def test2(): device = torch.device("cuda") data = np.load('./data/seg_data.npy') X_train = data[0][0][:40] print(X_train.shape) X_train = StandardScaler().fit_transform(X_train) X_train = X_train.reshape(-1, 1, 1920) X_train = torch.from_numpy(X_train) X_train = X_train.cuda().float() print(X_train.device) model = ConvNet(2).to(device) model.load_state_dict(torch.load("./DNNmodel.ckpt", map_location=device)) # model.eval() outputs = model(X_train)
np.random.seed(100) torch.manual_seed(100) torch.cuda.manual_seed(100) hidden = 16 lr = 0.01 weight_decay = 5e-4 dropout = 0.5 epochs = 200 # Model and optimizer model = GCN(nfeat=features.shape[1], nhid=hidden, out_dim=25, dropout=dropout) optimizer = optim.Adam(model.parameters(), lr=lr, weight_decay=weight_decay) model.cuda() features = features.cuda() adj = changed_adj.cuda() labels = labels.cuda() idx_train = idx_train.cuda() idx_val = idx_val.cuda() idx_test = idx_test.cuda() def train(epoch): t = time.time() model.train() optimizer.zero_grad() output = model(features, adj) loss_train = F.nll_loss(output[idx_train], labels[idx_train]) acc_train = accuracy(output[idx_train], labels[idx_train]) loss_train.backward()