Functions — github.com/TrustAGI-Lab/MTGODE

CodeHub MCPcopy Create free account

hub / github.com/TrustAGI-Lab/MTGODE / functions

Functions85 in github.com/TrustAGI-Lab/MTGODE

⨍ Functions 85 ◇ Types & classes 25

↓ 16 callersMethod__init__

(self, cin, cout, alpha=2.0, method='rk4', time=1.0, step_size=1.0, rtol=1e-4, atol=1e-3, adj

↓ 5 callersMethodinverse_transform

↓ 4 callersFunctionevaluate

(data, X, Y, model, evaluateL2, evaluateL1, batch_size, runid, save_prediction=False)

run_single_step.py:81

↓ 4 callersMethodget_iterator

↓ 3 callersMethod__init__

(self, buildA_true, num_nodes, device, predefined_A=None, static_feat=None, dropout=0.3, subg

↓ 3 callersMethod_batchify

(self, idx_set, horizon)

↓ 2 callersMethodeval

(self, input, real_val)

↓ 2 callersMethodget_batches

(self, inputs, targets, batch_size, shuffle=True)

↓ 2 callersFunctionmetric

↓ 2 callersMethodsetIntermediate

(self, dilation)

↓ 2 callersMethodtrain

(self, input, real_val, idx=None)

↓ 1 callersMethod_normalized

(self, normalize)

↓ 1 callersMethod_split

(self, train, valid, test)

↓ 1 callersFunctioncalculate_normalized_laplacian

# L = D^-1/2 (D-A) D^-1/2 = I - D^-1/2 A D^-1/2 # D = diag(A 1) :param adj: :return:

↓ 1 callersFunctionload_adj

↓ 1 callersFunctionload_dataset

(dataset_dir, batch_size, valid_batch_size=None, test_batch_size=None)

↓ 1 callersFunctionload_pickle

↓ 1 callersFunctionmain

run_single_step.py:196

↓ 1 callersFunctionmain

run_multi_step.py:76

↓ 1 callersFunctionmasked_mae

(preds, labels, null_val=np.nan)

↓ 1 callersFunctionmasked_mape

(preds, labels, null_val=np.nan)

↓ 1 callersFunctionmasked_mse

(preds, labels, null_val=np.nan)

↓ 1 callersFunctionmasked_rmse

(preds, labels, null_val=np.nan)

↓ 1 callersFunctionnormal_std

↓ 1 callersMethodreset_parameters

↓ 1 callersMethodreset_parameters

↓ 1 callersMethodsetGraph

↓ 1 callersMethodset_adj

↓ 1 callersMethodset_x0

↓ 1 callersMethodshuffle

↓ 1 callersFunctiontrain

(data, X, Y, model, criterion, optim, batch_size, clip=None)

run_single_step.py:143

↓ 1 callersMethodtransform

(self,c_in,c_out,bias=True)

(self,c_in,c_out,gdep,dropout,alpha)

(self, c_in, c_out, gdep, dropout, alpha)

(self,c_in,c_out,gdep,dropout,alpha)

(self, cin, cout, dilation_factor=2)

(self, cin, cout, dilation_factor=1)

(self, nnodes, k, dim, device, alpha=3, static_feat=None)

(self, nnodes, k, dim, device, alpha=3, static_feat=None)

(self, nnodes, k, dim, device, alpha=3, static_feat=None)

(self, nnodes, k, dim, device, alpha=3, static_feat=None)

(self, normalized_shape, eps=1e-5, elementwise_affine=True)

(self, c_in, c_out, init_alpha)

(self, cgpfunc, method, step_size, rtol, atol, adjoint, perturb, estimated_nfe)

(self, file_name, train, valid, device, horizon, window, normalize=2)

:param xs: :param ys: :param batch_size: :param pad_with_last_sample: pad with the last sample to make number of samp

(self, mean, std)

(self, model, lrate, wdecay, clip, step_size, seq_out_len, scaler, device, cl=True)

(self, odefunc, method, step_size, rtol, atol, adjoint=False, perturb=False)

(self, receptive_field, dilation, hidden_channels, dropout, method, time, step_size, alpha, r

Functionasym_adj

Asymmetrically normalize adjacency matrix.

Functioncalculate_scaled_laplacian

(adj_mx, lambda_max=2, undirected=True)

Methodextra_repr

(self, input, idx)

(self, input, idx=None)

Functionload_node_feature

Functionstr_to_bool

run_single_step.py:12

Functionstr_to_bool

run_multi_step.py:10

Functionsym_adj

Symmetrically normalize adjacency matrix.