This section will guide you through the steps required to install, configure,
and use Sulie to generate accurate forecasts for your time series data.
Getting started with Sulie is straightforward and requires minimal setup.
Start by creating an account through the Sulie dashboard.
Once registered, navigate to the API Keys section where you can generate API
keys for your organization. Each organization can have multiple API keys to
manage different applications or environments. API keys are essential for
authenticating your requests to the Sulie platform and accessing its forecasting
and fine-tuning capabilities.
At the core of Sulie is the Dataset abstraction, which builds on Pandas
DataFrames to represent your time series data. While Datasets can contain
arbitrary data, they must include your target variable for forecasting.
Here’s a simple example showing how to upload a dataset and generate forecasts:
Copy
import osimport pandas as pdfrom sulie import Sulieclient = Sulie( api_key=os.environ.get("SULIE_API_KEY"))# Prepare your datadf = pd.DataFrame(your_data)# Upload a datasetdataset = client.upload_dataset( name="product-purchases-v1", df=df)# Forecast on time-series data forecast = client.forecast( dataset="product-purchases-v1", horizon=30, # 30 time steps ahead target_col="y")