AI-MaxEntropy
view release on metacpan or search on metacpan
my $me3 = AI::MaxEntropy->new(
algorithm => { m => 7, epsilon => 1e-4 },
smoother => { type => 'gaussian', sigma => 0.8 }
);
see
Let the Maximum Entropy learner see a sample.
my $me = AI::MaxEntropy->new;
# see a sample with default weight 1.0
$me->see(['red', 'round'] => 'apple');
# see a sample with specified weight 0.5
$me->see(['yellow', 'long'] => 'banana' => 0.5);
The sample can be also represented in the attribute-value form, which
like
$me->see({color => 'yellow', shape => 'long'} => 'banana');
$me->see({color => ['red', 'green'], shape => 'round'} => 'apple');
param_n => ...
}
type
The entry "type => ..." specifies which algorithm is used for the
optimization. Valid values include:
'lbfgs' Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS)
'gis' General Iterative Scaling (GIS)
If ommited, 'lbfgs' is used by default.
progress_cb
The entry "progress_cb => ..." specifies the progress callback
subroutine which is used to trace the process of the algorithm. The
specified callback routine will be called at each iteration of the
algorithm.
For L-BFGS, "progress_cb" will be directly passed to "fmin" in
Algorithm::LBFGS. f(x) is the negative log-likelihood of current lambda
vector.
progress_cb(i, lambda, d_lambda, lambda_norm, d_lambda_norm)
"i" is the number of the iterations, "lambda" is an array ref containing
the current lambda vector, "d_lambda" is an array ref containing the
delta of the lambda vector in current iteration, "lambda_norm" and
"d_lambda_norm" are Euclid norms of "lambda" and "d_lambda"
respectively.
For both L-BFGS and GIS, the client program can also pass a string
'verbose' to "progress_cb" to use a default progress callback which
simply print out the progress on the screen.
"progress_cb" can also be omitted if the client program do not want to
trace the progress.
parameters
The rest entries are parameters for the specified algorithm. Each
parameter will be assigned with its default value when it is not given
explicitly.
For L-BFGS, the parameters will be directly passed to Algorithm::LBFGS
object, please refer to "Parameters" in Algorithm::LBFGS for details.
For GIS, there is only one parameter "epsilon", which controls the
precision of the algorithm (similar to the "epsilon" in
Algorithm::LBFGS). Generally speaking, a smaller "epsilon" produces a
more precise result. The default value of "epsilon" is 1e-3.
smoother
The smoother is a solution to the over-fitting problem. This property
chooses which type of smoother the client program want to apply and sets
the smoothing parameters.
Only one smoother have been implemented in this version of the module,
the Gaussian smoother.
One can apply the Gaussian smoother as following,
inc/Module/AutoInstall.pm view on Meta::CPAN
'' => 'Core Features', # XXX: deprecated
'-core' => 'Core Features',
);
# various lexical flags
my ( @Missing, @Existing, %DisabledTests, $UnderCPAN, $HasCPANPLUS );
my ( $Config, $CheckOnly, $SkipInstall, $AcceptDefault, $TestOnly );
my ( $PostambleActions, $PostambleUsed );
# See if it's a testing or non-interactive session
_accept_default( $ENV{AUTOMATED_TESTING} or ! -t STDIN );
_init();
sub _accept_default {
$AcceptDefault = shift;
}
sub missing_modules {
return @Missing;
}
sub do_install {
__PACKAGE__->install(
[
inc/Module/AutoInstall.pm view on Meta::CPAN
)
)
{
if ( $arg =~ /^--config=(.*)$/ ) {
$Config = [ split( ',', $1 ) ];
}
elsif ( $arg =~ /^--installdeps=(.*)$/ ) {
__PACKAGE__->install( $Config, @Missing = split( /,/, $1 ) );
exit 0;
}
elsif ( $arg =~ /^--default(?:deps)?$/ ) {
$AcceptDefault = 1;
}
elsif ( $arg =~ /^--check(?:deps)?$/ ) {
$CheckOnly = 1;
}
elsif ( $arg =~ /^--skip(?:deps)?$/ ) {
$SkipInstall = 1;
}
elsif ( $arg =~ /^--test(?:only)?$/ ) {
$TestOnly = 1;
}
}
}
# overrides MakeMaker's prompt() to automatically accept the default choice
sub _prompt {
goto &ExtUtils::MakeMaker::prompt unless $AcceptDefault;
my ( $prompt, $default ) = @_;
my $y = ( $default =~ /^[Yy]/ );
print $prompt, ' [', ( $y ? 'Y' : 'y' ), '/', ( $y ? 'n' : 'N' ), '] ';
print "$default\n";
return $default;
}
# the workhorse
sub import {
my $class = shift;
my @args = @_ or return;
my $core_all;
print "*** $class version " . $class->VERSION . "\n";
print "*** Checking for Perl dependencies...\n";
inc/Module/AutoInstall.pm view on Meta::CPAN
? ( ( ref($_) eq 'HASH' ) ? keys(%$_) : @{$_} )
: ''
}
map { +{@args}->{$_} }
grep { /^[^\-]/ or /^-core$/i } keys %{ +{@args} }
)[0]
);
while ( my ( $feature, $modules ) = splice( @args, 0, 2 ) ) {
my ( @required, @tests, @skiptests );
my $default = 1;
my $conflict = 0;
if ( $feature =~ m/^-(\w+)$/ ) {
my $option = lc($1);
# check for a newer version of myself
_update_to( $modules, @_ ) and return if $option eq 'version';
# sets CPAN configuration options
$Config = $modules if $option eq 'config';
inc/Module/AutoInstall.pm view on Meta::CPAN
$core_all = ( $modules =~ /^all$/i ) and next
if $option eq 'core';
next unless $option eq 'core';
}
print "[" . ( $FeatureMap{ lc($feature) } || $feature ) . "]\n";
$modules = [ %{$modules} ] if UNIVERSAL::isa( $modules, 'HASH' );
unshift @$modules, -default => &{ shift(@$modules) }
if ( ref( $modules->[0] ) eq 'CODE' ); # XXX: bugward combatability
while ( my ( $mod, $arg ) = splice( @$modules, 0, 2 ) ) {
if ( $mod =~ m/^-(\w+)$/ ) {
my $option = lc($1);
$default = $arg if ( $option eq 'default' );
$conflict = $arg if ( $option eq 'conflict' );
@tests = @{$arg} if ( $option eq 'tests' );
@skiptests = @{$arg} if ( $option eq 'skiptests' );
next;
}
printf( "- %-${maxlen}s ...", $mod );
if ( $arg and $arg =~ /^\D/ ) {
inc/Module/AutoInstall.pm view on Meta::CPAN
if (
!$SkipInstall
and (
$CheckOnly
or _prompt(
qq{==> Auto-install the }
. ( @required / 2 )
. ( $mandatory ? ' mandatory' : ' optional' )
. qq{ module(s) from CPAN?},
$default ? 'y' : 'n',
) =~ /^[Yy]/
)
)
{
push( @Missing, @required );
$DisabledTests{$_} = 1 for map { glob($_) } @skiptests;
}
elsif ( !$SkipInstall
and $default
and $mandatory
and
_prompt( qq{==> The module(s) are mandatory! Really skip?}, 'n', )
=~ /^[Nn]/ )
{
push( @Missing, @required );
$DisabledTests{$_} = 1 for map { glob($_) } @skiptests;
}
else {
inc/Module/Install/Makefile.pm view on Meta::CPAN
sub prompt {
shift;
# Infinite loop protection
my @c = caller();
if ( ++$seen{"$c[1]|$c[2]|$_[0]"} > 3 ) {
die "Caught an potential prompt infinite loop ($c[1]|$c[2]|$_[0])";
}
# In automated testing, always use defaults
if ( $ENV{AUTOMATED_TESTING} and ! $ENV{PERL_MM_USE_DEFAULT} ) {
local $ENV{PERL_MM_USE_DEFAULT} = 1;
goto &ExtUtils::MakeMaker::prompt;
} else {
goto &ExtUtils::MakeMaker::prompt;
}
}
sub makemaker_args {
my $self = shift;
lib/AI/MaxEntropy.pm view on Meta::CPAN
}
sub _free_cache {
my $self = shift;
$self->_free_cache_samples;
$self->_free_cache_f_map;
}
sub learn {
my $self = shift;
# cut 0 for default
$self->cut(0) if $self->{last_cut} == -1;
# initialize
$self->{lambda} = [map { 0 } (1 .. $self->{f_num})];
$self->_cache;
# optimize
my $type = $self->{algorithm}->{type} || 'lbfgs';
if ($type eq 'lbfgs') {
my $o = Algorithm::LBFGS->new(%{$self->{algorithm}});
$o->fmin(\&_neg_log_likelihood, $self->{lambda},
$self->{algorithm}->{progress_cb}, $self);
lib/AI/MaxEntropy.pm view on Meta::CPAN
algorithm => { m => 7, epsilon => 1e-4 },
smoother => { type => 'gaussian', sigma => 0.8 }
);
=head2 see
Let the Maximum Entropy learner see a sample.
my $me = AI::MaxEntropy->new;
# see a sample with default weight 1.0
$me->see(['red', 'round'] => 'apple');
# see a sample with specified weight 0.5
$me->see(['yellow', 'long'] => 'banana' => 0.5);
The sample can be also represented in the attribute-value form, which like
$me->see({color => 'yellow', shape => 'long'} => 'banana');
$me->see({color => ['red', 'green'], shape => 'round'} => 'apple');
lib/AI/MaxEntropy.pm view on Meta::CPAN
}
=head3 type
The entry C<type =E<gt> ...> specifies which algorithm is used for the
optimization. Valid values include:
'lbfgs' Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS)
'gis' General Iterative Scaling (GIS)
If ommited, C<'lbfgs'> is used by default.
=head3 progress_cb
The entry C<progress_cb =E<gt> ...> specifies the progress callback
subroutine which is used to trace the process of the algorithm.
The specified callback routine will be called at each iteration of the
algorithm.
For L-BFGS, C<progress_cb> will be directly passed to
L<Algorithm::LBFGS/fmin>. C<f(x)> is the negative log-likelihood of current
lib/AI/MaxEntropy.pm view on Meta::CPAN
For GIS, the C<progress_cb> is supposed to have a prototype like
progress_cb(i, lambda, d_lambda, lambda_norm, d_lambda_norm)
C<i> is the number of the iterations, C<lambda> is an array ref containing
the current lambda vector, C<d_lambda> is an array ref containing the
delta of the lambda vector in current iteration, C<lambda_norm> and
C<d_lambda_norm> are Euclid norms of C<lambda> and C<d_lambda> respectively.
For both L-BFGS and GIS, the client program can also pass a string
C<'verbose'> to C<progress_cb> to use a default progress callback
which simply print out the progress on the screen.
C<progress_cb> can also be omitted if the client program
do not want to trace the progress.
=head3 parameters
The rest entries are parameters for the specified algorithm.
Each parameter will be assigned with its default value when it is not
given explicitly.
For L-BFGS, the parameters will be directly passed to
L<Algorithm::LBFGS> object, please refer to L<Algorithm::LBFGS/Parameters>
for details.
For GIS, there is only one parameter C<epsilon>, which controls the
precision of the algorithm (similar to the C<epsilon> in
L<Algorithm::LBFGS>). Generally speaking, a smaller C<epsilon> produces
a more precise result. The default value of C<epsilon> is 1e-3.
=head2 smoother
The smoother is a solution to the over-fitting problem.
This property chooses which type of smoother the client program want to
apply and sets the smoothing parameters.
Only one smoother have been implemented in this version of the module,
the Gaussian smoother.
If this option is given, a single patch file will be created if
any changes are suggested. This requires a working diff program
to be installed on your system.
=head2 --copy=I<suffix>
If this option is given, a copy of each file will be saved with
the given suffix that contains the suggested changes. This does
not require any external programs.
If neither C<--patch> or C<--copy> are given, the default is to
simply print the diffs for each file. This requires either
C<Text::Diff> or a C<diff> program to be installed.
=head2 --diff=I<program>
Manually set the diff program and options to use. The default
is to use C<Text::Diff>, when installed, and output unified
context diffs.
=head2 --compat-version=I<version>
Tell F<ppport.h> to check for compatibility with the given
Perl version. The default is to check for compatibility with Perl
version 5.003. You can use this option to reduce the output
of F<ppport.h> if you intend to be backward compatible only
up to a certain Perl version.
=head2 --cplusplus
Usually, F<ppport.h> will detect C++ style comments and
replace them with C style comments for portability reasons.
Using this option instructs F<ppport.h> to leave C++
comments untouched.
Don't output any hints. Hints often contain useful portability
notes.
=head2 --nochanges
Don't suggest any changes. Only give diagnostic output and hints
unless these are also deactivated.
=head2 --nofilter
Don't filter the list of input files. By default, files not looking
like source code (i.e. not *.xs, *.c, *.cc, *.cpp or *.h) are skipped.
=head2 --list-provided
Lists the API elements for which compatibility is provided by
F<ppport.h>. Also lists if it must be explicitly requested,
if it has dependencies, and if there are hints for it.
=head2 --list-unsupported
sv_setpvf_mg_nocontext() NEED_sv_setpvf_mg_nocontext NEED_sv_setpvf_mg_nocontext_GLOBAL
vnewSVpvf() NEED_vnewSVpvf NEED_vnewSVpvf_GLOBAL
To avoid namespace conflicts, you can change the namespace of the
explicitly exported functions using the C<DPPP_NAMESPACE> macro.
Just C<#define> the macro before including C<ppport.h>:
#define DPPP_NAMESPACE MyOwnNamespace_
#include "ppport.h"
The default namespace is C<DPPP_>.
=back
The good thing is that most of the above can be checked by running
F<ppport.h> on your source code. See the next section for
details.
=head1 EXAMPLES
To verify whether F<ppport.h> is needed for your module, whether you
( run in 0.607 second using v1.01-cache-2.11-cpan-0a6323c29d9 )