gensphere

Quickstart Guide

Welcome to the GenSphere Quickstart Guide! This guide will walk you through creating and executing your first GenSphere workflow. By the end of this guide, you will have a functional workflow that interacts with the GenSphere platform.

Table of Contents

1. Setting Up Environment Variables
2. Importing GenSphere Modules
3. Running Your First Flow
   • Pulling a Project from the Platform
   • Executing the Flow
   • Accessing Outputs
4. Next Steps

Setting Up Environment Variables

Before you can run GenSphere workflows, you need to set up the necessary environment variables, including your API keys.

1. Obtain API Keys

Ensure you have the following API keys:

• OpenAI API Key: Obtain here.
• Composio API Key (optional): Obtain here.

2. Define Environment Variables

Set your environment variables in your terminal or include them in a .env file. Using a .env file is recommended for convenience and security.

Using a .env File

Create a file named .env in your project directory and add the following lines:

OPENAI_API_KEY="YOUR_OPENAI_API_KEY"

Substitute “YOUR_API_KEY” by your actual api key.

Loading Environment Variables

Ensure that your Python environment loads the .env file. GenSphere uses the python-dotenv package to load environment variables from a .env file automatically.

If you haven’t installed python-dotenv, install it using pip:

pip install python-dotenv

Alternatively, you can manually set environment variables in your terminal session:

# On Windows
set OPENAI_API_KEY=YOUR_OPENAI_API_KEY

# On macOS and Linux
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY

Importing GenSphere Modules

Begin by importing the necessary GenSphere modules in your Python script or Jupyter Notebook.

import logging
import traceback


# Set up logging configuration before importing other modules
logging.basicConfig(
    level=logging.DEBUG,
    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
    handlers=[
        logging.FileHandler("app.log", mode='w'),
        logging.StreamHandler()
    ]
)

from gensphere import genflow, yaml_utils
from gensphere.genflow import GenFlow
from gensphere.yaml_utils import YamlCompose
from gensphere.visualizer import Visualizer
from gensphere.hub import Hub

Explanation of Imports

• logging & traceback: For logging and debugging.
• gensphere modules: Core classes and utilities for defining and running workflows.

Running Your First Flow

This section guides you through pulling a pre-built workflow from the GenSphere platform, executing it, and accessing its outputs.

Pulling a Project from the Platform

GenSphere allows you to pull pre-built workflows from its open platform using a unique push_id. This enables you to quickly start with existing workflows.

# Initialize the Hub
hub = Hub()

# Define the push_id and filenames to save the pulled files
push_id = '2c03079c-0e33-489e-bbbe-777da744d56f'
yaml_filename = 'simple_examples.yaml' #this will be the path of your YAML file after downloading it
functions_filename = 'simple_examples_functions.py' #this will be the path of your functions file after downloading it

# Pull the project from the platform
hub.pull(
    push_id=push_id,
    yaml_filename=yaml_filename,
    functions_filename=functions_filename,
    save_to_disk=True
)

Explanation:

• Hub: The Hub class manages interactions with the GenSphere platform.
• push_id: Unique identifier for the project you want to pull.
• yaml_filename & functions_filename: Names of the files to save locally.

Result:

This command downloads the specified YAML workflow and associated Python functions, saving them to your current working directory.

Executing the Flow

Once you’ve pulled the workflow files, you can execute the workflow using the GenFlow class.

# Initialize GenFlow with the pulled YAML and functions files
flow = GenFlow('simple_examples.yaml', 'simple_examples_functions.py')

# Parse the YAML file to construct the execution graph
flow.parse_yaml()

# Run the workflow
flow.run()

Explanation:

• GenFlow: Core class responsible for parsing, building, and executing workflows.
• parse_yaml(): Parses the YAML file and constructs the execution graph.
• run(): Executes the workflow based on the constructed graph.

Accessing Outputs

After running the workflow, you can access the outputs generated by each node.

# Access the outputs of the workflow
outputs = flow.outputs

# Print the outputs
for node_name, output in outputs.items():
    print(f"Outputs from node '{node_name}': {output}")

Explanation:

• flow.outputs: A dictionary containing the outputs from each node in the workflow.
• Iterating through outputs: Print or utilize the outputs as needed.

Example Output:

Outputs from node 'get_current_date': {'current_date': '2024-04-27'}
Outputs from node 'get_timewindow': {'time_window': 'last month'}
Outputs from node 'product_hunt_scrape': {'product_hunt_scrape_results': ...}

Next Steps

Congratulations! You’ve successfully installed GenSphere, pulled a workflow from the platform, executed it, and accessed its outputs. Here are some suggested next steps to further enhance your understanding and usage of GenSphere:

1. Explore the Tutorials: Dive into the Quickstart Tutorial for more detailed guidance.
2. Define Your Own Workflows: Start creating your own workflows by defining YAML files, custom functions, and schemas.
3. Integrate with Tools: Leverage integrations with LangChain and Composio to enhance your workflows.
4. Visualize Your Workflows: Use the Visualizer class to get a graphical representation of your workflows.
5. Contribute to GenSphere: Share your workflows and contribute to the GenSphere community by pushing your projects to the platform.

---

Continue to the Tutorials section to deepen your understanding of GenSphere’s capabilities! ```

This site is open source. Improve this page.