Extras
Extraction of complex objects (Anthropic)
Basics
- Basic use
- Basic use via mixin
- Handling errors with `Maybe` helper class
- Modes
- Making some fields optional
- Private vs public object field
- Automatic correction based on validation results
- Using attributes
- Using LLM API connections from config file
- Validation
- Custom validation using Symfony Validator
- Validation across multiple fields
Advanced
- Context caching
- Context caching
- Customize parameters of OpenAI client
- Custom prompts
- Using structured data as an input
- Extracting arguments of function or method
- Streaming partial updates during inference
- Providing example inputs and outputs
- Extracting scalar values
- Extracting sequences of objects
- Streaming
- Structures
Troubleshooting
LLM API Support
Extras
- Extraction of complex objects
- Extraction of complex objects (Anthropic)
- Extraction of complex objects (Cohere)
- Extraction of complex objects (Gemini)
- Embeddings
- Image processing - car damage detection
- Image to data (OpenAI)
- Image to data (Anthropic)
- Image to data (Gemini)
- Working directly with LLMs
- Working directly with LLMs and JSON - JSON mode
- Working directly with LLMs and JSON - JSON Schema mode
- Working directly with LLMs and JSON - MdJSON mode
- Working directly with LLMs and JSON - Tools mode
- Prompts
- Generating JSON Schema from PHP classes
- Generating JSON Schema dynamically
- Simple content summary
- Inference and tool use
- Create tasks from meeting transcription
- Translating UI text fields
- Web page to PHP objects
Extras
Extraction of complex objects (Anthropic)
Overview
This is an example of extraction of a very complex structure from the provided text with Anthropic Claude 3 model.
Example
<?php
$loader = require 'vendor/autoload.php';
$loader->add('Cognesy\\Instructor\\', __DIR__ . '../../src/');
use Cognesy\Instructor\Enums\Mode;
use Cognesy\Instructor\Extras\Sequence\Sequence;
use Cognesy\Instructor\Instructor;
$report = <<<'EOT'
[2021-09-01]
Acme Insurance project to implement SalesTech CRM solution is currently
in RED status due to delayed delivery of document production system, led
by 3rd party vendor - Alfatech. Customer (Acme) is discussing the resolution
with the vendor. Due to dependencies it will result in delay of the
ecommerce track by 2 sprints. System integrator (SysCorp) are working
to absorb some of the delay by deploying extra resources to speed up
development when the doc production is done. Another issue is that the
customer is not able to provide the test data for the ecommerce track.
SysCorp notified it will impact stabilization schedule unless resolved by
the end of the month. Steerco has been informed last week about the
potential impact of the issues, but insists on maintaining release schedule
due to marketing campaign already ongoing. Customer executives are asking
us - SalesTech team - to confirm SysCorp's assessment of the situation.
We're struggling with that due to communication issues - SysCorp team has
not shown up on 2 recent calls. Lack of insight has been escalated to
SysCorp's leadership team yesterday, but we've got no response yet. The
previously reported Integration Proxy connectivity issue which was blocking
policy track has been resolved on 2021-08-30 - the track is now GREEN.
Production deployment plan has been finalized on Aug 15th and awaiting
customer approval.
EOT;
echo "Extracting project events from the report:\n\n";
echo $report . "\n\n";
class ProjectEvents {
/**
* List of events extracted from the text
* @var ProjectEvent[]
*/
public array $events = [];
}
/** Represents a project event */
class ProjectEvent {
/** Title of the event - this should be a short, descriptive title of the event */
public string $title = '';
/** Concise, informative description of the event */
public string $description = '';
/** Type of the event */
public ProjectEventType $type = ProjectEventType::Other;
/** Status of the event */
public ProjectEventStatus $status = ProjectEventStatus::Unknown;
/** Stakeholders involved in the event */
/** @var Stakeholder[] */
public array $stakeholders = [];
/** Date of the event if reported in the text */
public ?string $date = '';
}
/** Represents status of project event */
enum ProjectEventStatus: string {
case Open = 'open';
case Closed = 'closed';
case Unknown = 'unknown';
}
/** Represents type of project event */
enum ProjectEventType: string {
case Risk = 'risk';
case Issue = 'issue';
case Action = 'action';
case Progress = 'progress';
case Other = 'other';
}
/** Represents a project stakeholder */
class Stakeholder {
/** Name of the stakeholder */
public string $name = '';
/** Role of the stakeholder, if specified */
public StakeholderRole $role = StakeholderRole::Other;
/** Any details on the stakeholder, if specified - any mentions of company, organization, structure, group, team, function */
public ?string $details = '';
}
enum StakeholderRole: string {
case Customer = 'customer';
case Vendor = 'vendor';
case SystemIntegrator = 'system integrator';
case Other = 'other';
}
$instructor = (new Instructor)->withConnection('anthropic');
echo "PROJECT EVENTS:\n\n";
$events = $instructor
->onSequenceUpdate(fn($sequence) => displayEvent($sequence->last()))
->request(
messages: $report,
responseModel: Sequence::of(ProjectEvent::class),
model: 'claude-3-5-sonnet-20240620', // 'claude-3-haiku-20240307'
prompt: 'Extract a list of project events with all the details from the provided input in JSON format using schema: <|json_schema|>',
mode: Mode::Json,
examples: [['input' => 'Acme Insurance project to implement SalesTech CRM solution is currently in RED status due to delayed delivery of document production system, led by 3rd party vendor - Alfatech. Customer (Acme) is discussing the resolution with the vendor. Production deployment plan has been finalized on Aug 15th and awaiting customer approval.', 'output' => [["type" => "object", "title" => "sequenceOfProjectEvent", "description" => "A sequence of ProjectEvent", "properties" => ["list" => [["title" => "Absorbing delay by deploying extra resources", "description" => "System integrator (SysCorp) are working to absorb some of the delay by deploying extra resources to speed up development when the doc production is done.", "type" => "action", "status" => "open", "stakeholders" => [["name" => "SysCorp", "role" => "system integrator", "details" => "System integrator",],], "date" => "2021-09-01",], ["title" => "Finalization of production deployment plan", "description" => "Production deployment plan has been finalized on Aug 15th and awaiting customer approval.", "type" => "progress", "status" => "open", "stakeholders" => [["name" => "Acme", "role" => "customer", "details" => "Customer",],], "date" => "2021-08-15",],],]]]]],
options: [
'max_tokens' => 4096,
'stream' => true,
])
->get();
echo "TOTAL EVENTS: " . count($events) . "\n";
//dump($events->list);
function displayEvent(ProjectEvent $event) : void {
echo "Event: {$event->title}\n";
echo " - Descriptions: {$event->description}\n";
echo " - Type: {$event->type->value}\n";
echo " - Status: {$event->status->value}\n";
echo " - Date: {$event->date}\n";
if (empty($event->stakeholders)) {
echo " - Stakeholders: none\n";
} else {
echo " - Stakeholders:\n";
foreach($event->stakeholders as $stakeholder) {
echo " - {$stakeholder->name} ({$stakeholder->role->value})\n";
}
}
echo "\n";
}
?>