raffaello-camoriano/Enitor

Issues

• Remove filter hyperparameter cross validation logic from nystromUniformIncremental

  #36 opened 5 years ago by raffaello-camoriano
  0

• Decouple loss function from dataset

  #37 opened 5 years ago by raffaello-camoriano
  1

• Add time/memory speedup options to computeSqDistMat@gaussianKernel

  #38 opened 5 years ago by raffaello-camoriano
  0

• Add option for additional loss functions to dataset class

  #42 opened 5 years ago by raffaello-camoriano
  0

• obj.errorStorageMode in kigdesc and ksgdesc shoud be a flag, not a string.

  #45 opened 5 years ago by raffaello-camoriano
  0

• Datasets to be added

  #47 opened 5 years ago by raffaello-camoriano
  0

• set up error evaluation step in incRF and incNys

  #50 opened 5 years ago by raffaello-camoriano
  0

• Add option for rank-k update in randomFeaturesGaussianIncremental and nystromUniformIncremental

  #48 opened 5 years ago by raffaello-camoriano
  0

• Unify randomFeaturesGaussianIncremental and nystromUniformIncremental compute() method

  #49 opened 5 years ago by raffaello-camoriano
  0

• Fix Hessenberger version of Tikhonov

  #11 opened 5 years ago by raffaello-camoriano
  0

• Add configuration parameter for validation part in experiment class

  #19 opened 5 years ago by raffaello-camoriano
  0

• The 1st guess for m in incremental and batch Nystrom should never be 1

  #20 opened 5 years ago by raffaello-camoriano
  1

• remove the 'hasRealValues' bottleneck

  #21 opened 5 years ago by raffaello-camoriano
  1

• Add "strict" option to "horizonSharpStop" stopping rule

  #24 opened 5 years ago by raffaello-camoriano
  0

• Make the validationPart parameter explicit and easily accessible

  #26 opened 5 years ago by raffaello-camoriano
  0

• fix nystromUniform interface

  #31 opened 5 years ago by raffaello-camoriano
  0

• Add "formulation" tag to nrls algorithm

  #35 opened 5 years ago by raffaello-camoriano
  0

• Avoid matrix indexing in RFinc and NysInc

  #51 opened 5 years ago by raffaello-camoriano
  0

• Change name of filter class

  #52 opened 5 years ago by raffaello-camoriano
  0

• Make bandWaterfall plot function

  #54 opened 5 years ago by raffaello-camoriano
  0

• Add member functions to dataset class to convert between codings in multiclass scenarios

  #55 opened 5 years ago by raffaello-camoriano
  0

• Add RGB-D dataset with CaffeNet fc6 features

  #57 opened 5 years ago by raffaello-camoriano
  0

• Add Falcon

  #58 opened 5 years ago by raffaello-camoriano
  1

• Add mixUpTrainIdx and mixUpTestIdx to dataset to shuffle indexes without resampling them from scratch

  #56 opened 9 years ago by raffaello-camoriano
  0

• Do not recompute trainval, traintest and traintrain kernels from scratch at every filter iteration

  #53 opened 9 years ago by raffaello-camoriano
  1

• Incremental RF provides results which are incompatible with batch RF

  #34 opened 9 years ago by raffaello-camoriano
  2

• use try/catch for efficiency in SIsubGD_dual_hinge_loss && SsubGD_dual_hinge_loss

  #46 opened 9 years ago by raffaello-camoriano
  0

• In SGD, add option to compute CV performance only for the last step, instead of for every iteration

  #44 opened 9 years ago by raffaello-camoriano
  1

• Add cross-validation mechanism to SGD

  #41 opened 9 years ago by raffaello-camoriano
  1

• Restore the 'retraining' option

  #29 opened 9 years ago by raffaello-camoriano
  1

• Add perceptron loss

  #40 opened 9 years ago by raffaello-camoriano
  0

• Add full kernel precomputation option to kigdesc

  #43 opened 9 years ago by raffaello-camoriano
  1

• merge S-IGD filters

  #39 opened 9 years ago by raffaello-camoriano
  0

• add iteration timing option to incremental nystrom and incremental random features

  #33 opened 9 years ago by raffaello-camoriano
  1

• in incrementalNkrls, choosing numNysParGuesses == 1 provides unreliable results

  #30 opened 9 years ago by raffaello-camoriano
  1

• add minimum rank option to incremental nystrom and incremental random features

  #32 opened 9 years ago by raffaello-camoriano
  0

• Update the function signature of rfrls and ffrls

  #25 opened 9 years ago by raffaello-camoriano
  1

• fix mapParRangeSamples limit

  #28 opened 9 years ago by raffaello-camoriano
  0

• Possible bug in incrementalrfrls

  #27 opened 9 years ago by raffaello-camoriano
  1

• Add Comparison between maximum size KRLS on a part of the dataset and DACKRLS, incremental NKRLS and RFKRLS on the full dataset

  #18 opened 9 years ago by raffaello-camoriano
  0

• Add rebalancing methods

  #10 opened 9 years ago by raffaello-camoriano
  1

• add optional stopping rule to incrementalNKRLS

  #16 opened 9 years ago by raffaello-camoriano
  1

• add RFKRLS to Nystrom comparison experiment

  #17 opened 9 years ago by raffaello-camoriano
  1

• Add support for generating smaller datasets with a subset of classes in the constructor of Cifar10

  #22 opened 9 years ago by raffaello-camoriano
  1

• Fix incremental Nystrom stopping rule criterion

  #23 opened 9 years ago by raffaello-camoriano
  1

• Add full training, testing and validation performance matrix saving options to algorithms

  #14 opened 9 years ago by raffaello-camoriano
  1

• Adjust lambda range in experiments using Gaussian kernel

  #13 opened 9 years ago by raffaello-camoriano
  1

• Keep previously sampled columns in batch experiments

  #12 opened 9 years ago by raffaello-camoriano
  1

• inputParser not compatible with cell array input arguments

  #15 opened 9 years ago by raffaello-camoriano
  1

• retraining to be fixed in nrls

  #9 opened 10 years ago by raffaello-camoriano
  1