Last active
March 19, 2024 01:34
-
-
Save renalpha/f37fd7ef7b8e3be904976ebd313d0e2b to your computer and use it in GitHub Desktop.
RedisVectorConnector.php
This file contains hidden or 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 | |
private function indexVector(SplFileInfo $file, string $content): void | |
{ | |
$metaData = $this->getMetaData($file, $content); | |
/** @var RedisVector $redisVectorService */ | |
$redisVectorService = app(RedisVector::class); | |
/** @var \Exdeliver\LaraPal\Connectors\VoyageAIEmbedding $voyage */ | |
$voyage = app(VoyageAIEmbedding::class); | |
$redisVectorService->store([ | |
'id' => Str::uuid()->toString(), | |
'embedding' => $voyage->generateEmbedding($content), | |
'metaData' => json_encode($metaData, JSON_THROW_ON_ERROR), | |
'file' => $file->getFilename(), | |
]); | |
} |
This file contains hidden or 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 | |
/** @var RedisVector $redisVector */ | |
$redisVector = app(RedisVector::class); | |
$searchResults = $redisVector->search($embedding, 5); |
This file contains hidden or 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 | |
namespace Exdeliver\LaraPal\Connectors; | |
use Redis; | |
final class RedisVector | |
{ | |
private ?string $index; | |
private Redis $redis; | |
public function __construct(?string $index = null) | |
{ | |
$this->index = $index ?? config('services.redis.index'); | |
$this->redis = new Redis(); | |
$this->redis->connect(config('database.redis.default.host'), config('database.redis.default.port')); | |
$this->redis->auth(config('database.redis.default.password')); | |
$this->redis->select(config('database.redis.default.database')); | |
} | |
public function index(): array | |
{ | |
$keys = $this->redis->keys("{$this->index}:*"); | |
$vectors = []; | |
foreach ($keys as $key) { | |
$vector = $this->redis->hGetAll($key); | |
$vectors[] = $vector; | |
} | |
return $vectors; | |
} | |
public function store($data): bool|array|Redis | |
{ | |
$namespace = config('services.redis.namespace'); | |
$key = "{$this->index}:{$namespace}:{$data['id']}"; | |
// Convert the embedding array to a string representation | |
$embeddingString = implode(',', $data['embedding']); | |
// Store the embedding string as the value for the 'embedding' field | |
$this->redis->hSet($key, 'embedding', $embeddingString); | |
// Store other fields in the hash, if any | |
foreach ($data as $field => $value) { | |
if ($field !== 'id' && $field !== 'embedding') { | |
$this->redis->hSet($key, $field, $value); | |
} | |
} | |
return $this->redis->hGetAll($key); | |
} | |
public function query(array $data): Redis|array|bool | |
{ | |
$namespace = $data['namespace'] ?? config('services.redis.namespace'); | |
$key = "{$this->index}:{$namespace}:{$data['id']}"; | |
return $this->redis->hGetAll($key); | |
} | |
public function empty($data = ['deleteAll' => true]): array | |
{ | |
$namespace = config('services.redis.namespace'); | |
if ($data['deleteAll']) { | |
$keys = $this->redis->keys("{$this->index}:{$namespace}:*"); | |
$this->redis->del($keys); | |
} else { | |
$key = "{$this->index}:{$namespace}:{$data['id']}"; | |
$this->redis->del($key); | |
} | |
return ['success' => true]; | |
} | |
public function search($queryEmbedding, $topK = 5): array | |
{ | |
$namespace = config('services.redis.namespace'); | |
$keys = $this->redis->keys("{$this->index}:{$namespace}:*"); | |
$similarities = []; | |
foreach ($keys as $key) { | |
$embeddingString = $this->redis->hGet($key, 'embedding'); | |
$embedding = explode(',', $embeddingString); | |
$similarity = $this->cosineSimilarity($queryEmbedding, $embedding); | |
$similarities[$key] = $similarity; | |
} | |
arsort($similarities); | |
$results = array_slice($similarities, 0, $topK, true); | |
$searchResults = []; | |
foreach ($results as $key => $similarity) { | |
$data = $this->redis->hGetAll($key); | |
$data['similarity'] = $similarity; | |
$searchResults[] = $data; | |
} | |
return $searchResults; | |
} | |
private function cosineSimilarity($embedding1, $embedding2): float|int | |
{ | |
$dotProduct = 0; | |
$magnitude1 = 0; | |
$magnitude2 = 0; | |
foreach ($embedding1 as $i => $value1) { | |
$value2 = $embedding2[$i]; | |
$dotProduct += $value1 * $value2; | |
$magnitude1 += $value1 ** 2; | |
$magnitude2 += $value2 ** 2; | |
} | |
$magnitude1 = sqrt($magnitude1); | |
$magnitude2 = sqrt($magnitude2); | |
if ($magnitude1 == 0 || $magnitude2 == 0) { | |
return 0; | |
} | |
return $dotProduct / ($magnitude1 * $magnitude2); | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Build with/for laravel.