DFT-PRED: Make predictions for test cases using Dirichlet diffusion trees. Dft-pred prints the log of the predictive joint probability (or density) for the target values in each test case given. Target values and average log probabilities can be displayed as well. Usage: dft-pred options { log-file range } [ / test-inputs [ test-targets ] ] The final optional arguments give the source of inputs and targets for the cases to look at; they default to the test data specification in the first log file given. The Dirichlet diffusion trees to use for the predictions are taken from the records with the given ranges of indexes in the given log files. Generation of new data points from each of these trees is simulated to produce an approximation to the predictive distribution. An index range can have one of the forms "[low][:[high]][%mod]" or "[low][:[high]]+num", or one of these forms preceded by "@". When "@" is present, "low" and "high" are given in terms of cpu time, otherwise they are iteration numbers. When just "low" is given, only that index is used. If the colon is included, but "high" is not, the range extends to the highest index in the log file. The "mod" form allows iterations to be selected whose numbers are multiples of "mod", with the default being "mod" of one. The "num" form allows the total number of iterations used to be specified; they are distributed as evenly as possible within the specified range. Note that it is possible that the number of diffusion trees used in the end may not equal this number, if records with some indexes are missing. The 'options' argument consists of one or more of the following letters: t Display the target values for each case. r Use the raw form of the target values, before transformation. p Display the log of the joint probability (or probability density) of the true targets. Logs are to base e. b Suppress headings and averages - just bare numbers for each case. The numbers are printed in exponential format, to high precision. a Display only the average log probability (density), suppressing the results for individual cases. (Makes sense only if 'p' is specified as well, but not 't'.) The predictive distribution is calculated by a stratified Monte Carlo method. For each tree, paths directed to each of the training cases are simulated in turn, which diverge from the path to this training case at some random time. The location and time of this divergence define a Gaussian distribution. For real-valued data, the predictive distribution is approximated by an equal mixture of all these Gaussian distributions. For binary data, values from each Gaussian distribution are chosen at random, and used to define 0/1 probabilities that are mixed to produce the predictive distribution. These Monte Carlo estimates are found using a random number stream initialized by setting the seed to one at the start of the program. Accuracy can be increased by repeating the same log-file/range combination several times, effectively increasing the sample size use. The average log probability figure is accompanied by +- its standard error (as long as there is more than one test case). If only targets are to be displayed (no predictions), one may give just a single log file with no range. Otherwise, at least one diffusion tree must be specified. Copyright (c) 1995-2003 by Radford M. Neal