Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Improvements to the Embedding functionality #776

Merged
merged 4 commits into from
Jun 11, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
105 changes: 100 additions & 5 deletions includes/Classifai/Providers/Azure/Embeddings.php
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
namespace Classifai\Providers\Azure;

use Classifai\Providers\OpenAI\EmbeddingCalculations;
use Classifai\Providers\OpenAI\Tokenizer;
use Classifai\Normalizer;
use Classifai\Features\Classification;
use Classifai\Features\Feature;
Expand Down Expand Up @@ -400,11 +401,29 @@ public function generate_embeddings_for_post( int $post_id, bool $force = false

// Get the embeddings for each chunk.
if ( ! empty( $content_chunks ) ) {
foreach ( $content_chunks as $chunk ) {
$embedding = $this->generate_embedding( $chunk );
$tokenizer = new Tokenizer( $this->get_max_tokens() );
$total_tokens = $tokenizer->tokens_in_content( $content );

if ( $embedding && ! is_wp_error( $embedding ) ) {
$embeddings[] = array_map( 'floatval', $embedding );
// If we have a lot of tokens, we need to get embeddings for each chunk individually.
if ( $this->max_tokens < $total_tokens ) {
foreach ( $content_chunks as $chunk ) {
$embedding = $this->generate_embedding( $chunk );

if ( $embedding && ! is_wp_error( $embedding ) ) {
$embeddings[] = array_map( 'floatval', $embedding );
}
}
} else {
// Otherwise let's get all embeddings in a single request.
$all_embeddings = $this->generate_embeddings( $content_chunks );

if ( $all_embeddings && ! is_wp_error( $all_embeddings ) ) {
$embeddings = array_map(
function ( $embedding ) {
return array_map( 'floatval', $embedding );
},
$all_embeddings
);
}
}
}
Expand Down Expand Up @@ -833,6 +852,7 @@ public function generate_embedding( string $text = '', Feature $feature = null )
'Content-Type' => 'application/json',
],
'body' => wp_json_encode( $body ),
'timeout' => 60, // phpcs:ignore WordPressVIPMinimum.Performance.RemoteRequestTimeout.timeout_timeout
]
);
$response = $this->get_result( $response );
Expand Down Expand Up @@ -863,6 +883,81 @@ public function generate_embedding( string $text = '', Feature $feature = null )
return $return;
}

/**
* Generate embeddings for an array of text.
*
* @param array $strings Array of text to generate embeddings for.
* @param Feature|null $feature Feature instance.
* @return array|boolean|WP_Error
*/
public function generate_embeddings( array $strings = [], $feature = null ) {
if ( ! $feature ) {
$feature = new Classification();
}

$settings = $feature->get_settings();

// Ensure the feature is enabled.
if ( ! $feature->is_feature_enabled() ) {
return new WP_Error( 'not_enabled', esc_html__( 'Classification is disabled or OpenAI authentication failed. Please check your settings.', 'classifai' ) );
}

/**
* Filter the request body before sending to OpenAI.
*
* @since 3.1.0
* @hook classifai_azure_openai_embeddings_request_body
*
* @param {array} $body Request body that will be sent to OpenAI.
* @param {array} $strings Array of text we are getting embeddings for.
*
* @return {array} Request body.
*/
$body = apply_filters(
'classifai_azure_openai_embeddings_request_body',
[
'input' => $strings,
'dimensions' => $this->get_dimensions(),
],
$strings
);

// Make our API request.
$response = wp_remote_post(
$this->prep_api_url( $feature ),
[
'headers' => [
'api-key' => $settings[ static::ID ]['api_key'],
'Content-Type' => 'application/json',
],
'body' => wp_json_encode( $body ),
'timeout' => 60, // phpcs:ignore WordPressVIPMinimum.Performance.RemoteRequestTimeout.timeout_timeout
]
);
$response = $this->get_result( $response );

if ( is_wp_error( $response ) ) {
return $response;
}

if ( empty( $response['data'] ) ) {
return new WP_Error( 'no_data', esc_html__( 'No data returned from OpenAI.', 'classifai' ) );
}

$return = [];

// Parse out the embeddings response.
foreach ( $response['data'] as $data ) {
if ( ! isset( $data['embedding'] ) || ! is_array( $data['embedding'] ) ) {
continue;
}

$return[] = $data['embedding'];
}

return $return;
}

/**
* Chunk content into smaller pieces with an overlap.
*
Expand All @@ -889,7 +984,7 @@ public function chunk_content( string $content = '', int $chunk_size = 150, $ove
array_slice(
$words,
max( $i - $overlap_size, 0 ),
$i + $chunk_size
$chunk_size + $overlap_size
)
);

Expand Down
100 changes: 95 additions & 5 deletions includes/Classifai/Providers/OpenAI/Embeddings.php
Original file line number Diff line number Diff line change
Expand Up @@ -514,11 +514,29 @@ public function generate_embeddings_for_post( int $post_id, bool $force = false

// Get the embeddings for each chunk.
if ( ! empty( $content_chunks ) ) {
foreach ( $content_chunks as $chunk ) {
$embedding = $this->generate_embedding( $chunk );
$tokenizer = new Tokenizer( $this->get_max_tokens() );
$total_tokens = $tokenizer->tokens_in_content( $content );

if ( $embedding && ! is_wp_error( $embedding ) ) {
$embeddings[] = array_map( 'floatval', $embedding );
// If we have a lot of tokens, we need to get embeddings for each chunk individually.
if ( $this->max_tokens < $total_tokens ) {
foreach ( $content_chunks as $chunk ) {
$embedding = $this->generate_embedding( $chunk );

if ( $embedding && ! is_wp_error( $embedding ) ) {
$embeddings[] = array_map( 'floatval', $embedding );
}
}
} else {
// Otherwise let's get all embeddings in a single request.
$all_embeddings = $this->generate_embeddings( $content_chunks );

if ( $all_embeddings && ! is_wp_error( $all_embeddings ) ) {
$embeddings = array_map(
function ( $embedding ) {
return array_map( 'floatval', $embedding );
},
$all_embeddings
);
}
}
}
Expand Down Expand Up @@ -975,6 +993,78 @@ public function generate_embedding( string $text = '', $feature = null ) {
return $return;
}

/**
* Generate embeddings for an array of text.
*
* @param array $strings Array of text to generate embeddings for.
* @param Feature|null $feature Feature instance.
* @return array|boolean|WP_Error
*/
public function generate_embeddings( array $strings = [], $feature = null ) {
if ( ! $feature ) {
$feature = new Classification();
}

$settings = $feature->get_settings();

// Ensure the feature is enabled.
if ( ! $feature->is_feature_enabled() ) {
return new WP_Error( 'not_enabled', esc_html__( 'Classification is disabled or OpenAI authentication failed. Please check your settings.', 'classifai' ) );
}

$request = new APIRequest( $settings[ static::ID ]['api_key'] ?? '', $feature->get_option_name() );

/**
* Filter the request body before sending to OpenAI.
*
* @since 2.2.0
* @hook classifai_openai_embeddings_request_body
*
* @param {array} $body Request body that will be sent to OpenAI.
* @param {array} $strings Array of text we are getting embeddings for.
*
* @return {array} Request body.
*/
$body = apply_filters(
'classifai_openai_embeddings_request_body',
[
'model' => $this->get_model(),
'input' => $strings,
'dimensions' => $this->get_dimensions(),
],
$strings
);

// Make our API request.
$response = $request->post(
$this->get_api_url(),
[
'body' => wp_json_encode( $body ),
]
);

if ( is_wp_error( $response ) ) {
return $response;
}

if ( empty( $response['data'] ) ) {
return new WP_Error( 'no_data', esc_html__( 'No data returned from OpenAI.', 'classifai' ) );
}

$return = [];

// Parse out the embeddings response.
foreach ( $response['data'] as $data ) {
if ( ! isset( $data['embedding'] ) || ! is_array( $data['embedding'] ) ) {
continue;
}

$return[] = $data['embedding'];
}

return $return;
}

/**
* Chunk content into smaller pieces with an overlap.
*
Expand All @@ -1001,7 +1091,7 @@ public function chunk_content( string $content = '', int $chunk_size = 150, $ove
array_slice(
$words,
max( $i - $overlap_size, 0 ),
$i + $chunk_size
$chunk_size + $overlap_size
)
);

Expand Down
Loading