DSPyGen JS

DSPyGen JS is a JavaScript/TypeScript library designed to streamline the development of data processing and generation pipelines, with a special focus on integrating language models and other AI-driven processes. This README provides an overview of the key components and functionalities within the library, tailored specifically to the TypeScript code snippets provided earlier.

Overview

DSPyGen JS leverages modern TypeScript features to offer developers a robust framework for building and managing complex data processing and generation pipelines. Key functionalities include interfacing with language models, handling events with a custom actor model, and validating data against JSON schemas.

Key Features

• GroqLM Integration: A class for interacting with Groq language models, supporting basic requests and forwarding prompts to the model.
• Event Handling: Utilize the Actor model for concurrent operations and message handling within the pipeline.
• Data Validation: Classes that validate data objects against predefined JSON schemas, ensuring the integrity of data throughout the pipeline.
• Utilities: Helper functions and classes for common tasks, such as extracting JSON from strings and handling custom event data.

Installation

To use DSPyGen JS in your project, ensure you have Node.js and npm or yarn installed. Then, add DSPyGen JS to your project dependencies using npm or yarn:

npm install dpgjs
# or
yarn add dpgjs

Usage

GroqLM Integration

The GroqLM class allows for seamless interaction with Groq's language models. It's designed to handle various communication errors gracefully and provides a simple interface for making predictions:

import { GroqLM, GroqModels } from "dpgjs";

const model = new GroqLM(GroqModels.llama2);
model.forward(prompt)
  .then(response => console.log(response))
  .catch(error => console.error(error));

Actor System for Event Handling

DSPyGen JS employs an actor system to manage concurrency and facilitate message passing between different components of your pipeline:

import { ActorSystem, BaseActor, BaseMessage } from "dpgjs";

class MyActor extends BaseActor {
  handleMessage(message: BaseMessage) {
    // Handle message
  }
}

const system = new ActorSystem();
const actor = system.actorOf(MyActor);

Data Validation with JSON Schema

Validating data objects against JSON schemas is streamlined with the VEvent class, which automatically validates its properties upon creation:

import { VEvent } from "dpgjs";

const eventData = {
  // Event data following the VEvent schema
};
const event = new VEvent(eventData);

Extracting JSON from Strings

The utility function extract can be used to parse strings and extract JSON objects, simplifying the process of dealing with unstructured data:

import { extract } from "dpgjs";

const jsonString = extract('{"key": "value"}');
console.log(jsonString); // Outputs the parsed JSON object

Testing

DSPyGen JS comes with built-in support for unit testing using Vitest, allowing you to ensure the reliability and correctness of your pipeline components:

import { describe, it, expect } from "vitest";
import { MyComponent } from "dpgjs";

describe("MyComponent", () => {
  it("should behave correctly", () => {
    const component = new MyComponent();
    expect(component.doSomething()).toBe(true);
  });
});

Conclusion

DSPyGen JS provides a comprehensive set of tools for building data processing and generation pipelines, emphasizing ease of use, flexibility, and reliability. Whether you're working with complex language models or need a robust system for event handling and data validation, DSPyGen JS offers a solid foundation for your projects.