188 references to LastOccurrenceWins
Microsoft.ML.Data (76)
Commands\CrossValidationCommand.cs (13)
42[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for features", ShortName = "feat", SortOrder = 2)] 45[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab", SortOrder = 3)] 48[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for example weight", ShortName = "weight", SortOrder = 4)] 51[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for grouping", ShortName = "group", SortOrder = 5)] 57[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for stratification", ShortName = "strat", SortOrder = 7)] 60[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Columns with custom kinds declared through key assignments, for example, col[Kind]=Name to assign column named 'Name' kind 'Kind'", 64[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of folds in k-fold cross-validation", ShortName = "k")] 67[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Use threads", ShortName = "threads")] 70[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Normalize option for the feature column", ShortName = "norm")] 73[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether we should cache input training data", ShortName = "cache")] 86[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of instances to train the calibrator", ShortName = "numcali")] 89[Argument(ArgumentType.LastOccurrenceWins, HelpText = "File to save per-instance predictions and metrics to", 99[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether we should load predictor from input model and use it as the initial model state", ShortName = "cont")]
Commands\DataCommand.cs (1)
53[Argument(ArgumentType.LastOccurrenceWins, Visibility = ArgumentAttribute.VisibilityType.CmdLineOnly,
Commands\EvaluateCommand.cs (9)
126[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab", SortOrder = 3)] 129[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for example weight", ShortName = "weight", SortOrder = 4)] 132[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for grouping", ShortName = "group", SortOrder = 5)] 135[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Columns with custom kinds declared through key assignments, for example, col[Kind]=Name to assign column named 'Name' kind 'Kind'", 176[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab", SortOrder = 3)] 179[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for example weight", ShortName = "weight", SortOrder = 4)] 182[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for grouping", ShortName = "group", SortOrder = 5)] 188[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Columns with custom kinds declared through key assignments, for example, col[Kind]=Name to assign column named 'Name' kind 'Kind'", 198[Argument(ArgumentType.LastOccurrenceWins, HelpText = "File to save per-instance predictions and metrics to",
Commands\ScoreCommand.cs (2)
46[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for features when scorer is not defined", ShortName = "feat")] 63[Argument(ArgumentType.LastOccurrenceWins, HelpText = "File to save the data", ShortName = "dout")]
Commands\TestCommand.cs (4)
30[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab", SortOrder = 3)] 33[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for example weight", ShortName = "weight", SortOrder = 4)] 36[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for grouping", ShortName = "group", SortOrder = 5)] 42[Argument(ArgumentType.LastOccurrenceWins,
Commands\TrainCommand.cs (9)
43[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for features", ShortName = "feat", SortOrder = 2)] 46[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab", SortOrder = 3)] 49[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for example weight", ShortName = "weight", SortOrder = 4)] 52[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for grouping", ShortName = "group", SortOrder = 5)] 58[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Columns with custom kinds declared through key assignments, for example, col[Kind]=Name to assign column named 'Name' kind 'Kind'", 62[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Normalize option for the feature column", ShortName = "norm")] 74[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether we should cache input training data", ShortName = "cache")] 80[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of instances to train the calibrator", ShortName = "numcali")] 83[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether we should load predictor from input model and use it as the initial model state", ShortName = "cont")]
Commands\TrainTestCommand.cs (9)
40[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for features", ShortName = "feat", SortOrder = 2)] 43[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab", SortOrder = 3)] 46[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for example weight", ShortName = "weight", SortOrder = 4)] 49[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for grouping", ShortName = "group", SortOrder = 5)] 55[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Columns with custom kinds declared through key assignments, for example, col[Kind]=Name to assign column named 'Name' kind 'Kind'", 59[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Normalize option for the feature column", ShortName = "norm")] 65[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether we should cache input training data", ShortName = "cache")] 71[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of instances to train the calibrator", ShortName = "numcali")] 78[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether we should load predictor from input model and use it as the initial model state", ShortName = "cont")]
DataLoadSave\Binary\BinaryLoader.cs (2)
43[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The number of worker decompressor threads to use", ShortName = "t")] 48[Argument(ArgumentType.LastOccurrenceWins, HelpText = "When shuffling, the number of blocks worth of data to keep in the shuffle pool. " +
DataLoadSave\Binary\BinarySaver.cs (5)
36[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The compression scheme to use for the blocks", ShortName = "comp")] 39[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The block-size heuristic will choose no more than this many rows to have per block, can be set to null to indicate that there is no inherent limit", ShortName = "rpb")] 42[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The block-size heuristic will attempt to have about this many bytes across all columns per block, can be set to null to accept the indicated max-rows-per-block as the number of rows per block", ShortName = "bpb")] 45[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If true, this forces a deterministic block order during writing", ShortName = "det")] 48[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Suppress any info output (not warnings or errors)", Hide = true)]
DataLoadSave\Text\TextSaver.cs (1)
44[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Suppress any info output (not warnings or errors)", Hide = true)]
DataLoadSave\Transpose\TransposeLoader.cs (1)
38[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The number of worker decompresser threads to use", ShortName = "t")]
DataLoadSave\Transpose\TransposeSaver.cs (2)
35[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Write a copy of the data in row-wise format, in addition to the transposed data. This will increase performance for mixed applications while taking significantly more space.", ShortName = "row")] 38[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Suppress any info output (not warnings or errors)", Hide = true)]
Dirty\ChooseColumnsByIndexTransform.cs (1)
30[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If true, selected columns are dropped instead of kept, with the order of kept columns being the same as the original", ShortName = "d")]
Evaluators\QuantileRegressionEvaluator.cs (1)
468[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Quantile index to select", ShortName = "ind")]
Prediction\Calibrator.cs (2)
1637[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The slope parameter of f(x) = 1 / (1 + exp(slope * x + offset)", ShortName = "a")] 1640[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The offset parameter of f(x) = 1 / (1 + exp(slope * x + offset)", ShortName = "b")]
Training\TrainerInputBase.cs (1)
48[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether trainer should cache input training data", ShortName = "cache", SortOrder = 6, Visibility = ArgumentAttribute.VisibilityType.EntryPointsOnly)]
Transforms\BootstrapSamplingTransformer.cs (1)
50[Argument(ArgumentType.LastOccurrenceWins, HelpText = "When shuffling the output, the number of output rows to keep in that pool. Note that shuffling of output is completely distinct from shuffling of input.", ShortName = "pool")]
Transforms\RowShufflingTransformer.cs (5)
46[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The pool will have this many rows", ShortName = "rows")] 50[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If true, the transform will not attempt to shuffle the input cursor but only shuffle based on the pool. This parameter has no effect if the input data was not itself shufflable.", ShortName = "po")] 53[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If true, the transform will always provide a shuffled view.", ShortName = "force")] 56[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If true, the transform will always shuffle the input. The default value is the same as forceShuffle.", ShortName = "forceSource")] 59[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The random seed to use for forced shuffling.", ShortName = "seed")]
Transforms\TrainAndScoreTransformer.cs (6)
27[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for features when scorer is not defined", 111[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for features when scorer is not defined", 115[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab", SortOrder = 103, 119[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for grouping", ShortName = "group", 123[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for example weight", ShortName = "weight", 155[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of instances to train the calibrator", ShortName = "numcali")]
Utils\LossFunctions.cs (1)
582[Argument(ArgumentType.LastOccurrenceWins, HelpText =
Microsoft.ML.FastTree (59)
FastTreeArguments.cs (46)
69[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Option for using derivatives optimized for unbalanced sets", ShortName = "us")] 172[Argument(ArgumentType.LastOccurrenceWins, HelpText = 227[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Comma-separated list of gains associated to each relevance label.", ShortName = "gains")] 234[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Train DCG instead of NDCG", ShortName = "dcg")] 239[Argument(ArgumentType.LastOccurrenceWins, 255[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Use shifted NDCG", Hide = true)] 265[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Distance weight 2 adjustment to cost", ShortName = "dw", Hide = true)] 270[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Normalize query lambdas", ShortName = "nql", Hide = true)] 359[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The number of threads to use", ShortName = "t", NullName = "<Auto>")] 372[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The seed of the random number generator", ShortName = "r1")] 379[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The seed of the active feature selection", ShortName = "r3", Hide = true)] 386[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The entropy (regularization) coefficient between 0 and 1", ShortName = "e")] 393[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The number of histograms in the pool (between 2 and numLeaves)", ShortName = "ps")] 399[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether to utilize the disk or the data's native transposition facilities (where applicable) when performing the transpose", ShortName = "dt")] 405[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether to collectivize features during dataset preparation to speed up training", ShortName = "flocks", Hide = true)] 411[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether to do split based on multiple categorical feature values.", ShortName = "cat")] 417[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Maximum categorical split groups to consider when splitting on a categorical feature. " + 425[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Maximum categorical split points to consider when splitting on a categorical feature.", ShortName = "maxcat")] 431[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Minimum categorical example percentage in a bin to consider for a split.", ShortName = "mdop")] 437[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Minimum categorical example count in a bin to consider for a split.", ShortName = "mdo")] 443[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Bias for calculating gradient for each feature bin for a categorical feature.", ShortName = "bias")] 449[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Bundle low population bins. " + 460[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Maximum number of distinct values (bins) per feature", ShortName = "mb")] 466[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Sparsity level needed to use sparse feature representation", ShortName = "sp")] 472[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The feature first use penalty coefficient", ShortName = "ffup")] 478[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The feature re-use penalty (regularization) coefficient", ShortName = "frup")] 488[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Tree fitting gain confidence requirement (should be in the range [0,1) ).", ShortName = "gainconf")] 513[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The max number of leaves in each regression tree", ShortName = "nl", SortOrder = 2)] 523[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The minimal number of examples allowed in a leaf of a regression tree, out of the subsampled data", ShortName = "mil", SortOrder = 3)] 532[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Total number of decision trees to create in the ensemble", ShortName = "iter", SortOrder = 1)] 583[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The level of feature compression to use", ShortName = "fcomp", Hide = true)] 590[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Compress the tree Ensemble", ShortName = "cmp", Hide = true)] 598[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Print metrics graph for the first test set", ShortName = "graph", Hide = true)] 607[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Print Train and Validation metrics in graph", ShortName = "graphtv", Hide = true)] 614[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Calculate metric values for train/valid/test every k rounds", ShortName = "tf")] 655[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Option for using best regression step trees", ShortName = "bsr")] 661[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Should we use line search for a step size", ShortName = "ls")] 667[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of post-bracket line search steps", ShortName = "lssteps")] 673[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Minimum line search step size", ShortName = "minstep")] 687[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Optimization algorithm to be used (GradientDescent, AcceleratedGradientDescent)", ShortName = "oa")] 734[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Use window and tolerance for pruning", ShortName = "prtol")] 754[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The learning rate", ShortName = "lr", SortOrder = 4)] 796[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Training starts from random ordering (determined by /r1)", ShortName = "rs", Hide = true)] 826[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Freeform defining the scores that should be used as the baseline ranker", ShortName = "basescores", Hide = true)] 834[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Baseline alpha for tradeoffs of risk (0 is normal training)", ShortName = "basealpha", Hide = true)] 842[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The discount freeform which specifies the per position discounts of examples in a query (uses a single variable P for position where P=0 is first position)",
GamClassification.cs (1)
69[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Should we use derivatives optimized for unbalanced sets", ShortName = "us")]
GamTrainer.cs (10)
40[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The entropy (regularization) coefficient between 0 and 1", ShortName = "e")] 50[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Tree fitting gain confidence requirement (should be in the range [0,1) ).", ShortName = "gainconf")] 56[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Total number of iterations over all features", ShortName = "iter", SortOrder = 1)] 64[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The number of threads to use", ShortName = "t", NullName = "<Auto>")] 70[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The learning rate", ShortName = "lr", SortOrder = 4)] 78[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether to utilize the disk or the data's native transposition facilities (where applicable) when performing the transpose", ShortName = "dt")] 84[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Maximum number of distinct values (bins) per feature", ShortName = "mb")] 102[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The seed of the random number generator", ShortName = "r1")] 108[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Minimum number of training instances required to form a partition", ShortName = "mi", SortOrder = 3)] 116[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether to collectivize features during dataset preparation to speed up training", ShortName = "flocks", Hide = true)]
RandomForestRegression.cs (1)
305[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Shuffle the labels on every iteration. " +
SumupPerformanceCommand.cs (1)
64[Argument(ArgumentType.LastOccurrenceWins,
Microsoft.ML.Maml (1)
HelpCommand.cs (1)
57[Argument(ArgumentType.LastOccurrenceWins, Hide = true, SignatureType = typeof(SignatureModuleGenerator))]
Microsoft.ML.Mkl.Components (1)
OlsLinearRegression.cs (1)
90[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether to calculate per parameter significance statistics", ShortName = "sig")]
Microsoft.ML.Parquet (2)
ParquetLoader.cs (2)
80[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of column chunk values to cache while reading from parquet file", ShortName = "chunkSize")] 83[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If true, will read large numbers as dates", ShortName = "bigIntDates")]
Microsoft.ML.Predictor.Tests (2)
CmdLine\CmdLine.cs (1)
183[Argument(ArgumentType.LastOccurrenceWins)]
CmdLine\CmdLineReverseTest.cs (1)
111[Argument(ArgumentType.LastOccurrenceWins)]
Microsoft.ML.StandardTrainers (2)
Standard\LogisticRegression\LbfgsPredictorBase.cs (1)
112[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Init weights diameter", ShortName = "initwts, InitWtsDiameter", SortOrder = 140)]
Standard\MulticlassClassification\MetaMulticlassTrainer.cs (1)
30[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of instances to train the calibrator", SortOrder = 150, ShortName = "numcali")]
Microsoft.ML.Sweeper (32)
Algorithms\Grid.cs (2)
35[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of tries to generate distinct parameter sets.", ShortName = "r")] 116[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Limit for the number of combinations to generate the entire grid.", ShortName = "maxpoints")]
Algorithms\KdoSweeper.cs (9)
47[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If iteration point is outside parameter definitions, should it be projected?", ShortName = "project")] 50[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of points to use for random initialization", ShortName = "nip")] 53[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Minimum mutation spread", ShortName = "mms")] 56[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Maximum length of history to retain", ShortName = "hlen")] 59[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If true, draws samples from independent Beta distributions, rather than multivariate Gaussian", ShortName = "beta")] 62[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If true, uses simpler mutation and concentration model")] 65[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Proportion of trials, between 0 and 1, that are uniform random draws", ShortName = "prand")] 68[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Maximum power for rescaling (the larger the number, the stronger the exploitation of good points)", ShortName = "wrp")] 74[Argument(ArgumentType.LastOccurrenceWins, HelpText = "(Deprecated) Use legacy discrete parameter behavior.", ShortName = "legacy", Hide = true, Visibility = ArgumentAttribute.VisibilityType.CmdLineOnly)]
Algorithms\NelderMead.cs (8)
26[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The sweeper used to get the initial results.", ShortName = "init", SignatureType = typeof(SignatureSweeperFromParameterList))] 32[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Simplex diameter for stopping", ShortName = "dstop")] 35[Argument(ArgumentType.LastOccurrenceWins, 40[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Reflection parameter", ShortName = "dr")] 43[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Expansion parameter", ShortName = "de")] 46[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Inside contraction parameter", ShortName = "dic")] 49[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Outside contraction parameter", ShortName = "doc")] 52[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Shrinkage parameter", ShortName = "ds")]
Algorithms\SmacSweeper.cs (9)
34[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If iteration point is outside parameter definitions, should it be projected?", ShortName = "project")] 37[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of regression trees in forest", ShortName = "numtrees")] 40[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Minimum number of data points required to be in a node if it is to be split further", ShortName = "nmin")] 43[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of points to use for random initialization", ShortName = "nip")] 46[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of search parents to use for local search in maximizing EI acquisition function", ShortName = "lsp")] 49[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of random configurations when maximizing EI acquisition function", ShortName = "nrcan")] 52[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Fraction of eligible dimensions to split on (i.e., split ratio)", ShortName = "sr")] 55[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Epsilon threshold for ending local searches", ShortName = "eps")] 58[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of neighbors to sample for locally searching each numerical parameter", ShortName = "nnnp")]
Parameters.cs (3)
35[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Number of steps for grid runthrough.", ShortName = "steps")] 38[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Amount of increment between steps (multiplicative if log).", ShortName = "inc")] 41[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Log scale.", ShortName = "log")]
SweepResultEvaluator.cs (1)
23[Argument(ArgumentType.LastOccurrenceWins, HelpText = "The sweeper used to get the initial results.", ShortName = "m")]
Microsoft.ML.Transforms (13)
LearnerFeatureSelection.cs (8)
32[Argument(ArgumentType.LastOccurrenceWins, HelpText = "If the corresponding absolute value of the weight for a slot is greater than this threshold, the slot is preserved", ShortName = "ft", SortOrder = 2)] 46[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for features", ShortName = "feat,col", SortOrder = 3, Purpose = SpecialPurpose.ColumnName)] 49[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab", SortOrder = 4, Purpose = SpecialPurpose.ColumnName)] 52[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for example weight", ShortName = "weight", SortOrder = 5, Purpose = SpecialPurpose.ColumnName)] 55[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for grouping", ShortName = "group", Purpose = SpecialPurpose.ColumnName)] 61[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Columns with custom kinds declared through key assignments, for example, col[Kind]=Name to assign column named 'Name' kind 'Kind'", 65[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Normalize option for the feature column", ShortName = "norm")] 68[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Whether we should cache input training data", ShortName = "cache")]
MutualInformationFeatureSelection.cs (1)
94[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Column to use for labels", ShortName = "lab",
RandomFourierFeaturizing.cs (2)
50[Argument(ArgumentType.LastOccurrenceWins, 67[Argument(ArgumentType.LastOccurrenceWins,
SvmLight\SvmLightSaver.cs (2)
31[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Write the variant of SVM-light format where feature indices start from 0, not 1", ShortName = "z")] 34[Argument(ArgumentType.LastOccurrenceWins, HelpText = "Format output labels for a binary classification problem (-1 for negative, 1 for positive)", ShortName = "b")]