module Ann_patterns: sig .. end
Patterns for neural networks.
This module allows to read pattern files, print patterns, read and normalize
network's inputs.
val read_patterns : int -> int -> string -> int list option -> (float array * float array) array
read_patterns dim_x dim_y filename columns reads a patterns file
and returns them as an array of couples (x,t) where
x is an input vector of dimension dim_x, and t is a target
vector of dimension dim_y.
The file filename should be made such that each vector x is followed
on the same line by the target vector t.
If columns is the option None, then all columns of the file are used
to build the input vectors. Otherwise, columns may be an option Some l
where l is a selection of columns tu use as input (list of integers,
first columns being indexed 0).
The target vectors are always the last dim_y columns.
val fprint_pattern : out_channel -> float array * float array -> unit
fprint_pattern ch (x,t) prints the input vector x and the
target vector t and channel ch.
val stats : (float array * 'a) array -> float array * float array
stats patterns returns the mean value and standard deviation of
the input vectors of patterns
val normalize : (float array * 'a) array -> (float array * 'a) array
normalize patterns normalizes the input vectors by removing the
average value and dividing by the standard deviation.
It returns a fresh array of patterns.
val read_inputs : int -> string -> int list option -> float array array
read_inputs dim_x filename columns reads a file of input vectors
of dimension dim_x, from a file filename. If columns is the
option None, then all columns of the file are used to build the input
vectors. Otherwise, columns may be an option Some l where l is a
selection of columns tu use as input (list of integers, first columns
being indexed 0).
The function returns an array of input vectors (as float arrays)