Getting Started with OpenAI

Use kanoa with OpenAI’s GPT models or Azure OpenAI deployments.

OpenAI API

Prerequisites

Python 3.11 or higher
kanoa installed (pip install kanoa)
OpenAI API key

Step 1: Get Your API Key

Visit OpenAI Platform
Sign in or create an account
Click “Create new secret key”
Copy the API key

Step 2: Configure Authentication

Option A: Configuration File (Recommended)

Store your API key in ~/.config/kanoa/.env:

mkdir -p ~/.config/kanoa
echo "OPENAI_API_KEY=your-api-key-here" >> ~/.config/kanoa/.env

Option B: Environment Variable

export OPENAI_API_KEY="your-api-key-here"  # pragma: allowlist secret

Step 3: Use OpenAI Models

import matplotlib.pyplot as plt
import numpy as np
from kanoa import AnalyticsInterpreter

# Create sample data
x = np.linspace(0, 10, 100)
y = np.sin(x) * np.exp(-x/10)

plt.figure(figsize=(10, 6))
plt.plot(x, y)
plt.title("Damped Sine Wave")
plt.xlabel("Time")
plt.ylabel("Amplitude")

# Use OpenAI GPT models
interpreter = AnalyticsInterpreter(
    backend='openai',
    model='gpt-4o'  # or 'gpt-4-turbo', 'gpt-3.5-turbo'
)

result = interpreter.interpret(
    fig=plt.gcf(),
    context="Signal processing analysis",
    focus="Identify the damping characteristics"
)

print(result.text)
print(f"\nCost: ${result.usage.total_cost:.4f}")

Azure OpenAI

For Azure OpenAI deployments, provide your Azure endpoint and credentials:

from kanoa import AnalyticsInterpreter

interpreter = AnalyticsInterpreter(
    backend='openai',
    api_base='https://your-resource.openai.azure.com/openai/deployments/your-deployment',
    api_key='your-azure-key',  # pragma: allowlist secret
    api_version='2024-02-01'  # Azure API version
)

result = interpreter.interpret(
    fig=plt.gcf(),
    context="Business metrics",
    focus="Summarize trends"
)

Azure Configuration

You can also store Azure credentials in your config file:

# ~/.config/kanoa/.env
AZURE_OPENAI_API_KEY=your-azure-key
AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com
AZURE_OPENAI_DEPLOYMENT=your-deployment-name

Then use:

interpreter = AnalyticsInterpreter(backend='openai')

Model Selection

Recommended Models

gpt-4o: Latest and most capable multimodal model
gpt-4-turbo: Fast, high-quality responses
gpt-3.5-turbo: Cost-effective for simpler tasks

See OpenAI Backend Reference for detailed model information.

Next Steps

Cost Management: Learn about Cost Management
Knowledge Bases: Explore Knowledge Bases Guide
Backend Details: Check OpenAI Backend Reference

Troubleshooting

Authentication failed

Verify your API key is set correctly:

import os
print(os.getenv('OPENAI_API_KEY'))  # Should show your key

Rate limit errors

Reduce request frequency
Upgrade your OpenAI account tier
See OpenAI rate limits documentation

Model not found

Ensure the model name matches OpenAI’s current offerings. Check OpenAI models documentation.