svd分解matlab代码-Code-for-ETLMSC:ETLMSC代码

  • j2_323096
    了解作者
  • 62.1MB
    文件大小
  • zip
    文件格式
  • 0
    收藏次数
  • VIP专享
    资源类型
  • 0
    下载次数
  • 2022-06-15 02:50
    上传日期
svd分解matlab代码
Code-for-ETLMSC-master.zip
  • Code-for-ETLMSC-master
  • ClusteringMeasure
  • compute_f.m
    502B
  • hungarian.m
    11.1KB
  • bestMap.m
    709B
  • compute_nmi.m
    1.5KB
  • Contingency.m
    563B
  • RandIndex.m
    1.4KB
  • Accuracy.m
    94B
  • ETLMSC_UCI_DIGIT.m
    2.7KB
  • UCI_3view.mat
    1.3MB
  • twist
  • Yi2X.m
    177B
  • X2Yi.m
    59B
  • ConfusionMatrices
  • compute_confusion_matrix.m
    714B
  • main.m
    1.5KB
  • rotateXLabels.m
    13.6KB
  • draw_cm.m
    1KB
  • wshrinkObj.m
    2.1KB
  • Trans_Faces.m
    122B
  • COIL20_3VIEWS.mat
    50.5MB
  • ETLMSC_BBC.m
    2.6KB
  • ETLMSC_COIL20.m
    2.8KB
  • code_coregspectral
  • construct_similarity.m
    272B
  • compute_f.m
    502B
  • baseline_spectral_onkernel.m
    2.2KB
  • EuDist2.m
    1.1KB
  • run_synth.m
    1.8KB
  • baseline_spectral.m
    2KB
  • optSigma.m
    131B
  • spectral_centroid_multiview.m
    5.1KB
  • compute_nmi.m
    1.5KB
  • Contingency.m
    563B
  • spectral_centroid.m
    9.1KB
  • RandIndex.m
    1.4KB
  • spectral_pairwise_multview.m
    6.5KB
  • baseline_spectral_onRW_acc.m
    2.3KB
  • spectral_centroid_multiview_onkernel.m
    5.2KB
  • spectral_mindis.m
    2.9KB
  • spectral_pairwise.m
    8.2KB
  • synth3views_2clusters.mat
    38KB
  • spectral_pairwise_multview_onkernel.m
    6.5KB
  • README.txt
    2KB
  • constructKernel.m
    2.5KB
  • spectral_mindis_onkernel.m
    2.9KB
  • run_synth.asv
    1.8KB
  • Readme.txt
    1.5KB
  • bbcsport_2view.mat
    582.7KB
  • LRR
  • inexact_alm_lrr_l21.m
    1.7KB
  • NormalizeData.m
    142B
  • inexact_alm_lrr_l1.m
    1.7KB
  • yale.mat
    9.8MB
  • demo_LRR.m
    709B
  • solve_lrr.m
    1.2KB
  • SpectralClustering.m
    1.2KB
  • exact_alm_lrr_l21v2.m
    2.7KB
  • exact_alm_lrr_l1v2.m
    2.8KB
  • solve_l1l2.m
    273B
  • demo.m
    1.2KB
  • .gitattributes
    66B
内容介绍
Paper: Coregularized multiview spectral clustering, NIPS, 2011. ================================================================= Code ======== run_synth.m: example code showing the usage of the two multiview approaches proposed in the paper baseline_spectral.m: single view spectral clustering using normalized Laplacian (symmetric) spectral_mindis.m: Two view spectral clustering of De Sa et al spectral_mindis_onkernel.m: Two view spectral clustering of De Sa et al (on similarity matrices) spectral_pairwise.m: Pairwise coregularized spectral clustering spectral_pairwise_multiview.m: Pairwise coregularized spectral clustering (more than two views) spectral_pairwise_multiview_onkernel.m: Pairwise coregularized spectral clustering (more than two views) on similarity matrices spectral_centroid.m: Centroid based coregularized spectral clustering spectral_centroid_multiview.m: Centroid based coregularized spectral clustering (more than two views) spectral_centroid_multiview_onkernel.m: Centroid based coregularized spectral clustering (more than two views) on similarity matrices Note ======= 1. All the codes use Gaussian kernel for construction of similarity matrix. 2. Pairwise spectral approach reports the clsutering performance on the first view in each iteration. If you have prior knowledge about which view is more informative about the clusters, use that as the first view. E.g. X{1} = X1; X{2} = X2; sigma(1) = sigma1; sigma(2) = sigma2; lambda = 0.5; numiter = 5; numClust = 2; num_views = 2; [F P R nmi avgent AR] = spectral_pairwise_multview(X,num_views,numClust,sigma,lambda,truth,numiter); The performance measures are reported on X{1} in each iteration in the above function call. If you want the performance to be reported on X{2} instead, please do as follows: X{1} = X2; X{2} = X1; sigma(1) = sigma2; sigma(2) = sigma1; lambda = 0.5; numiter = 5; numClust = 2; num_views = 2; [F P R nmi avgent AR] = spectral_pairwise_multview(X,num_views,numClust,sigma,lambda,truth,numiter); 3. Performance can be sensitive to the Lambda value chosen.
评论
    相关推荐