/usr/include/ngram/ngram-relentropy.h is in libngram-dev 1.3.0-1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 | // Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
// Copyright 2005-2016 Brian Roark and Google, Inc.
// Stolcke relative entropy style model shrinking derived class.
#ifndef NGRAM_NGRAM_RELENTROPY_H_
#define NGRAM_NGRAM_RELENTROPY_H_
#include <ngram/ngram-shrink.h>
namespace ngram {
class NGramRelEntropy : public NGramShrink<StdArc> {
public:
// Constructs an NGramRelEntropy object that prunes an LM using relative
// entropy criterion.
NGramRelEntropy(StdMutableFst *infst, double theta, int shrink_opt = 0,
double tot_uni = -1.0, Label backoff_label = 0,
double norm_eps = kNormEps, bool check_consistency = false)
: NGramShrink<StdArc>(infst, shrink_opt, tot_uni, backoff_label, norm_eps,
check_consistency) {
// Threshold provided in real domain, convert to log
theta_ = log(theta + 1); // e^D - 1 <= theta_ -> D <= log(theta_ + 1)
}
// Shrink n-gram model, based on initialized parameters
bool ShrinkNGramModel() {
return NGramShrink<StdArc>::ShrinkNGramModel(true);
}
// Returns a theta that will yield the target number of ngrams and no more.
// In relative entropy shrinking, theta is initially in real domain, then
// converted to log domain for pruning. In this function we convert back
// from log domain to real domain for the threshold.
void CalculateTheta(int target_number_of_ngrams) {
theta_ = ThetaForMaxNGrams(target_number_of_ngrams);
}
protected:
// provide the pruning threshold
double GetTheta(StateId state) const override { return theta_; }
// Compute shrink score for transition based on Stolcke (KL) formula
// D(p||p') = -p(h) { p(w|h) [ log p(w|h') + log \alpha'(h) - log p(w|h) ] +
// \alpha_numerator(h) [ log \alpha'(h) - log \alpha (h) ] }
// return exp(D(p||p')) - 1
double ShrinkScore(const ShrinkStateStats &state,
const ShrinkArcStats &arc) const override {
if (arc.log_prob == -StdArc::Weight::Zero().Value() ||
state.log_prob == -StdArc::Weight::Zero().Value()) {
return -StdArc::Weight::Zero().Value();
}
double new_log_backoff = CalcNewLogBackoff(arc);
double score = arc.log_backoff_prob + new_log_backoff - arc.log_prob;
double secondterm =
new_log_backoff + (GetNLogBackoffNum() - GetNLogBackoffDenom());
secondterm *= exp(-GetNLogBackoffNum());
score *= exp(arc.log_prob);
score += secondterm;
score *= -exp(state.log_prob);
return score;
}
private:
double theta_; // Shrinking parameter
DISALLOW_COPY_AND_ASSIGN(NGramRelEntropy);
};
} // namespace ngram
#endif // NGRAM_NGRAM_RELENTROPY_H_
|