Skip to main content
Get started with Synthefy’s Cloud Forecasting API in minutes. This guide shows you the simplest way to make your first forecast request.

​
Prerequisites

​
Install the Client

pip install synthefy

​
API Key Setup

Get your API key from the Synthefy Console.
API Key Images: Detailed instructions with screenshots on how to obtain your API key from the console will be added here.

​
Simple Example

Here’s the simplest way to use the Synthefy Cloud API: 
import pandas as pd
from synthefy import SynthefyAPIClient

# Constants
HISTORY_PERIODS = 5
HISTORY_SALES = [100, 105, 98, 110, 102]
TARGET_PERIODS = 7

# Prepare your data
history_df = pd.DataFrame({
    'date': pd.date_range('2024-01-01', periods=HISTORY_PERIODS, freq='D'),
    'sales': HISTORY_SALES
})

target_df = pd.DataFrame({
    'date': pd.date_range('2024-01-06', periods=TARGET_PERIODS, freq='D'),
    'sales': [None] * TARGET_PERIODS  # What we want to predict
})

# Make the forecast
api_key = "your_api_key_here"  # Get from https://console.synthefy.com/
with SynthefyAPIClient(
    api_key=api_key,
    base_url="https://forecast.synthefy.com",
    endpoint="/v2/forecast"
) as client:
    print("\nSending request...")
    forecast_dfs = client.forecast_dfs(
        history_dfs=[history_df],
        target_dfs=[target_df],
        target_col='sales',
        timestamp_col='date',
        metadata_cols=[],
        leak_cols=[],
        model='Migas-1.0'
    )

# Get your predictions
forecast_df = forecast_dfs[0]
print(forecast_df)
That’s it! The forecast_dfs method handles all the complexity for you.

​
Next Steps

API Reference

Explore the complete API client documentation.

On-Premise Deployment

Interested in deploying on-premise? Check out our on-premise quickstart guide.