Last active
October 7, 2015 13:43
-
-
Save igorw/ba5e4c9562d19b89ef56 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
<?php | |
$n = 6; | |
$rules = []; | |
$ngrams = []; | |
foreach (read_csv('tweets.csv') as $data) { | |
$source = $data['text']; | |
$source = htmlspecialchars_decode($source); | |
$source = preg_replace('#http.*(…|\s|$)#', ' ', $source); | |
$source = preg_replace('/\s+/', ' ', $source); | |
$source = trim($source); | |
if ($source === '' || strlen($source) < 2*$n) { | |
continue; | |
} | |
foreach (range($n, strlen($source)-4) as $i) { | |
$ngram_a = substr($source, $i - $n, $n); | |
$ngram_b = substr($source, $i, $n); | |
if (!isset($rules[$ngram_a][$ngram_b])) { | |
$rules[$ngram_a][$ngram_b] = 0; | |
} | |
$rules[$ngram_a][$ngram_b]++; | |
} | |
$ngrams[] = substr($source, 0, $n); | |
foreach (str_split($source) as $i => $char) { | |
$ngram = substr($source, $i + 1, $n); | |
if ($char === ' ' && strlen($ngram) === $n) { | |
$ngrams[] = substr($source, $i + 1, $n); | |
} | |
} | |
} | |
// a few passes of removing states that do not loop back | |
// because we don't want to implement mark and sweep-ish | |
// ugh, this is super slow | |
foreach (range(1, 3) as $i) { | |
foreach ($rules as $state => $new_states) { | |
foreach ($new_states as $ngram => $weight) { | |
if (!isset($rules[$ngram])) { | |
unset($rules[$state][$ngram]); | |
} | |
} | |
if (count($rules[$state]) === 0) { | |
unset($rules[$state]); | |
} | |
} | |
} | |
while (true) { | |
$state = array_pick($ngrams); | |
foreach (range(1, 50) as $i) { | |
echo $state; | |
if (!isset($rules[$state])) { | |
// no transition found for state | |
break; | |
} | |
$state = weighted_pick($rules[$state]); | |
} | |
echo "\n\n"; | |
sleep(2); | |
} | |
function array_pick($values) { | |
return $values[array_rand($values)]; | |
} | |
function weighted_pick(array $weighted_values) { | |
$rand = mt_rand(1, (int) array_sum($weighted_values)); | |
foreach ($weighted_values as $ngram => $weight) { | |
$rand -= $weight; | |
if ($rand <= 0) { | |
return $ngram; | |
} | |
} | |
} | |
function read_csv($filename) { | |
$csv = new SplFileObject($filename); | |
$cols = $csv->fgetcsv(); | |
while (!$csv->eof() && [null] !== ($row = $csv->fgetcsv())) { | |
$data = array_combine($cols, $row); | |
yield $data; | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment