MCPcopy Create free account

hub / github.com/davidhallac/TICC / functions

Functions112 in github.com/davidhallac/TICC

⨍ Functions 112◇ Types & classes 4
↓ 15 callersFunctiongetValue
(arr, index, length)
paper code/solveCrossTime.py:932
↓ 14 callersMethodAddNode
(self, NId, Objective=__default_objective,\ Constraints=__default_constraints)
paper code/solveCrossTime.py:537
↓ 13 callersMethod__GetEdgeTup
(self, NId1, NId2)
paper code/solveCrossTime.py:565
↓ 11 callersMethodNodes
(self)
paper code/solveCrossTime.py:88
↓ 11 callersMethodfit
Main method for TICC solver. Parameters: - input_file: location of the data file
TICC_solver.py:50
↓ 10 callersMethodGetNodeVariables
(self, NId)
paper code/solveCrossTime.py:627
↓ 10 callersFunctioncompute_confusion_matrix
computes a confusion matrix and returns it
paper code/network_accuracy.py:175
↓ 10 callersFunctioncompute_confusion_matrix
computes a confusion matrix and returns it
paper code/TICC.py:175
↓ 9 callersMethodEdges
(self)
paper code/solveCrossTime.py:96
↓ 8 callersMethodAddEdge
(self, SrcNId, DstNId, ObjectiveFunc=None, Objective=__default_objective, Constraints=__default_c
paper code/solveCrossTime.py:582
↓ 8 callersMethodGetNodeValue
(self, NId, Name)
paper code/solveCrossTime.py:480
↓ 8 callersMethodSetEdgeObjective
(self, SrcNId, DstNId, Objective)
paper code/solveCrossTime.py:602
↓ 8 callersFunctionij2symmetric
(i,j,size)
paper code/solveCrossTime.py:973
↓ 7 callersFunctioncompute_confusion_matrix
computes a confusion matrix and returns it
paper code/car.py:167
↓ 6 callersMethodSetNodeObjective
(self, NId, Objective)
paper code/solveCrossTime.py:545
↓ 6 callersMethodSolve
(self, M=Minimize, UseADMM=True, NumProcessors=0, Rho=1.0, MaxIters=250, EpsAbs=0.01, EpsRel=0.
paper code/solveCrossTime.py:108
↓ 6 callersMethod__VerifyNId
(self, NId)
paper code/solveCrossTime.py:506
↓ 5 callersFunctioncomputeF1_macro
computes the macro F1 score confusion matrix : requres permutation matching according to which matrix must be permuted
paper code/network_accuracy.py:232
↓ 5 callersFunctioncomputeF1_macro
computes the macro F1 score confusion matrix : requres permutation matching according to which matrix must be permuted
paper code/TICC.py:232
↓ 5 callersFunctionwriteValue
(sharedarr, index, nparr, length)
paper code/solveCrossTime.py:937
↓ 4 callersMethodSetEdgeConstraints
(self, SrcNId, DstNId, Constraints)
paper code/solveCrossTime.py:612
↓ 4 callersMethod__VerifyEdgeTup
(self, ETup)
paper code/solveCrossTime.py:569
↓ 3 callersMethodSetNodeConstraints
(self, NId, Constraints)
paper code/solveCrossTime.py:555
↓ 3 callersFunctionfind_matching
returns the perfect matching
paper code/car.py:118
↓ 3 callersFunctionfind_matching
returns the perfect matching
paper code/network_accuracy.py:125
↓ 3 callersFunctionfind_matching
returns the perfect matching
paper code/TICC.py:125
↓ 3 callersMethodpredict_clusters
Given the current trained model, predict clusters. If the cluster segmentation has not been optimized yet, than this will be part of
TICC_solver.py:375
↓ 2 callersMethod__ExtractVariableList
(self, Objective)
paper code/solveCrossTime.py:524
↓ 2 callersMethod__SolveADMM
(self, numProcessors, rho_param, maxIters, eps_abs, eps_rel, verbose)
paper code/solveCrossTime.py:276
↓ 2 callersMethod__UpdateAllVariables
(self, NId, Objective)
paper code/solveCrossTime.py:511
↓ 2 callersMethodij2symmetric
(self, i,j,size)
src/admm_solver.py:18
↓ 2 callersFunctionupdateClusters
Takes in LLE_node_vals matrix and computes the path that minimizes the total cost over the path Note the LLE's are negative of the true LLE's actu
paper code/car.py:76
↓ 1 callersMethodADMM_u
(self)
src/admm_solver.py:76
↓ 1 callersMethodADMM_x
(self)
src/admm_solver.py:36
↓ 1 callersMethodADMM_z
(self, index_penalty = 1)
src/admm_solver.py:43
↓ 1 callersMethodCheckConvergence
(self, z_old, e_abs, e_rel, verbose)
src/admm_solver.py:89
↓ 1 callersFunctionProx_logdet
(S, A, eta)
paper code/solveCrossTime.py:956
↓ 1 callersMethodProx_logdet
(self, S, A, eta)
src/admm_solver.py:29
↓ 1 callersMethod__CheckConvergence
(self, A, A_tr, x, z, z_old, u, rho, p, n, e_abs, e_rel, verbose)
paper code/solveCrossTime.py:458
↓ 1 callersMethod__ClusterGraph
(self,clusterSize)
paper code/solveCrossTime.py:790
↓ 1 callersMethod__SolveClusterADMM
(self,M,UseADMM,superNodes,numProcessors, rho_param, maxIters, eps_abs, eps_rel,v
paper code/solveCrossTime.py:169
↓ 1 callersFunctioncomputeBIC
empirical covariance and inverse_covariance should be dicts K is num clusters T is num samples
src/TICC_helper.py:183
↓ 1 callersFunctioncomputeNetworkAccuracy
Takes in the matching for the clusters takes the computed clusters computes the average F1 score over the network
paper code/network_accuracy.py:256
↓ 1 callersFunctioncompute_BIC_score
compute BIC score for the clusters
paper code/car.py:184
↓ 1 callersMethodcompute_f_score
(self, matching_EM, matching_GMM, matching_Kmeans, train_confusion_matrix_EM, train_co
TICC_solver.py:217
↓ 1 callersMethodcompute_matches
(self, train_confusion_matrix_EM, train_confusion_matrix_GMM, train_confusion_matrix_kmeans)
TICC_solver.py:236
↓ 1 callersFunctiongenInvCov
(size, low = 0.3 , upper = 0.6, portion = 0.2,symmetric = True)
paper code/generate_synthetic_data.py:29
↓ 1 callersFunctiongenRandInv
(size,low = 0.3, upper=0.6, portion = 0.2)
paper code/generate_synthetic_data.py:45
↓ 1 callersFunctiongenerate_inverse
(rand_seed)
paper code/generate_synthetic_data.py:27
↓ 1 callersFunctiongetTrainTestSplit
- m: number of observations - num_blocks: window_size + 1 - num_stacked: window_size Returns: - sorted list of training indices
src/TICC_helper.py:4
↓ 1 callersFunctionhex_to_rgb
Return (red, green, blue) for the color given as #rrggbb.
paper code/scalability_test.py:208
↓ 1 callersFunctionhex_to_rgb
Return (red, green, blue) for the color given as #rrggbb.
paper code/car.py:218
↓ 1 callersFunctionhex_to_rgb
Return (red, green, blue) values Input is hexadecimal color code: #rrggbb.
paper code/network_accuracy.py:301
↓ 1 callersFunctionhex_to_rgb
Return (red, green, blue) values Input is hexadecimal color code: #rrggbb.
paper code/TICC.py:301
↓ 1 callersMethodload_data
(self, input_file)
TICC_solver.py:363
↓ 1 callersMethodlog_parameters
(self)
TICC_solver.py:369
↓ 1 callersMethodoptimize_clusters
(self, computed_covariance, len_train_clusters, log_det_values, optRes, train_cluster_inverse)
TICC_solver.py:293
↓ 1 callersMethodprepare_out_directory
(self)
TICC_solver.py:351
↓ 1 callersMethodsmoothen_clusters
(self, cluster_mean_info, computed_covariance, cluster_mean_stacked_info, complete_D
TICC_solver.py:264
↓ 1 callersMethodstack_training_data
(self, Data, n, num_train_points, training_indices)
TICC_solver.py:342
↓ 1 callersMethodtrain_clusters
(self, cluster_mean_info, cluster_mean_stacked_info, complete_D_train, empirical_covariances,
TICC_solver.py:312
↓ 1 callersFunctionupdateClusters
Takes in LLE_node_vals matrix and computes the path that minimizes the total cost over the path Note the LLE's are negative of the true LLE's actu
paper code/scalability_test.py:65
↓ 1 callersFunctionupdateClusters
Uses the Viterbi path dynamic programming algorithm to compute the optimal cluster assigments Takes in LLE_node_vals matrix and computes the path
paper code/network_accuracy.py:82
↓ 1 callersFunctionupdateClusters
Uses the Viterbi path dynamic programming algorithm to compute the optimal cluster assigments Takes in LLE_node_vals matrix and computes the path
paper code/TICC.py:82
↓ 1 callersFunctionupper2Full
(a, eps = 0)
paper code/scalability_test.py:54
↓ 1 callersFunctionupper2Full
(a, eps = 0)
paper code/car.py:65
↓ 1 callersFunctionupper2Full
(a)
paper code/solveCrossTime.py:965
↓ 1 callersFunctionupper2Full
(a, eps = 0)
paper code/network_accuracy.py:71
↓ 1 callersFunctionupper2Full
(a, eps = 0)
paper code/runTest.py:9
↓ 1 callersFunctionupper2Full
(a, eps = 0)
paper code/TICC.py:71
↓ 1 callersMethodupper2Full
(self, a)
src/admm_solver.py:21
↓ 1 callersFunctionupperToFull
(a, eps=0)
src/TICC_helper.py:29
↓ 1 callersFunctionvisualize
Takes in a np.matrix of size m x n and Visualizes the cross time correlation structure stores the plot using file_name passed in @params m : n
paper code/Visualization_function.py:4
↓ 1 callersMethodwrite_plot
(self, clustered_points, str_NULL, training_indices)
TICC_solver.py:253
FunctionADMM_u
(entry)
paper code/solveCrossTime.py:1116
FunctionADMM_x
(entry)
paper code/solveCrossTime.py:977
FunctionADMM_z
(entry, index_penalty = 1)
paper code/solveCrossTime.py:1015
MethodAddEdgeObjectives
(self, ObjFunc, Filename=None, EdgeIDs=None,\ SrcIdCol=None, DstIdCol=None)
paper code/solveCrossTime.py:712
MethodAddNodeObjectives
(self, Filename, ObjFunc, NodeIDs=None, IdCol=None)
paper code/solveCrossTime.py:645
MethodGetEdgeConstraints
(self, SrcNId, DstNId)
paper code/solveCrossTime.py:617
MethodGetEdgeObjective
(self, SrcNId, DstNId)
paper code/solveCrossTime.py:607
MethodGetNodeConstraints
(self, NId)
paper code/solveCrossTime.py:559
MethodGetNodeObjective
(self, NId)
paper code/solveCrossTime.py:551
MethodGetTotalProblemValue
(self)
paper code/solveCrossTime.py:435
FunctionLoadEdgeList
(Filename)
paper code/solveCrossTime.py:15
MethodPrintSolution
(self, Filename=None)
paper code/solveCrossTime.py:490
FunctionSetRho
(Rho=None)
paper code/solveCrossTime.py:887
FunctionSetRhoUpdateFunc
(Func=None)
paper code/solveCrossTime.py:894
Method__call__
(self, maxIters, eps_abs, eps_rel, verbose)
src/admm_solver.py:109
Method__init__
Parameters: - window_size: size of the sliding window - number_of_clusters: number of clusters - lambda_p
TICC_solver.py:17
Method__init__
(self, lamb, num_stacked, size_blocks, rho, S, rho_update_func=None)
src/admm_solver.py:4
Method__init__
(self, Graph=None)
paper code/solveCrossTime.py:59
FunctioncomputeF1Score
computes the F1 scores and returns a list of values
paper code/car.py:138
FunctioncomputeF1Score_delete
computes the F1 scores and returns a list of values
src/TICC_helper.py:112
FunctioncomputeF1Score_delete
computes the F1 scores and returns a list of values
paper code/scalability_test.py:127
FunctioncomputeF1Score_delete
computes the F1 scores and returns a list of values
paper code/network_accuracy.py:145
FunctioncomputeF1Score_delete
computes the F1 scores and returns a list of values
paper code/TICC.py:145
FunctioncomputeF1_macro
computes the macro F1 score confusion matrix : requres permutation matching according to which matrix must be permuted
src/TICC_helper.py:159
FunctioncomputeF1_macro
computes the macro F1 score confusion matrix : requres permutation matching according to which matrix must be permuted
paper code/scalability_test.py:173
FunctioncomputeNetworkAccuracy
Takes in the matching for the clusters takes the computed clusters computes the average F1 score over the network
paper code/TICC.py:256
next →1–100 of 112, ranked by callers