG', {}) ) ) else: raise NotImplementedError # regression loss if losses_cfg.LOSS_REG == 'WeightedSmoothL1Loss': self.add_module( 'reg_loss_func', loss_utils.WeightedSmoothL1Loss( code_weights=losses_cfg.LOSS_WEIGHTS.get('code_weights', None), **losses_cfg.get('LOSS_REG_CONFIG', {}) ) ) elif losses_cfg.LOSS_REG == 'WeightedL1Loss': self.add_module( 'reg_loss_func', loss_utils.WeightedL1Loss( code_weights=losses_cfg.LOSS_WEIGHTS.get('code_weights', None) ) ) else: raise NotImplementedError # sasa loss loss_sasa_cfg = losses_cfg.get('LOSS_SASA_CONFIG', None) if loss_sasa_cfg is not None: self.enable_sasa = True self.add_module( 'loss_point_sasa', loss_utils.PointSASALoss(**loss_sasa_cfg) ) else: self.enable_sasa = False def make_fc_layers(self, input_channels, output_channels, fc_list): fc_layers = [] pre_channel = input_channels for k in range(0, fc_list.__len__()): fc_layers.extend([ nn.Conv1d(pre_channel, fc_list[k], kernel_size=1, bias=False), nn.BatchNorm1d(fc_list[k]), nn.ReLU() ]) pre_channel = fc_list[k] fc_layers.append(nn.Conv1d(pre_channel, output_channels, kernel_size=1, bias=True)) fc_layers = nn.Sequential(*fc_layers) return fc_layers def assign_stack_targets_simple(self, points, gt_boxes, extend_gt_boxes=None, set_ignore_flag=True): """ Args: points: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] gt_boxes: (B, M, 8) extend_gt_boxes: (B, M, 8), required if set ignore flag set_ignore_flag: Returns: point_cls_labels: (N1 + N2 + N3 + ...), long type, 0:background, -1:ignore point_reg_labels: (N1 + N2 + N3 + ..., 3), corresponding object centroid """ assert len(points.shape) == 2 and points.shape[1] == 4, 'points.shape=%s' % str(points.shape) assert len(gt_boxes.shape) == 3, 'gt_boxes.shape=%s' % str(gt_boxes.shape) assert extend_gt_boxes is None or len(extend_gt_boxes.shape) == 3, \ 'extend_gt_boxes.shape=%s' % str(extend_gt_boxes.shape) assert not set_ignore_flag or extend_gt_boxes is not None batch_size = gt_boxes.shape[0] bs_idx = points[:, 0] point_cls_labels = points.new_zeros(points.shape[0]).long() point_reg_labels = gt_boxes.new_zeros((points.shape[0], 3)) for k in range(batch_size): bs_mask = (bs_idx == k) points_single = points[bs_mask][:, 1:4] point_cls_labels_single = point_cls_labels.new_zeros(bs_mask.sum()) box_idxs_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( points_single.unsqueeze(dim=0), gt_boxes[k:k + 1, :, 0:7].contiguous() ).long().squeeze(dim=0) box_fg_flag = (box_idxs_of_pts >= 0) if extend_gt_boxes is not None: extend_box_idx_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( points_single.unsqueeze(dim=0), extend_gt_boxes[k:k + 1, :, 0:7].contiguous() ).long().squeeze(dim=0) fg_flag = box_fg_flag ignore_flag = fg_flag ^ (extend_box_idx_of_pts >= 0) point_cls_labels_single[ignore_flag] = -1 gt_box_of_fg_points = gt_boxes[k][box_idxs_of_pts[box_fg_flag]] point_cls_labels_single[box_fg_flag] = 1 point_cls_labels[bs_mask] = point_cls_labels_single point_reg_labels_single = point_reg_labels.new_zeros((bs_mask.sum(), 3)) point_reg_labels_single[box_fg_flag] = gt_box_of_fg_points[:, 0:3] point_reg_labels[bs_mask] = point_reg_labels_single targets_dict = { 'point_cls_labels': point_cls_labels, 'point_reg_labels': point_reg_labels, } return targets_dict def assign_targets_simple(self, points, gt_boxes, extra_width=None, set_ignore_flag=True): """ Args: points: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] gt_boxes: (B, M, 8) extra_width: (dx, dy, dz) extra width applied to gt boxes assign_method: binary or distance set_ignore_flag: Returns: point_vote_labels: (N1 + N2 + N3 + ..., 3) """ assert gt_boxes.shape.__len__() == 3, 'gt_boxes.shape=%s' % str(gt_boxes.shape) assert points.shape.__len__() in [2], 'points.shape=%s' % str(points.shape) batch_size = gt_boxes.shape[0] extend_gt_boxes = box_utils.enlarge_box3d( gt_boxes.view(-1, gt_boxes.shape[-1]), extra_width=extra_width ).view(batch_size, -1, gt_boxes.shape[-1]) \ if extra_width is not None else gt_boxes if set_ignore_flag: targets_dict = self.assign_stack_targets_simple(points=points, gt_boxes=gt_boxes, extend_gt_boxes=extend_gt_boxes, set_ignore_flag=set_ignore_flag) else: targets_dict = self.assign_stack_targets_simple(points=points, gt_boxes=extend_gt_boxes, set_ignore_flag=set_ignore_flag) return targets_dict def assign_stack_targets_mask(self, points, gt_boxes, extend_gt_boxes=None, set_ignore_flag=True, use_ball_constraint=False, central_radius=2.0): """ Args: points: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] gt_boxes: (B, M, 8) extend_gt_boxes: [B, M, 8] set_ignore_flag: use_ball_constraint: central_radius: Returns: point_cls_labels: (N1 + N2 + N3 + ...), long type, 0:background, -1:ignored point_reg_labels: (N1 + N2 + N3 + ..., code_size) point_box_labels: (N1 + N2 + N3 + ..., 7) """ assert len(points.shape) == 2 and points.shape[1] == 4, 'points.shape=%s' % str(points.shape) assert len(gt_boxes.shape) == 3, 'gt_boxes.shape=%s' % str(gt_boxes.shape) assert extend_gt_boxes is None or len(extend_gt_boxes.shape) == 3, \ 'extend_gt_boxes.shape=%s' % str(extend_gt_boxes.shape) assert set_ignore_flag != use_ball_constraint, 'Choose one only!' batch_size = gt_boxes.shape[0] bs_idx = points[:, 0] point_cls_labels = gt_boxes.new_zeros(points.shape[0]).long() point_reg_labels = gt_boxes.new_zeros((points.shape[0], self.box_coder.code_size)) point_box_labels = gt_boxes.new_zeros((points.shape[0], gt_boxes.size(2) - 1)) for k in range(batch_size): bs_mask = (bs_idx == k) points_single = points[bs_mask][:, 1:4] point_cls_labels_single = point_cls_labels.new_zeros(bs_mask.sum()) box_idxs_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( points_single.unsqueeze(dim=0), gt_boxes[k:k + 1, :, 0:7].contiguous() ).long().squeeze(dim=0) box_fg_flag = (box_idxs_of_pts >= 0) if set_ignore_flag: extend_box_idxs_of_pts = roiaware_pool3d_utils.points_in_boxes_gpu( points_single.unsqueeze(dim=0), extend_gt_boxes[k:k+1, :, 0:7].contiguous() ).long().squeeze(dim=0) fg_flag = box_fg_flag ignore_flag = fg_flag ^ (extend_box_idxs_of_pts >= 0) point_cls_labels_single[ignore_flag] = -1 elif use_ball_constraint: box_centers = gt_boxes[k][box_idxs_of_pts][:, 0:3].clone() ball_flag = ((box_centers - points_single).norm(dim=1) < central_radius) fg_flag = box_fg_flag & ball_flag ignore_flag = fg_flag ^ box_fg_flag point_cls_labels_single[ignore_flag] = -1 else: raise NotImplementedError gt_box_of_fg_points = gt_boxes[k][box_idxs_of_pts[fg_flag]] point_cls_labels_single[fg_flag] = 1 if self.num_class == 1 else gt_box_of_fg_points[:, -1].long() point_cls_labels[bs_mask] = point_cls_labels_single if gt_box_of_fg_points.shape[0] > 0: point_reg_labels_single = point_reg_labels.new_zeros((bs_mask.sum(), self.box_coder.code_size)) fg_point_box_labels = self.box_coder.encode_torch( gt_boxes=gt_box_of_fg_points[:, :-1], points=points_single[fg_flag], gt_classes=gt_box_of_fg_points[:, -1].long() ) point_reg_labels_single[fg_flag] = fg_point_box_labels point_reg_labels[bs_mask] = point_reg_labels_single point_box_labels_single = point_box_labels.new_zeros((bs_mask.sum(), gt_boxes.size(2) - 1)) point_box_labels_single[fg_flag] = gt_box_of_fg_points[:, :-1] point_box_labels[bs_mask] = point_box_labels_single targets_dict = { 'point_cls_labels': point_cls_labels, 'point_reg_labels': point_reg_labels, 'point_box_labels': point_box_labels } return targets_dict def assign_stack_targets_iou(self, points, pred_boxes, gt_boxes, pos_iou_threshold=0.5, neg_iou_threshold=0.35): """ Args: points: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] pred_boxes: (N, 7/8) gt_boxes: (B, M, 8) pos_iou_threshold: neg_iou_threshold: Returns: point_cls_labels: (N1 + N2 + N3 + ...), long type, 0:background, -1:ignored point_reg_labels: (N1 + N2 + N3 + ..., code_size) point_box_labels: (N1 + N2 + N3 + ..., 7) """ assert len(points.shape) == 2 and points.shape[1] == 4, 'points.shape=%s' % str(points.shape) assert len(pred_boxes.shape) == 2 and pred_boxes.shape[1] >= 7, 'pred_boxes.shape=%s' % str(pred_boxes.shape) assert len(gt_boxes.shape) == 3 and gt_boxes.shape[2] == 8, 'gt_boxes.shape=%s' % str(gt_boxes.shape) batch_size = gt_boxes.shape[0] bs_idx = points[:, 0] point_cls_labels = gt_boxes.new_zeros(pred_boxes.shape[0]).long() point_reg_labels = gt_boxes.new_zeros((pred_boxes.shape[0], self.box_coder.code_size)) point_box_labels = gt_boxes.new_zeros((pred_boxes.shape[0], 7)) for k in range(batch_size): bs_mask = (bs_idx == k) points_single = points[bs_mask][:, 1:4] pred_boxes_single = pred_boxes[bs_mask] point_cls_labels_single = point_cls_labels.new_zeros(bs_mask.sum()) pred_boxes_iou = iou3d_nms_utils.boxes_iou3d_gpu( pred_boxes_single, gt_boxes[k][:, :7] ) pred_boxes_iou, box_idxs_of_pts = torch.max(pred_boxes_iou, dim=-1) fg_flag = pred_boxes_iou > pos_iou_threshold ignore_flag = (pred_boxes_iou > neg_iou_threshold) ^ fg_flag gt_box_of_fg_points = gt_boxes[k][box_idxs_of_pts[fg_flag]] point_cls_labels_single[fg_flag] = 1 if self.num_class == 1 else gt_box_of_fg_points[:, -1].long() point_cls_labels_single[ignore_flag] = -1 point_cls_labels[bs_mask] = point_cls_labels_single if gt_box_of_fg_points.shape[0] > 0: point_reg_labels_single = point_reg_labels.new_zeros((bs_mask.sum(), self.box_coder.code_size)) fg_point_box_labels = self.box_coder.encode_torch( gt_boxes=gt_box_of_fg_points[:, :-1], points=points_single[fg_flag], gt_classes=gt_box_of_fg_points[:, -1].long() ) point_reg_labels_single[fg_flag] = fg_point_box_labels point_reg_labels[bs_mask] = point_reg_labels_single point_box_labels_single = point_box_labels.new_zeros((bs_mask.sum(), 7)) point_box_labels_single[fg_flag] = gt_box_of_fg_points[:, :-1] point_box_labels[bs_mask] = point_box_labels_single targets_dict = { 'point_cls_labels': point_cls_labels, 'point_reg_labels': point_reg_labels, 'point_box_labels': point_box_labels } return targets_dict def assign_targets(self, input_dict): """ Args: input_dict: batch_size: point_coords: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] gt_boxes (optional): (B, M, 8) Returns: point_part_labels: (N1 + N2 + N3 + ..., 3) """ assign_method = self.model_cfg.TARGET_CONFIG.ASSIGN_METHOD # mask or iou if assign_method == 'mask': points = input_dict['point_vote_coords'] gt_boxes = input_dict['gt_boxes'] assert points.shape.__len__() == 2, 'points.shape=%s' % str(points.shape) assert gt_boxes.shape.__len__() == 3, 'gt_boxes.shape=%s' % str(gt_boxes.shape) central_radius = self.model_cfg.TARGET_CONFIG.get('GT_CENTRAL_RADIUS', 2.0) targets_dict = self.assign_stack_targets_mask( points=points, gt_boxes=gt_boxes, set_ignore_flag=False, use_ball_constraint=True, central_radius=central_radius ) elif assign_method == 'iou': points = input_dict['point_vote_coords'] pred_boxes = input_dict['point_box_preds'] gt_boxes = input_dict['gt_boxes'] assert points.shape.__len__() == 2, 'points.shape=%s' % str(points.shape) assert gt_boxes.shape.__len__() == 3, 'gt_boxes.shape=%s' % str(gt_boxes.shape) assert pred_boxes.shape.__len__() == 2, 'pred_boxes.shape=%s' % str(pred_boxes.shape) pos_iou_threshold = self.model_cfg.TARGET_CONFIG.POS_IOU_THRESHOLD neg_iou_threshold = self.model_cfg.TARGET_CONFIG.NEG_IOU_THRESHOLD targets_dict = self.assign_stack_targets_iou( points=points, pred_boxes=pred_boxes, gt_boxes=gt_boxes, pos_iou_threshold=pos_iou_threshold, neg_iou_threshold=neg_iou_threshold ) else: raise NotImplementedError return targets_dict def get_vote_layer_loss(self, tb_dict=None): pos_mask = self.forward_ret_dict['vote_cls_labels'] > 0 vote_reg_labels = self.forward_ret_dict['vote_reg_labels'] vote_reg_preds = self.forward_ret_dict['point_vote_coords'] reg_weights = pos_mask.float() pos_normalizer = pos_mask.sum().float() reg_weights /= torch.clamp(pos_normalizer, min=1.0) vote_loss_reg_src = self.reg_loss_func( vote_reg_preds[None, ...], vote_reg_labels[None, ...], weights=reg_weights[None, ...]) vote_loss_reg = vote_loss_reg_src.sum() loss_weights_dict = self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS vote_loss_reg = vote_loss_reg * loss_weights_dict['vote_reg_weight'] if tb_dict is None: tb_dict = {} tb_dict.update({'vote_loss_reg': vote_loss_reg.item()}) return vote_loss_reg, tb_dict @torch.no_grad() def generate_centerness_label(self, point_base, point_box_labels, pos_mask, epsilon=1e-6): """ Args: point_base: (N1 + N2 + N3 + ..., 3) point_box_labels: (N1 + N2 + N3 + ..., 7) pos_mask: (N1 + N2 + N3 + ...) epsilon: Returns: centerness_label: (N1 + N2 + N3 + ...) """ centerness = point_box_labels.new_zeros(pos_mask.shape) point_box_labels = point_box_labels[pos_mask, :] canonical_xyz = point_base[pos_mask, :] - point_box_labels[:, :3] rys = point_box_labels[:, -1] canonical_xyz = common_utils.rotate_points_along_z( canonical_xyz.unsqueeze(dim=1), -rys ).squeeze(dim=1) distance_front = point_box_labels[:, 3] / 2 - canonical_xyz[:, 0] distance_back = point_box_labels[:, 3] / 2 + canonical_xyz[:, 0] distance_left = point_box_labels[:, 4] / 2 - canonical_xyz[:, 1] distance_right = point_box_labels[:, 4] / 2 + canonical_xyz[:, 1] distance_top = point_box_labels[:, 5] / 2 - canonical_xyz[:, 2] distance_bottom = point_box_labels[:, 5] / 2 + canonical_xyz[:, 2] centerness_l = torch.min(distance_front, distance_back) / torch.max(distance_front, distance_back) centerness_w = torch.min(distance_left, distance_right) / torch.max(distance_left, distance_right) centerness_h = torch.min(distance_top, distance_bottom) / torch.max(distance_top, distance_bottom) centerness_pos = torch.clamp(centerness_l * centerness_w * centerness_h, min=epsilon) ** (1 / 3.0) centerness[pos_mask] = centerness_pos return centerness def get_axis_aligned_iou_loss_lidar(self, pred_boxes: torch.Tensor, gt_boxes: torch.Tensor): """ Args: pred_boxes: (N, 7) float Tensor. gt_boxes: (N, 7) float Tensor. Returns: iou_loss: (N) float Tensor. """ assert pred_boxes.shape[0] == gt_boxes.shape[0] pos_p, len_p, *cps = torch.split(pred_boxes, 3, dim=-1) pos_g, len_g, *cgs = torch.split(gt_boxes, 3, dim=-1) len_p = torch.clamp(len_p, min=1e-5) len_g = torch.clamp(len_g, min=1e-5) vol_p = len_p.prod(dim=-1) vol_g = len_g.prod(dim=-1) min_p, max_p = pos_p - len_p / 2, pos_p + len_p / 2 min_g, max_g = pos_g - len_g / 2, pos_g + len_g / 2 min_max = torch.min(max_p, max_g) max_min = torch.max(min_p, min_g) diff = torch.clamp(min_max - max_min, min=0) intersection = diff.prod(dim=-1) union = vol_p + vol_g - intersection iou_axis_aligned = intersection / torch.clamp(union, min=1e-5) iou_loss = 1 - iou_axis_aligned return iou_loss def get_corner_loss_lidar(self, pred_boxes: torch.Tensor, gt_boxes: torch.Tensor): """ Args: pred_boxes: (N, 7) float Tensor. gt_boxes: (N, 7) float Tensor. Returns: corner_loss: (N) float Tensor. """ assert pred_boxes.shape[0] == gt_boxes.shape[0] pred_box_corners = box_utils.boxes_to_corners_3d(pred_boxes) gt_box_corners = box_utils.boxes_to_corners_3d(gt_boxes) gt_boxes_flip = gt_boxes.clone() gt_boxes_flip[:, 6] += np.pi gt_box_corners_flip = box_utils.boxes_to_corners_3d(gt_boxes_flip) # (N, 8, 3) corner_loss = loss_utils.WeightedSmoothL1Loss.smooth_l1_loss(pred_box_corners - gt_box_corners, 1.0) corner_loss_flip = loss_utils.WeightedSmoothL1Loss.smooth_l1_loss(pred_box_corners - gt_box_corners_flip, 1.0) corner_loss = torch.min(corner_loss.sum(dim=2), corner_loss_flip.sum(dim=2)) return corner_loss.mean(dim=1) def get_cls_layer_loss(self, tb_dict=None): point_cls_labels = self.forward_ret_dict['point_cls_labels'].view(-1) point_cls_preds = self.forward_ret_dict['point_cls_preds'].view(-1, self.num_class) positives = point_cls_labels > 0 negatives = point_cls_labels == 0 cls_weights = positives * 1.0 + negatives * 1.0 one_hot_targets = point_cls_preds.new_zeros(*list(point_cls_labels.shape), self.num_class + 1) one_hot_targets.scatter_(-1, (point_cls_labels * (point_cls_labels >= 0).long()).unsqueeze(dim=-1).long(), 1.0) self.forward_ret_dict['point_cls_labels_onehot'] = one_hot_targets loss_cfgs = self.model_cfg.LOSS_CONFIG if 'WithCenterness' in loss_cfgs.LOSS_CLS: point_base = self.forward_ret_dict['point_vote_coords'] point_box_labels = self.forward_ret_dict['point_box_labels'] centerness_label = self.generate_centerness_label(point_base, point_box_labels, positives) loss_cls_cfg = loss_cfgs.get('LOSS_CLS_CONFIG', None) centerness_min = loss_cls_cfg['centerness_min'] if loss_cls_cfg is not None else 0.0 centerness_max = loss_cls_cfg['centerness_max'] if loss_cls_cfg is not None else 1.0 centerness_label = centerness_min + (centerness_max - centerness_min) * centerness_label one_hot_targets *= centerness_label.unsqueeze(dim=-1) point_loss_cls = self.cls_loss_func(point_cls_preds, one_hot_targets[..., 1:], weights=cls_weights) loss_weights_dict = self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS point_loss_cls = point_loss_cls * loss_weights_dict['point_cls_weight'] if tb_dict is None: tb_dict = {} tb_dict.update({ 'point_pos_num': positives.sum().item() }) return point_loss_cls, cls_weights, tb_dict # point_loss_cls: (N) def get_box_layer_loss(self, tb_dict=None): pos_mask = self.forward_ret_dict['point_cls_labels'] > 0 point_reg_preds = self.forward_ret_dict['point_reg_preds'] point_reg_labels = self.forward_ret_dict['point_reg_labels'] reg_weights = pos_mask.float() loss_weights_dict = self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS if tb_dict is None: tb_dict = {} point_loss_offset_reg = self.reg_loss_func( point_reg_preds[None, :, :6], point_reg_labels[None, :, :6], weights=reg_weights[None, ...] ) point_loss_offset_reg = point_loss_offset_reg.sum(dim=-1).squeeze() if hasattr(self.box_coder, 'pred_velo') and self.box_coder.pred_velo: point_loss_velo_reg = self.reg_loss_func( point_reg_preds[None, :, 6 + 2 * self.box_coder.angle_bin_num:8 + 2 * self.box_coder.angle_bin_num], point_reg_labels[None, :, 6 + 2 * self.box_coder.angle_bin_num:8 + 2 * self.box_coder.angle_bin_num], weights=reg_weights[None, ...] ) point_loss_velo_reg = point_loss_velo_reg.sum(dim=-1).squeeze() point_loss_offset_reg = point_loss_offset_reg + point_loss_velo_reg point_loss_offset_reg *= loss_weights_dict['point_offset_reg_weight'] if isinstance(self.box_coder, box_coder_utils.PointBinResidualCoder): point_angle_cls_labels = \ point_reg_labels[:, 6:6 + self.box_coder.angle_bin_num] point_loss_angle_cls = F.cross_entropy( # angle bin cls point_reg_preds[:, 6:6 + self.box_coder.angle_bin_num], point_angle_cls_labels.argmax(dim=-1), reduction='none') * reg_weights point_angle_reg_preds = point_reg_preds[:, 6 + self.box_coder.angle_bin_num:6 + 2 * self.box_coder.angle_bin_num] point_angle_reg_labels = point_reg_labels[:, 6 + self.box_coder.angle_bin_num:6 + 2 * self.box_coder.angle_bin_num] point_angle_reg_preds = (point_angle_reg_preds * point_angle_cls_labels).sum(dim=-1, keepdim=True) point_angle_reg_labels = (point_angle_reg_labels * point_angle_cls_labels).sum(dim=-1, keepdim=True) point_loss_angle_reg = self.reg_loss_func( point_angle_reg_preds[None, ...], point_angle_reg_labels[None, ...], weights=reg_weights[None, ...] ) point_loss_angle_reg = point_loss_angle_reg.squeeze() point_loss_angle_cls *= loss_weights_dict['point_angle_cls_weight'] point_loss_angle_reg *= loss_weights_dict['point_angle_reg_weight'] point_loss_box = point_loss_offset_reg + point_loss_angle_cls + point_loss_angle_reg # (N) else: point_angle_reg_preds = point_reg_preds[:, 6:] point_angle_reg_labels = point_reg_labels[:, 6:] point_loss_angle_reg = self.reg_loss_func( point_angle_reg_preds[None, ...], point_angle_reg_labels[None, ...], weights=reg_weights[None, ...] ) point_loss_angle_reg *= loss_weights_dict['point_angle_reg_weight'] point_loss_box = point_loss_offset_reg + point_loss_angle_reg if reg_weights.sum() > 0: point_box_preds = self.forward_ret_dict['point_box_preds'] point_box_labels = self.forward_ret_dict['point_box_labels'] point_loss_box_aux = 0 if self.model_cfg.LOSS_CONFIG.get('AXIS_ALIGNED_IOU_LOSS_REGULARIZATION', False): point_loss_iou = self.get_axis_aligned_iou_loss_lidar( point_box_preds[pos_mask, :], point_box_labels[pos_mask, :] ) point_loss_iou *= self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS['point_iou_weight'] point_loss_box_aux = point_loss_box_aux + point_loss_iou if self.model_cfg.LOSS_CONFIG.get('CORNER_LOSS_REGULARIZATION', False): point_loss_corner = self.get_corner_loss_lidar( point_box_preds[pos_mask, 0:7], point_box_labels[pos_mask, 0:7] ) point_loss_corner *= self.model_cfg.LOSS_CONFIG.LOSS_WEIGHTS['point_corner_weight'] point_loss_box_aux = point_loss_box_aux + point_loss_corner point_loss_box[pos_mask] = point_loss_box[pos_mask] + point_loss_box_aux return point_loss_box, reg_weights, tb_dict # point_loss_box: (N) def get_sasa_layer_loss(self, tb_dict=None): if self.enable_sasa: point_loss_sasa_list = self.loss_point_sasa.loss_forward( self.forward_ret_dict['point_sasa_preds'], self.forward_ret_dict['point_sasa_labels'] ) point_loss_sasa = 0 tb_dict = dict() for i in range(len(point_loss_sasa_list)): cur_point_loss_sasa = point_loss_sasa_list[i] if cur_point_loss_sasa is None: continue point_loss_sasa = point_loss_sasa + cur_point_loss_sasa tb_dict['point_loss_sasa_layer_%d' % i] = point_loss_sasa_list[i].item() tb_dict['point_loss_sasa'] = point_loss_sasa.item() return point_loss_sasa, tb_dict else: return None, None def get_loss(self, tb_dict=None): tb_dict = {} if tb_dict is None else tb_dict point_loss_vote, tb_dict_0 = self.get_vote_layer_loss() point_loss_cls, cls_weights, tb_dict_1 = self.get_cls_layer_loss() point_loss_box, box_weights, tb_dict_2 = self.get_box_layer_loss() point_loss_cls = point_loss_cls.sum() / torch.clamp(cls_weights.sum(), min=1.0) point_loss_box = point_loss_box.sum() / torch.clamp(box_weights.sum(), min=1.0) tb_dict.update({ 'point_loss_vote': point_loss_vote.item(), 'point_loss_cls': point_loss_cls.item(), 'point_loss_box': point_loss_box.item() }) point_loss = point_loss_vote + point_loss_cls + point_loss_box tb_dict.update(tb_dict_0) tb_dict.update(tb_dict_1) tb_dict.update(tb_dict_2) point_loss_sasa, tb_dict_3 = self.get_sasa_layer_loss() if point_loss_sasa is not None: tb_dict.update(tb_dict_3) point_loss += point_loss_sasa return point_loss, tb_dict def forward(self, batch_dict): """ Args: batch_dict: batch_size: point_features: (N1 + N2 + N3 + ..., C) point_coords: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] point_scores (optional): (B, N) gt_boxes (optional): (B, M, 8) Returns: batch_dict: point_cls_scores: (N1 + N2 + N3 + ..., 1) point_part_offset: (N1 + N2 + N3 + ..., 3) """ batch_size = batch_dict['batch_size'] point_coords = batch_dict['point_coords'] point_features = batch_dict['point_features'] batch_idx, point_coords = point_coords[:, 0], point_coords[:, 1:4] batch_idx = batch_idx.view(batch_size, -1, 1) point_coords = point_coords.view(batch_size, -1, 3).contiguous() point_features = point_features.reshape( batch_size, point_coords.size(1), -1 ).permute(0, 2, 1).contiguous() # candidate points sampling sample_range = self.model_cfg.SAMPLE_RANGE sample_batch_idx = batch_idx[:, sample_range[0]:sample_range[1], :].contiguous() candidate_coords = point_coords[:, sample_range[0]:sample_range[1], :].contiguous() candidate_features = point_features[:, :, sample_range[0]:sample_range[1]].contiguous() # generate vote points vote_offsets = self.vote_layers(candidate_features) # (B, 3, N) vote_translation_range = np.array(self.vote_cfg.MAX_TRANSLATION_RANGE, dtype=np.float32) vote_translation_range = torch.from_numpy(vote_translation_range).cuda().unsqueeze(dim=0).unsqueeze(dim=-1) vote_offsets = torch.max(vote_offsets, -vote_translation_range) vote_offsets = torch.min(vote_offsets, vote_translation_range) vote_coords = candidate_coords + vote_offsets.permute(0, 2, 1).contiguous() ret_dict = {'batch_size': batch_size, 'point_candidate_coords': candidate_coords.view(-1, 3).contiguous(), 'point_vote_coords': vote_coords.view(-1, 3).contiguous()} sample_batch_idx_flatten = sample_batch_idx.view(-1, 1).contiguous() # (N, 1) batch_dict['batch_index'] = sample_batch_idx_flatten.squeeze(-1) batch_dict['point_candidate_coords'] = torch.cat( # (N, 4) (sample_batch_idx_flatten, ret_dict['point_candidate_coords']), dim=-1) batch_dict['point_vote_coords'] = torch.cat( # (N, 4) (sample_batch_idx_flatten, ret_dict['point_vote_coords']), dim=-1) if self.training: # assign targets for vote loss extra_width = self.model_cfg.TARGET_CONFIG.get('VOTE_EXTRA_WIDTH', None) targets_dict = self.assign_targets_simple(batch_dict['point_candidate_coords'], batch_dict['gt_boxes'], extra_width=extra_width, set_ignore_flag=False) ret_dict['vote_cls_labels'] = targets_dict['point_cls_labels'] # (N) ret_dict['vote_reg_labels'] = targets_dict['point_reg_labels'] # (N, 3) _, point_features, _ = self.SA_module( point_coords, point_features, new_xyz=vote_coords ) point_features = self.shared_fc_layer(point_features) point_cls_preds = self.cls_layers(point_features) point_reg_preds = self.reg_layers(point_features) point_cls_preds = point_cls_preds.permute(0, 2, 1).contiguous() point_cls_preds = point_cls_preds.view(-1, point_cls_preds.shape[-1]).contiguous() point_reg_preds = point_reg_preds.permute(0, 2, 1).contiguous() point_reg_preds = point_reg_preds.view(-1, point_reg_preds.shape[-1]).contiguous() point_cls_scores = torch.sigmoid(point_cls_preds) batch_dict['point_cls_scores'] = point_cls_scores point_box_preds = self.box_coder.decode_torch(point_reg_preds, ret_dict['point_vote_coords']) batch_dict['point_box_preds'] = point_box_preds ret_dict.update({'point_cls_preds': point_cls_preds, 'point_reg_preds': point_reg_preds, 'point_box_preds': point_box_preds, 'point_cls_scores': point_cls_scores}) if self.training: targets_dict = self.assign_targets(batch_dict) ret_dict['point_cls_labels'] = targets_dict['point_cls_labels'] ret_dict['point_reg_labels'] = targets_dict['point_reg_labels'] ret_dict['point_box_labels'] = targets_dict['point_box_labels'] if self.enable_sasa: point_sasa_labels = self.loss_point_sasa( batch_dict['point_coords_list'], batch_dict['point_scores_list'], batch_dict['gt_boxes'] ) ret_dict.update({ 'point_sasa_preds': batch_dict['point_scores_list'], 'point_sasa_labels': point_sasa_labels }) if not self.training or self.predict_boxes_when_training: point_cls_preds, point_box_preds = self.generate_predicted_boxes( points=batch_dict['point_vote_coords'][:, 1:4], point_cls_preds=point_cls_preds, point_box_preds=point_reg_preds ) batch_dict['batch_cls_preds'] = point_cls_preds batch_dict['batch_box_preds'] = point_box_preds batch_dict['cls_preds_normalized'] = False self.forward_ret_dict = ret_dict return batch_dict
37,917
45.754624
127
py
SASA
SASA-main/pcdet/models/dense_heads/point_intra_part_head.py
import torch from ...utils import box_coder_utils, box_utils from .point_head_template import PointHeadTemplate class PointIntraPartOffsetHead(PointHeadTemplate): """ Point-based head for predicting the intra-object part locations. Reference Paper: https://arxiv.org/abs/1907.03670 From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Network """ def __init__(self, num_class, input_channels, model_cfg, predict_boxes_when_training=False, **kwargs): super().__init__(model_cfg=model_cfg, num_class=num_class) self.predict_boxes_when_training = predict_boxes_when_training self.cls_layers = self.make_fc_layers( fc_cfg=self.model_cfg.CLS_FC, input_channels=input_channels, output_channels=num_class ) self.part_reg_layers = self.make_fc_layers( fc_cfg=self.model_cfg.PART_FC, input_channels=input_channels, output_channels=3 ) target_cfg = self.model_cfg.TARGET_CONFIG if target_cfg.get('BOX_CODER', None) is not None: self.box_coder = getattr(box_coder_utils, target_cfg.BOX_CODER)( **target_cfg.BOX_CODER_CONFIG ) self.box_layers = self.make_fc_layers( fc_cfg=self.model_cfg.REG_FC, input_channels=input_channels, output_channels=self.box_coder.code_size ) else: self.box_layers = None def assign_targets(self, input_dict): """ Args: input_dict: point_features: (N1 + N2 + N3 + ..., C) batch_size: point_coords: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] gt_boxes (optional): (B, M, 8) Returns: point_cls_labels: (N1 + N2 + N3 + ...), long type, 0:background, -1:ignored point_part_labels: (N1 + N2 + N3 + ..., 3) """ point_coords = input_dict['point_coords'] gt_boxes = input_dict['gt_boxes'] assert gt_boxes.shape.__len__() == 3, 'gt_boxes.shape=%s' % str(gt_boxes.shape) assert point_coords.shape.__len__() in [2], 'points.shape=%s' % str(point_coords.shape) batch_size = gt_boxes.shape[0] extend_gt_boxes = box_utils.enlarge_box3d( gt_boxes.view(-1, gt_boxes.shape[-1]), extra_width=self.model_cfg.TARGET_CONFIG.GT_EXTRA_WIDTH ).view(batch_size, -1, gt_boxes.shape[-1]) targets_dict = self.assign_stack_targets( points=point_coords, gt_boxes=gt_boxes, extend_gt_boxes=extend_gt_boxes, set_ignore_flag=True, use_ball_constraint=False, ret_part_labels=True, ret_box_labels=(self.box_layers is not None) ) return targets_dict def get_loss(self, tb_dict=None): tb_dict = {} if tb_dict is None else tb_dict point_loss_cls, tb_dict = self.get_cls_layer_loss(tb_dict) point_loss_part, tb_dict = self.get_part_layer_loss(tb_dict) point_loss = point_loss_cls + point_loss_part if self.box_layers is not None: point_loss_box, tb_dict = self.get_box_layer_loss(tb_dict) point_loss += point_loss_box return point_loss, tb_dict def forward(self, batch_dict): """ Args: batch_dict: batch_size: point_features: (N1 + N2 + N3 + ..., C) or (B, N, C) point_coords: (N1 + N2 + N3 + ..., 4) [bs_idx, x, y, z] point_labels (optional): (N1 + N2 + N3 + ...) gt_boxes (optional): (B, M, 8) Returns: batch_dict: point_cls_scores: (N1 + N2 + N3 + ..., 1) point_part_offset: (N1 + N2 + N3 + ..., 3) """ point_features = batch_dict['point_features'] point_cls_preds = self.cls_layers(point_features) # (total_points, num_class) point_part_preds = self.part_reg_layers(point_features) ret_dict = { 'point_cls_preds': point_cls_preds, 'point_part_preds': point_part_preds, } if self.box_layers is not None: point_box_preds = self.box_layers(point_features) ret_dict['point_box_preds'] = point_box_preds point_cls_scores = torch.sigmoid(point_cls_preds) point_part_offset = torch.sigmoid(point_part_preds) batch_dict['point_cls_scores'], _ = point_cls_scores.max(dim=-1) batch_dict['point_part_offset'] = point_part_offset if self.training: targets_dict = self.assign_targets(batch_dict) ret_dict['point_cls_labels'] = targets_dict['point_cls_labels'] ret_dict['point_part_labels'] = targets_dict.get('point_part_labels') ret_dict['point_box_labels'] = targets_dict.get('point_box_labels') if self.box_layers is not None and (not self.training or self.predict_boxes_when_training): point_cls_preds, point_box_preds = self.generate_predicted_boxes( points=batch_dict['point_coords'][:, 1:4], point_cls_preds=point_cls_preds, point_box_preds=ret_dict['point_box_preds'] ) batch_dict['batch_cls_preds'] = point_cls_preds batch_dict['batch_box_preds'] = point_box_preds batch_dict['batch_index'] = batch_dict['point_coords'][:, 0] batch_dict['cls_preds_normalized'] = False self.forward_ret_dict = ret_dict return batch_dict
5,568
42.507813
107
py
SASA
SASA-main/pcdet/models/dense_heads/target_assigner/anchor_generator.py
import torch class AnchorGenerator(object): def __init__(self, anchor_range, anchor_generator_config): super().__init__() self.anchor_generator_cfg = anchor_generator_config self.anchor_range = anchor_range self.anchor_sizes = [config['anchor_sizes'] for config in anchor_generator_config] self.anchor_rotations = [config['anchor_rotations'] for config in anchor_generator_config] self.anchor_heights = [config['anchor_bottom_heights'] for config in anchor_generator_config] self.align_center = [config.get('align_center', False) for config in anchor_generator_config] assert len(self.anchor_sizes) == len(self.anchor_rotations) == len(self.anchor_heights) self.num_of_anchor_sets = len(self.anchor_sizes) def generate_anchors(self, grid_sizes): assert len(grid_sizes) == self.num_of_anchor_sets all_anchors = [] num_anchors_per_location = [] for grid_size, anchor_size, anchor_rotation, anchor_height, align_center in zip( grid_sizes, self.anchor_sizes, self.anchor_rotations, self.anchor_heights, self.align_center): num_anchors_per_location.append(len(anchor_rotation) * len(anchor_size) * len(anchor_height)) if align_center: x_stride = (self.anchor_range[3] - self.anchor_range[0]) / grid_size[0] y_stride = (self.anchor_range[4] - self.anchor_range[1]) / grid_size[1] x_offset, y_offset = x_stride / 2, y_stride / 2 else: x_stride = (self.anchor_range[3] - self.anchor_range[0]) / (grid_size[0] - 1) y_stride = (self.anchor_range[4] - self.anchor_range[1]) / (grid_size[1] - 1) x_offset, y_offset = 0, 0 x_shifts = torch.arange( self.anchor_range[0] + x_offset, self.anchor_range[3] + 1e-5, step=x_stride, dtype=torch.float32, ).cuda() y_shifts = torch.arange( self.anchor_range[1] + y_offset, self.anchor_range[4] + 1e-5, step=y_stride, dtype=torch.float32, ).cuda() z_shifts = x_shifts.new_tensor(anchor_height) num_anchor_size, num_anchor_rotation = anchor_size.__len__(), anchor_rotation.__len__() anchor_rotation = x_shifts.new_tensor(anchor_rotation) anchor_size = x_shifts.new_tensor(anchor_size) x_shifts, y_shifts, z_shifts = torch.meshgrid([ x_shifts, y_shifts, z_shifts ]) # [x_grid, y_grid, z_grid] anchors = torch.stack((x_shifts, y_shifts, z_shifts), dim=-1) # [x, y, z, 3] anchors = anchors[:, :, :, None, :].repeat(1, 1, 1, anchor_size.shape[0], 1) anchor_size = anchor_size.view(1, 1, 1, -1, 3).repeat([*anchors.shape[0:3], 1, 1]) anchors = torch.cat((anchors, anchor_size), dim=-1) anchors = anchors[:, :, :, :, None, :].repeat(1, 1, 1, 1, num_anchor_rotation, 1) anchor_rotation = anchor_rotation.view(1, 1, 1, 1, -1, 1).repeat([*anchors.shape[0:3], num_anchor_size, 1, 1]) anchors = torch.cat((anchors, anchor_rotation), dim=-1) # [x, y, z, num_size, num_rot, 7] anchors = anchors.permute(2, 1, 0, 3, 4, 5).contiguous() #anchors = anchors.view(-1, anchors.shape[-1]) anchors[..., 2] += anchors[..., 5] / 2 # shift to box centers all_anchors.append(anchors) return all_anchors, num_anchors_per_location if __name__ == '__main__': from easydict import EasyDict config = [ EasyDict({ 'anchor_sizes': [[2.1, 4.7, 1.7], [0.86, 0.91, 1.73], [0.84, 1.78, 1.78]], 'anchor_rotations': [0, 1.57], 'anchor_heights': [0, 0.5] }) ] A = AnchorGenerator( anchor_range=[-75.2, -75.2, -2, 75.2, 75.2, 4], anchor_generator_config=config ) import pdb pdb.set_trace() A.generate_anchors([[188, 188]])
3,990
48.8875
122
py
SASA
SASA-main/pcdet/models/dense_heads/target_assigner/axis_aligned_target_assigner.py
import numpy as np import torch from ....ops.iou3d_nms import iou3d_nms_utils from ....utils import box_utils class AxisAlignedTargetAssigner(object): def __init__(self, model_cfg, class_names, box_coder, match_height=False): super().__init__() anchor_generator_cfg = model_cfg.ANCHOR_GENERATOR_CONFIG anchor_target_cfg = model_cfg.TARGET_ASSIGNER_CONFIG self.box_coder = box_coder self.match_height = match_height self.class_names = np.array(class_names) self.anchor_class_names = [config['class_name'] for config in anchor_generator_cfg] self.pos_fraction = anchor_target_cfg.POS_FRACTION if anchor_target_cfg.POS_FRACTION >= 0 else None self.sample_size = anchor_target_cfg.SAMPLE_SIZE self.norm_by_num_examples = anchor_target_cfg.NORM_BY_NUM_EXAMPLES self.matched_thresholds = {} self.unmatched_thresholds = {} for config in anchor_generator_cfg: self.matched_thresholds[config['class_name']] = config['matched_threshold'] self.unmatched_thresholds[config['class_name']] = config['unmatched_threshold'] self.use_multihead = model_cfg.get('USE_MULTIHEAD', False) self.seperate_multihead = model_cfg.get('SEPARATE_MULTIHEAD', False) if self.seperate_multihead: rpn_head_cfgs = model_cfg.RPN_HEAD_CFGS self.gt_remapping = {} for rpn_head_cfg in rpn_head_cfgs: for idx, name in enumerate(rpn_head_cfg['HEAD_CLS_NAME']): self.gt_remapping[name] = idx + 1 def assign_targets(self, all_anchors, gt_boxes_with_classes): """ Args: all_anchors: [(N, 7), ...] gt_boxes: (B, M, 8) Returns: """ bbox_targets = [] cls_labels = [] reg_weights = [] batch_size = gt_boxes_with_classes.shape[0] gt_classes = gt_boxes_with_classes[:, :, -1] gt_boxes = gt_boxes_with_classes[:, :, :-1] for k in range(batch_size): cur_gt = gt_boxes[k] cnt = cur_gt.__len__() - 1 while cnt > 0 and cur_gt[cnt].sum() == 0: cnt -= 1 cur_gt = cur_gt[:cnt + 1] cur_gt_classes = gt_classes[k][:cnt + 1].int() target_list = [] for anchor_class_name, anchors in zip(self.anchor_class_names, all_anchors): if cur_gt_classes.shape[0] > 1: mask = torch.from_numpy(self.class_names[cur_gt_classes.cpu() - 1] == anchor_class_name) else: mask = torch.tensor([self.class_names[c - 1] == anchor_class_name for c in cur_gt_classes], dtype=torch.bool) if self.use_multihead: anchors = anchors.permute(3, 4, 0, 1, 2, 5).contiguous().view(-1, anchors.shape[-1]) if self.seperate_multihead: selected_classes = cur_gt_classes[mask].clone() if len(selected_classes) > 0: new_cls_id = self.gt_remapping[anchor_class_name] selected_classes[:] = new_cls_id else: selected_classes = cur_gt_classes[mask] else: feature_map_size = anchors.shape[:3] anchors = anchors.view(-1, anchors.shape[-1]) selected_classes = cur_gt_classes[mask] single_target = self.assign_targets_single( anchors, cur_gt[mask], gt_classes=selected_classes, matched_threshold=self.matched_thresholds[anchor_class_name], unmatched_threshold=self.unmatched_thresholds[anchor_class_name] ) target_list.append(single_target) if self.use_multihead: target_dict = { 'box_cls_labels': [t['box_cls_labels'].view(-1) for t in target_list], 'box_reg_targets': [t['box_reg_targets'].view(-1, self.box_coder.code_size) for t in target_list], 'reg_weights': [t['reg_weights'].view(-1) for t in target_list] } target_dict['box_reg_targets'] = torch.cat(target_dict['box_reg_targets'], dim=0) target_dict['box_cls_labels'] = torch.cat(target_dict['box_cls_labels'], dim=0).view(-1) target_dict['reg_weights'] = torch.cat(target_dict['reg_weights'], dim=0).view(-1) else: target_dict = { 'box_cls_labels': [t['box_cls_labels'].view(*feature_map_size, -1) for t in target_list], 'box_reg_targets': [t['box_reg_targets'].view(*feature_map_size, -1, self.box_coder.code_size) for t in target_list], 'reg_weights': [t['reg_weights'].view(*feature_map_size, -1) for t in target_list] } target_dict['box_reg_targets'] = torch.cat( target_dict['box_reg_targets'], dim=-2 ).view(-1, self.box_coder.code_size) target_dict['box_cls_labels'] = torch.cat(target_dict['box_cls_labels'], dim=-1).view(-1) target_dict['reg_weights'] = torch.cat(target_dict['reg_weights'], dim=-1).view(-1) bbox_targets.append(target_dict['box_reg_targets']) cls_labels.append(target_dict['box_cls_labels']) reg_weights.append(target_dict['reg_weights']) bbox_targets = torch.stack(bbox_targets, dim=0) cls_labels = torch.stack(cls_labels, dim=0) reg_weights = torch.stack(reg_weights, dim=0) all_targets_dict = { 'box_cls_labels': cls_labels, 'box_reg_targets': bbox_targets, 'reg_weights': reg_weights } return all_targets_dict def assign_targets_single(self, anchors, gt_boxes, gt_classes, matched_threshold=0.6, unmatched_threshold=0.45 ): num_anchors = anchors.shape[0] num_gt = gt_boxes.shape[0] labels = torch.ones((num_anchors,), dtype=torch.int32, device=anchors.device) * -1 gt_ids = torch.ones((num_anchors,), dtype=torch.int32, device=anchors.device) * -1 if len(gt_boxes) > 0 and anchors.shape[0] > 0: anchor_by_gt_overlap = iou3d_nms_utils.boxes_iou3d_gpu(anchors[:, 0:7], gt_boxes[:, 0:7]) \ if self.match_height else box_utils.boxes3d_nearest_bev_iou(anchors[:, 0:7], gt_boxes[:, 0:7]) anchor_to_gt_argmax = torch.from_numpy(anchor_by_gt_overlap.cpu().numpy().argmax(axis=1)).cuda() anchor_to_gt_max = anchor_by_gt_overlap[ torch.arange(num_anchors, device=anchors.device), anchor_to_gt_argmax ] gt_to_anchor_argmax = torch.from_numpy(anchor_by_gt_overlap.cpu().numpy().argmax(axis=0)).cuda() gt_to_anchor_max = anchor_by_gt_overlap[gt_to_anchor_argmax, torch.arange(num_gt, device=anchors.device)] empty_gt_mask = gt_to_anchor_max == 0 gt_to_anchor_max[empty_gt_mask] = -1 anchors_with_max_overlap = (anchor_by_gt_overlap == gt_to_anchor_max).nonzero()[:, 0] gt_inds_force = anchor_to_gt_argmax[anchors_with_max_overlap] labels[anchors_with_max_overlap] = gt_classes[gt_inds_force] gt_ids[anchors_with_max_overlap] = gt_inds_force.int() pos_inds = anchor_to_gt_max >= matched_threshold gt_inds_over_thresh = anchor_to_gt_argmax[pos_inds] labels[pos_inds] = gt_classes[gt_inds_over_thresh] gt_ids[pos_inds] = gt_inds_over_thresh.int() bg_inds = (anchor_to_gt_max < unmatched_threshold).nonzero()[:, 0] else: bg_inds = torch.arange(num_anchors, device=anchors.device) fg_inds = (labels > 0).nonzero()[:, 0] if self.pos_fraction is not None: num_fg = int(self.pos_fraction * self.sample_size) if len(fg_inds) > num_fg: num_disabled = len(fg_inds) - num_fg disable_inds = torch.randperm(len(fg_inds))[:num_disabled] labels[disable_inds] = -1 fg_inds = (labels > 0).nonzero()[:, 0] num_bg = self.sample_size - (labels > 0).sum() if len(bg_inds) > num_bg: enable_inds = bg_inds[torch.randint(0, len(bg_inds), size=(num_bg,))] labels[enable_inds] = 0 # bg_inds = torch.nonzero(labels == 0)[:, 0] else: if len(gt_boxes) == 0 or anchors.shape[0] == 0: labels[:] = 0 else: labels[bg_inds] = 0 labels[anchors_with_max_overlap] = gt_classes[gt_inds_force] bbox_targets = anchors.new_zeros((num_anchors, self.box_coder.code_size)) if len(gt_boxes) > 0 and anchors.shape[0] > 0: fg_gt_boxes = gt_boxes[anchor_to_gt_argmax[fg_inds], :] fg_anchors = anchors[fg_inds, :] bbox_targets[fg_inds, :] = self.box_coder.encode_torch(fg_gt_boxes, fg_anchors) reg_weights = anchors.new_zeros((num_anchors,)) if self.norm_by_num_examples: num_examples = (labels >= 0).sum() num_examples = num_examples if num_examples > 1.0 else 1.0 reg_weights[labels > 0] = 1.0 / num_examples else: reg_weights[labels > 0] = 1.0 ret_dict = { 'box_cls_labels': labels, 'box_reg_targets': bbox_targets, 'reg_weights': reg_weights, } return ret_dict
9,874
45.14486
118
py
SASA
SASA-main/pcdet/models/dense_heads/target_assigner/atss_target_assigner.py
import torch from ....ops.iou3d_nms import iou3d_nms_utils from ....utils import common_utils class ATSSTargetAssigner(object): """ Reference: https://arxiv.org/abs/1912.02424 """ def __init__(self, topk, box_coder, match_height=False): self.topk = topk self.box_coder = box_coder self.match_height = match_height def assign_targets(self, anchors_list, gt_boxes_with_classes, use_multihead=False): """ Args: anchors: [(N, 7), ...] gt_boxes: (B, M, 8) Returns: """ if not isinstance(anchors_list, list): anchors_list = [anchors_list] single_set_of_anchor = True else: single_set_of_anchor = len(anchors_list) == 1 cls_labels_list, reg_targets_list, reg_weights_list = [], [], [] for anchors in anchors_list: batch_size = gt_boxes_with_classes.shape[0] gt_classes = gt_boxes_with_classes[:, :, -1] gt_boxes = gt_boxes_with_classes[:, :, :-1] if use_multihead: anchors = anchors.permute(3, 4, 0, 1, 2, 5).contiguous().view(-1, anchors.shape[-1]) else: anchors = anchors.view(-1, anchors.shape[-1]) cls_labels, reg_targets, reg_weights = [], [], [] for k in range(batch_size): cur_gt = gt_boxes[k] cnt = cur_gt.__len__() - 1 while cnt > 0 and cur_gt[cnt].sum() == 0: cnt -= 1 cur_gt = cur_gt[:cnt + 1] cur_gt_classes = gt_classes[k][:cnt + 1] cur_cls_labels, cur_reg_targets, cur_reg_weights = self.assign_targets_single( anchors, cur_gt, cur_gt_classes ) cls_labels.append(cur_cls_labels) reg_targets.append(cur_reg_targets) reg_weights.append(cur_reg_weights) cls_labels = torch.stack(cls_labels, dim=0) reg_targets = torch.stack(reg_targets, dim=0) reg_weights = torch.stack(reg_weights, dim=0) cls_labels_list.append(cls_labels) reg_targets_list.append(reg_targets) reg_weights_list.append(reg_weights) if single_set_of_anchor: ret_dict = { 'box_cls_labels': cls_labels_list[0], 'box_reg_targets': reg_targets_list[0], 'reg_weights': reg_weights_list[0] } else: ret_dict = { 'box_cls_labels': torch.cat(cls_labels_list, dim=1), 'box_reg_targets': torch.cat(reg_targets_list, dim=1), 'reg_weights': torch.cat(reg_weights_list, dim=1) } return ret_dict def assign_targets_single(self, anchors, gt_boxes, gt_classes): """ Args: anchors: (N, 7) [x, y, z, dx, dy, dz, heading] gt_boxes: (M, 7) [x, y, z, dx, dy, dz, heading] gt_classes: (M) Returns: """ num_anchor = anchors.shape[0] num_gt = gt_boxes.shape[0] # select topk anchors for each gt_boxes if self.match_height: ious = iou3d_nms_utils.boxes_iou3d_gpu(anchors[:, 0:7], gt_boxes[:, 0:7]) # (N, M) else: ious = iou3d_nms_utils.boxes_iou_bev(anchors[:, 0:7], gt_boxes[:, 0:7]) distance = (anchors[:, None, 0:3] - gt_boxes[None, :, 0:3]).norm(dim=-1) # (N, M) _, topk_idxs = distance.topk(self.topk, dim=0, largest=False) # (K, M) candidate_ious = ious[topk_idxs, torch.arange(num_gt)] # (K, M) iou_mean_per_gt = candidate_ious.mean(dim=0) iou_std_per_gt = candidate_ious.std(dim=0) iou_thresh_per_gt = iou_mean_per_gt + iou_std_per_gt + 1e-6 is_pos = candidate_ious >= iou_thresh_per_gt[None, :] # (K, M) # check whether anchor_center in gt_boxes, only check BEV x-y axes candidate_anchors = anchors[topk_idxs.view(-1)] # (KxM, 7) gt_boxes_of_each_anchor = gt_boxes[:, :].repeat(self.topk, 1) # (KxM, 7) xyz_local = candidate_anchors[:, 0:3] - gt_boxes_of_each_anchor[:, 0:3] xyz_local = common_utils.rotate_points_along_z( xyz_local[:, None, :], -gt_boxes_of_each_anchor[:, 6] ).squeeze(dim=1) xy_local = xyz_local[:, 0:2] lw = gt_boxes_of_each_anchor[:, 3:5][:, [1, 0]] # bugfixed: w ==> y, l ==> x in local coords is_in_gt = ((xy_local <= lw / 2) & (xy_local >= -lw / 2)).all(dim=-1).view(-1, num_gt) # (K, M) is_pos = is_pos & is_in_gt # (K, M) for ng in range(num_gt): topk_idxs[:, ng] += ng * num_anchor # select the highest IoU if an anchor box is assigned with multiple gt_boxes INF = -0x7FFFFFFF ious_inf = torch.full_like(ious, INF).t().contiguous().view(-1) # (MxN) index = topk_idxs.view(-1)[is_pos.view(-1)] ious_inf[index] = ious.t().contiguous().view(-1)[index] ious_inf = ious_inf.view(num_gt, -1).t() # (N, M) anchors_to_gt_values, anchors_to_gt_indexs = ious_inf.max(dim=1) # match the gt_boxes to the anchors which have maximum iou with them max_iou_of_each_gt, argmax_iou_of_each_gt = ious.max(dim=0) anchors_to_gt_indexs[argmax_iou_of_each_gt] = torch.arange(0, num_gt, device=ious.device) anchors_to_gt_values[argmax_iou_of_each_gt] = max_iou_of_each_gt cls_labels = gt_classes[anchors_to_gt_indexs] cls_labels[anchors_to_gt_values == INF] = 0 matched_gts = gt_boxes[anchors_to_gt_indexs] pos_mask = cls_labels > 0 reg_targets = matched_gts.new_zeros((num_anchor, self.box_coder.code_size)) reg_weights = matched_gts.new_zeros(num_anchor) if pos_mask.sum() > 0: reg_targets[pos_mask > 0] = self.box_coder.encode_torch(matched_gts[pos_mask > 0], anchors[pos_mask > 0]) reg_weights[pos_mask] = 1.0 return cls_labels, reg_targets, reg_weights
6,050
41.612676
117
py
SASA
SASA-main/pcdet/models/roi_heads/roi_head_template.py
import numpy as np import torch import torch.nn as nn import torch.nn.functional as F from ...utils import box_coder_utils, common_utils, loss_utils from ..model_utils.model_nms_utils import class_agnostic_nms from .target_assigner.proposal_target_layer import ProposalTargetLayer class RoIHeadTemplate(nn.Module): def __init__(self, num_class, model_cfg): super().__init__() self.model_cfg = model_cfg self.num_class = num_class self.box_coder = getattr(box_coder_utils, self.model_cfg.TARGET_CONFIG.BOX_CODER)( **self.model_cfg.TARGET_CONFIG.get('BOX_CODER_CONFIG', {}) ) self.proposal_target_layer = ProposalTargetLayer(roi_sampler_cfg=self.model_cfg.TARGET_CONFIG) self.build_losses(self.model_cfg.LOSS_CONFIG) self.forward_ret_dict = None def build_losses(self, losses_cfg): self.add_module( 'reg_loss_func', loss_utils.WeightedSmoothL1Loss(code_weights=losses_cfg.LOSS_WEIGHTS['code_weights']) ) def make_fc_layers(self, input_channels, output_channels, fc_list): fc_layers = [] pre_channel = input_channels for k in range(0, fc_list.__len__()): fc_layers.extend([ nn.Conv1d(pre_channel, fc_list[k], kernel_size=1, bias=False), nn.BatchNorm1d(fc_list[k]), nn.ReLU() ]) pre_channel = fc_list[k] if self.model_cfg.DP_RATIO >= 0 and k == 0: fc_layers.append(nn.Dropout(self.model_cfg.DP_RATIO)) fc_layers.append(nn.Conv1d(pre_channel, output_channels, kernel_size=1, bias=True)) fc_layers = nn.Sequential(*fc_layers) return fc_layers @torch.no_grad() def proposal_layer(self, batch_dict, nms_config): """ Args: batch_dict: batch_size: batch_cls_preds: (B, num_boxes, num_classes | 1) or (N1+N2+..., num_classes | 1) batch_box_preds: (B, num_boxes, 7+C) or (N1+N2+..., 7+C) cls_preds_normalized: indicate whether batch_cls_preds is normalized batch_index: optional (N1+N2+...) nms_config: Returns: batch_dict: rois: (B, num_rois, 7+C) roi_scores: (B, num_rois) roi_labels: (B, num_rois) """ batch_size = batch_dict['batch_size'] batch_box_preds = batch_dict['batch_box_preds'] batch_cls_preds = batch_dict['batch_cls_preds'] rois = batch_box_preds.new_zeros((batch_size, nms_config.NMS_POST_MAXSIZE, batch_box_preds.shape[-1])) roi_scores = batch_box_preds.new_zeros((batch_size, nms_config.NMS_POST_MAXSIZE)) roi_labels = batch_box_preds.new_zeros((batch_size, nms_config.NMS_POST_MAXSIZE), dtype=torch.long) for index in range(batch_size): if batch_dict.get('batch_index', None) is not None: assert batch_cls_preds.shape.__len__() == 2 batch_mask = (batch_dict['batch_index'] == index) else: assert batch_dict['batch_cls_preds'].shape.__len__() == 3 batch_mask = index box_preds = batch_box_preds[batch_mask] cls_preds = batch_cls_preds[batch_mask] cur_roi_scores, cur_roi_labels = torch.max(cls_preds, dim=1) if nms_config.MULTI_CLASSES_NMS: raise NotImplementedError else: selected, selected_scores = class_agnostic_nms( box_scores=cur_roi_scores, box_preds=box_preds, nms_config=nms_config ) rois[index, :len(selected), :] = box_preds[selected] roi_scores[index, :len(selected)] = cur_roi_scores[selected] roi_labels[index, :len(selected)] = cur_roi_labels[selected] batch_dict['rois'] = rois batch_dict['roi_scores'] = roi_scores batch_dict['roi_labels'] = roi_labels + 1 batch_dict['has_class_labels'] = True if batch_cls_preds.shape[-1] > 1 else False batch_dict.pop('batch_index', None) return batch_dict def assign_targets(self, batch_dict): batch_size = batch_dict['batch_size'] with torch.no_grad(): targets_dict = self.proposal_target_layer.forward(batch_dict) rois = targets_dict['rois'] # (B, N, 7 + C) gt_of_rois = targets_dict['gt_of_rois'] # (B, N, 7 + C + 1) targets_dict['gt_of_rois_src'] = gt_of_rois.clone().detach() # canonical transformation roi_center = rois[:, :, 0:3] roi_ry = rois[:, :, 6] % (2 * np.pi) gt_of_rois[:, :, 0:3] = gt_of_rois[:, :, 0:3] - roi_center gt_of_rois[:, :, 6] = gt_of_rois[:, :, 6] - roi_ry # transfer LiDAR coords to local coords gt_of_rois = common_utils.rotate_points_along_z( points=gt_of_rois.view(-1, 1, gt_of_rois.shape[-1]), angle=-roi_ry.view(-1) ).view(batch_size, -1, gt_of_rois.shape[-1]) # flip orientation if rois have opposite orientation heading_label = gt_of_rois[:, :, 6] % (2 * np.pi) # 0 ~ 2pi opposite_flag = (heading_label > np.pi * 0.5) & (heading_label < np.pi * 1.5) heading_label[opposite_flag] = (heading_label[opposite_flag] + np.pi) % (2 * np.pi) # (0 ~ pi/2, 3pi/2 ~ 2pi) flag = heading_label > np.pi heading_label[flag] = heading_label[flag] - np.pi * 2 # (-pi/2, pi/2) heading_label = torch.clamp(heading_label, min=-np.pi / 2, max=np.pi / 2) gt_of_rois[:, :, 6] = heading_label targets_dict['gt_of_rois'] = gt_of_rois return targets_dict def get_box_reg_layer_loss(self, forward_ret_dict): loss_cfgs = self.model_cfg.LOSS_CONFIG code_size = self.box_coder.code_size reg_valid_mask = forward_ret_dict['reg_valid_mask'].view(-1) gt_boxes3d_ct = forward_ret_dict['gt_of_rois'][..., 0:code_size] gt_of_rois_src = forward_ret_dict['gt_of_rois_src'][..., 0:code_size].view(-1, code_size) rcnn_reg = forward_ret_dict['rcnn_reg'] # (rcnn_batch_size, C) roi_boxes3d = forward_ret_dict['rois'] rcnn_batch_size = gt_boxes3d_ct.view(-1, code_size).shape[0] fg_mask = (reg_valid_mask > 0) fg_sum = fg_mask.long().sum().item() tb_dict = {} if loss_cfgs.REG_LOSS == 'smooth-l1': rois_anchor = roi_boxes3d.clone().detach().view(-1, code_size) rois_anchor[:, 0:3] = 0 rois_anchor[:, 6] = 0 reg_targets = self.box_coder.encode_torch( gt_boxes3d_ct.view(rcnn_batch_size, code_size), rois_anchor ) rcnn_loss_reg = self.reg_loss_func( rcnn_reg.view(rcnn_batch_size, -1).unsqueeze(dim=0), reg_targets.unsqueeze(dim=0), ) # [B, M, 7] rcnn_loss_reg = (rcnn_loss_reg.view(rcnn_batch_size, -1) * fg_mask.unsqueeze(dim=-1).float()).sum() / max(fg_sum, 1) rcnn_loss_reg = rcnn_loss_reg * loss_cfgs.LOSS_WEIGHTS['rcnn_reg_weight'] tb_dict['rcnn_loss_reg'] = rcnn_loss_reg.item() if loss_cfgs.CORNER_LOSS_REGULARIZATION and fg_sum > 0: # TODO: NEED to BE CHECK fg_rcnn_reg = rcnn_reg.view(rcnn_batch_size, -1)[fg_mask] fg_roi_boxes3d = roi_boxes3d.view(-1, code_size)[fg_mask] fg_roi_boxes3d = fg_roi_boxes3d.view(1, -1, code_size) batch_anchors = fg_roi_boxes3d.clone().detach() roi_ry = fg_roi_boxes3d[:, :, 6].view(-1) roi_xyz = fg_roi_boxes3d[:, :, 0:3].view(-1, 3) batch_anchors[:, :, 0:3] = 0 rcnn_boxes3d = self.box_coder.decode_torch( fg_rcnn_reg.view(batch_anchors.shape[0], -1, code_size), batch_anchors ).view(-1, code_size) rcnn_boxes3d = common_utils.rotate_points_along_z( rcnn_boxes3d.unsqueeze(dim=1), roi_ry ).squeeze(dim=1) rcnn_boxes3d[:, 0:3] += roi_xyz loss_corner = loss_utils.get_corner_loss_lidar( rcnn_boxes3d[:, 0:7], gt_of_rois_src[fg_mask][:, 0:7] ) loss_corner = loss_corner.mean() loss_corner = loss_corner * loss_cfgs.LOSS_WEIGHTS['rcnn_corner_weight'] rcnn_loss_reg += loss_corner tb_dict['rcnn_loss_corner'] = loss_corner.item() else: raise NotImplementedError return rcnn_loss_reg, tb_dict def get_box_cls_layer_loss(self, forward_ret_dict): loss_cfgs = self.model_cfg.LOSS_CONFIG rcnn_cls = forward_ret_dict['rcnn_cls'] rcnn_cls_labels = forward_ret_dict['rcnn_cls_labels'].view(-1) if loss_cfgs.CLS_LOSS == 'BinaryCrossEntropy': rcnn_cls_flat = rcnn_cls.view(-1) batch_loss_cls = F.binary_cross_entropy(torch.sigmoid(rcnn_cls_flat), rcnn_cls_labels.float(), reduction='none') cls_valid_mask = (rcnn_cls_labels >= 0).float() rcnn_loss_cls = (batch_loss_cls * cls_valid_mask).sum() / torch.clamp(cls_valid_mask.sum(), min=1.0) elif loss_cfgs.CLS_LOSS == 'CrossEntropy': batch_loss_cls = F.cross_entropy(rcnn_cls, rcnn_cls_labels, reduction='none', ignore_index=-1) cls_valid_mask = (rcnn_cls_labels >= 0).float() rcnn_loss_cls = (batch_loss_cls * cls_valid_mask).sum() / torch.clamp(cls_valid_mask.sum(), min=1.0) else: raise NotImplementedError rcnn_loss_cls = rcnn_loss_cls * loss_cfgs.LOSS_WEIGHTS['rcnn_cls_weight'] tb_dict = {'rcnn_loss_cls': rcnn_loss_cls.item()} return rcnn_loss_cls, tb_dict def get_loss(self, tb_dict=None): tb_dict = {} if tb_dict is None else tb_dict rcnn_loss = 0 rcnn_loss_cls, cls_tb_dict = self.get_box_cls_layer_loss(self.forward_ret_dict) rcnn_loss += rcnn_loss_cls tb_dict.update(cls_tb_dict) rcnn_loss_reg, reg_tb_dict = self.get_box_reg_layer_loss(self.forward_ret_dict) rcnn_loss += rcnn_loss_reg tb_dict.update(reg_tb_dict) tb_dict['rcnn_loss'] = rcnn_loss.item() return rcnn_loss, tb_dict def generate_predicted_boxes(self, batch_size, rois, cls_preds, box_preds): """ Args: batch_size: rois: (B, N, 7) cls_preds: (BN, num_class) box_preds: (BN, code_size) Returns: """ code_size = self.box_coder.code_size # batch_cls_preds: (B, N, num_class or 1) batch_cls_preds = cls_preds.view(batch_size, -1, cls_preds.shape[-1]) batch_box_preds = box_preds.view(batch_size, -1, code_size) roi_ry = rois[:, :, 6].view(-1) roi_xyz = rois[:, :, 0:3].view(-1, 3) local_rois = rois.clone().detach() local_rois[:, :, 0:3] = 0 batch_box_preds = self.box_coder.decode_torch(batch_box_preds, local_rois).view(-1, code_size) batch_box_preds = common_utils.rotate_points_along_z( batch_box_preds.unsqueeze(dim=1), roi_ry ).squeeze(dim=1) batch_box_preds[:, 0:3] += roi_xyz batch_box_preds = batch_box_preds.view(batch_size, -1, code_size) return batch_cls_preds, batch_box_preds
11,451
43.216216
128
py
SASA
SASA-main/pcdet/models/roi_heads/partA2_head.py
import numpy as np import spconv import torch import torch.nn as nn from ...ops.roiaware_pool3d import roiaware_pool3d_utils from .roi_head_template import RoIHeadTemplate class PartA2FCHead(RoIHeadTemplate): def __init__(self, input_channels, model_cfg, num_class=1): super().__init__(num_class=num_class, model_cfg=model_cfg) self.model_cfg = model_cfg self.SA_modules = nn.ModuleList() block = self.post_act_block c0 = self.model_cfg.ROI_AWARE_POOL.NUM_FEATURES // 2 self.conv_part = spconv.SparseSequential( block(4, 64, 3, padding=1, indice_key='rcnn_subm1'), block(64, c0, 3, padding=1, indice_key='rcnn_subm1_1'), ) self.conv_rpn = spconv.SparseSequential( block(input_channels, 64, 3, padding=1, indice_key='rcnn_subm2'), block(64, c0, 3, padding=1, indice_key='rcnn_subm1_2'), ) shared_fc_list = [] pool_size = self.model_cfg.ROI_AWARE_POOL.POOL_SIZE pre_channel = self.model_cfg.ROI_AWARE_POOL.NUM_FEATURES * pool_size * pool_size * pool_size for k in range(0, self.model_cfg.SHARED_FC.__len__()): shared_fc_list.extend([ nn.Conv1d(pre_channel, self.model_cfg.SHARED_FC[k], kernel_size=1, bias=False), nn.BatchNorm1d(self.model_cfg.SHARED_FC[k]), nn.ReLU() ]) pre_channel = self.model_cfg.SHARED_FC[k] if k != self.model_cfg.SHARED_FC.__len__() - 1 and self.model_cfg.DP_RATIO > 0: shared_fc_list.append(nn.Dropout(self.model_cfg.DP_RATIO)) self.shared_fc_layer = nn.Sequential(*shared_fc_list) self.cls_layers = self.make_fc_layers( input_channels=pre_channel, output_channels=self.num_class, fc_list=self.model_cfg.CLS_FC ) self.reg_layers = self.make_fc_layers( input_channels=pre_channel, output_channels=self.box_coder.code_size * self.num_class, fc_list=self.model_cfg.REG_FC ) self.roiaware_pool3d_layer = roiaware_pool3d_utils.RoIAwarePool3d( out_size=self.model_cfg.ROI_AWARE_POOL.POOL_SIZE, max_pts_each_voxel=self.model_cfg.ROI_AWARE_POOL.MAX_POINTS_PER_VOXEL ) self.init_weights(weight_init='xavier') def init_weights(self, weight_init='xavier'): if weight_init == 'kaiming': init_func = nn.init.kaiming_normal_ elif weight_init == 'xavier': init_func = nn.init.xavier_normal_ elif weight_init == 'normal': init_func = nn.init.normal_ else: raise NotImplementedError for m in self.modules(): if isinstance(m, nn.Conv2d) or isinstance(m, nn.Conv1d): if weight_init == 'normal': init_func(m.weight, mean=0, std=0.001) else: init_func(m.weight) if m.bias is not None: nn.init.constant_(m.bias, 0) nn.init.normal_(self.reg_layers[-1].weight, mean=0, std=0.001) def post_act_block(self, in_channels, out_channels, kernel_size, indice_key, stride=1, padding=0, conv_type='subm'): if conv_type == 'subm': m = spconv.SparseSequential( spconv.SubMConv3d(in_channels, out_channels, kernel_size, bias=False, indice_key=indice_key), nn.BatchNorm1d(out_channels, eps=1e-3, momentum=0.01), nn.ReLU(), ) elif conv_type == 'spconv': m = spconv.SparseSequential( spconv.SparseConv3d(in_channels, out_channels, kernel_size, stride=stride, padding=padding, bias=False, indice_key=indice_key), nn.BatchNorm1d(out_channels, eps=1e-3, momentum=0.01), nn.ReLU(), ) elif conv_type == 'inverseconv': m = spconv.SparseSequential( spconv.SparseInverseConv3d(in_channels, out_channels, kernel_size, indice_key=indice_key, bias=False), nn.BatchNorm1d(out_channels, eps=1e-3, momentum=0.01), nn.ReLU(), ) else: raise NotImplementedError return m def roiaware_pool(self, batch_dict): """ Args: batch_dict: batch_size: rois: (B, num_rois, 7 + C) point_coords: (num_points, 4) [bs_idx, x, y, z] point_features: (num_points, C) point_cls_scores: (N1 + N2 + N3 + ..., 1) point_part_offset: (N1 + N2 + N3 + ..., 3) Returns: """ batch_size = batch_dict['batch_size'] batch_idx = batch_dict['point_coords'][:, 0] point_coords = batch_dict['point_coords'][:, 1:4] point_features = batch_dict['point_features'] part_features = torch.cat(( batch_dict['point_part_offset'] if not self.model_cfg.get('DISABLE_PART', False) else point_coords, batch_dict['point_cls_scores'].view(-1, 1).detach() ), dim=1) part_features[part_features[:, -1] < self.model_cfg.SEG_MASK_SCORE_THRESH, 0:3] = 0 rois = batch_dict['rois'] pooled_part_features_list, pooled_rpn_features_list = [], [] for bs_idx in range(batch_size): bs_mask = (batch_idx == bs_idx) cur_point_coords = point_coords[bs_mask] cur_part_features = part_features[bs_mask] cur_rpn_features = point_features[bs_mask] cur_roi = rois[bs_idx][:, 0:7].contiguous() # (N, 7) pooled_part_features = self.roiaware_pool3d_layer.forward( cur_roi, cur_point_coords, cur_part_features, pool_method='avg' ) # (N, out_x, out_y, out_z, 4) pooled_rpn_features = self.roiaware_pool3d_layer.forward( cur_roi, cur_point_coords, cur_rpn_features, pool_method='max' ) # (N, out_x, out_y, out_z, C) pooled_part_features_list.append(pooled_part_features) pooled_rpn_features_list.append(pooled_rpn_features) pooled_part_features = torch.cat(pooled_part_features_list, dim=0) # (B * N, out_x, out_y, out_z, 4) pooled_rpn_features = torch.cat(pooled_rpn_features_list, dim=0) # (B * N, out_x, out_y, out_z, C) return pooled_part_features, pooled_rpn_features @staticmethod def fake_sparse_idx(sparse_idx, batch_size_rcnn): print('Warning: Sparse_Idx_Shape(%s) \r' % (str(sparse_idx.shape)), end='', flush=True) # at most one sample is non-empty, then fake the first voxels of each sample(BN needs at least # two values each channel) as non-empty for the below calculation sparse_idx = sparse_idx.new_zeros((batch_size_rcnn, 3)) bs_idxs = torch.arange(batch_size_rcnn).type_as(sparse_idx).view(-1, 1) sparse_idx = torch.cat((bs_idxs, sparse_idx), dim=1) return sparse_idx def forward(self, batch_dict): """ Args: batch_dict: Returns: """ targets_dict = self.proposal_layer( batch_dict, nms_config=self.model_cfg.NMS_CONFIG['TRAIN' if self.training else 'TEST'] ) if self.training: targets_dict = self.assign_targets(batch_dict) batch_dict['rois'] = targets_dict['rois'] batch_dict['roi_labels'] = targets_dict['roi_labels'] # RoI aware pooling pooled_part_features, pooled_rpn_features = self.roiaware_pool(batch_dict) batch_size_rcnn = pooled_part_features.shape[0] # (B * N, out_x, out_y, out_z, 4) # transform to sparse tensors sparse_shape = np.array(pooled_part_features.shape[1:4], dtype=np.int32) sparse_idx = pooled_part_features.sum(dim=-1).nonzero() # (non_empty_num, 4) ==> [bs_idx, x_idx, y_idx, z_idx] if sparse_idx.shape[0] < 3: sparse_idx = self.fake_sparse_idx(sparse_idx, batch_size_rcnn) if self.training: # these are invalid samples targets_dict['rcnn_cls_labels'].fill_(-1) targets_dict['reg_valid_mask'].fill_(-1) part_features = pooled_part_features[sparse_idx[:, 0], sparse_idx[:, 1], sparse_idx[:, 2], sparse_idx[:, 3]] rpn_features = pooled_rpn_features[sparse_idx[:, 0], sparse_idx[:, 1], sparse_idx[:, 2], sparse_idx[:, 3]] coords = sparse_idx.int() part_features = spconv.SparseConvTensor(part_features, coords, sparse_shape, batch_size_rcnn) rpn_features = spconv.SparseConvTensor(rpn_features, coords, sparse_shape, batch_size_rcnn) # forward rcnn network x_part = self.conv_part(part_features) x_rpn = self.conv_rpn(rpn_features) merged_feature = torch.cat((x_rpn.features, x_part.features), dim=1) # (N, C) shared_feature = spconv.SparseConvTensor(merged_feature, coords, sparse_shape, batch_size_rcnn) shared_feature = shared_feature.dense().view(batch_size_rcnn, -1, 1) shared_feature = self.shared_fc_layer(shared_feature) rcnn_cls = self.cls_layers(shared_feature).transpose(1, 2).contiguous().squeeze(dim=1) # (B, 1 or 2) rcnn_reg = self.reg_layers(shared_feature).transpose(1, 2).contiguous().squeeze(dim=1) # (B, C) if not self.training: batch_cls_preds, batch_box_preds = self.generate_predicted_boxes( batch_size=batch_dict['batch_size'], rois=batch_dict['rois'], cls_preds=rcnn_cls, box_preds=rcnn_reg ) batch_dict['batch_cls_preds'] = batch_cls_preds batch_dict['batch_box_preds'] = batch_box_preds batch_dict['cls_preds_normalized'] = False else: targets_dict['rcnn_cls'] = rcnn_cls targets_dict['rcnn_reg'] = rcnn_reg self.forward_ret_dict = targets_dict return batch_dict
10,039
43.622222
120
py
SASA
SASA-main/pcdet/models/roi_heads/__init__.py
from .partA2_head import PartA2FCHead from .pointrcnn_head import PointRCNNHead from .pvrcnn_head import PVRCNNHead from .roi_head_template import RoIHeadTemplate __all__ = { 'RoIHeadTemplate': RoIHeadTemplate, 'PartA2FCHead': PartA2FCHead, 'PVRCNNHead': PVRCNNHead, 'PointRCNNHead': PointRCNNHead }
317
25.5
46
py