The Sulie SDK revolutionizes time series forecasting by providing an effortless integration with the Sulie platform. Powered by Mimosa, a cutting-edge transformer-based foundation model, Sulie enables advanced forecasting with high accuracy through zero-shot inference and automatic fine-tuning. This means you can generate tailored forecasts without the need for extensive pre-training, allowing you to focus on leveraging your data's insights.

Key Features

1. Powerful Forecasting with Mimosa: Generate precise time series forecasts effortlessly, ensuring fast and reliable predictions even when training the model isn’t feasible, thanks to Mimosa's unique capabilities.

   import pandas as pd
   
   # Example time series data
   df = pd.DataFrame({
       'timestamp': pd.date_range(start='2023-01-01', periods=1000, freq='H'),
       'demand': [ ... ],           # Demand data
       'location': ['Plant A', ...] # Data for different locations
   })
   
   # Forecast demand for each location over the next 24 hours
   forecast = client.forecast(
       dataset=df,
       target='demand',
       group_by='location',
       date='timestamp',
       frequency='H',
       horizon=24,            # Predict 24 hours ahead
       num_samples=100        # Generate probabilistic forecasts
   )
   print(forecast)
   

2. Customized Fine-Tuning: Elevate your forecasting capabilities by optimizing Mimosa for your specific datasets and business scenarios using Weighted Quantile Loss (WQL) for precise evaluations.

   # Fine-tune Mimosa on custom dataset
   fine_tune_job = client.fine_tune(
       dataset=df,
       target="demand",
       description="Fine-tune for Plant A demand prediction"
   )
   
   # Check the fine-tuning job status
   print(f"Job status: {fine_tune_job.status}")
   

3. Efficient Dataset Management: Utilize Sulie’s Dataset API to manage and version your datasets seamlessly, making them accessible for collective use on forecasting and fine-tuning tasks.

   # Upload a dataset to Sulie
   dataset = client.upload_dataset(
       name="product-sales-data-v1",
       df=df,
       mode="append"  # Choose 'append' or 'overwrite'
   )
   
   # List available datasets
   datasets = client.list_datasets()
   print(f"Available datasets: {datasets}")
   

4. Advanced Forecasting with Custom Models: Select from your fine-tuned models for targeted forecasting, enhancing accuracy and relevance specific to your operational needs.

   # List custom and fine-tuned models
   custom_models = client.list_custom_models()
   
   # Select and forecast with a fine-tuned model
   model_name = custom_models[0].name
   custom_model = client.get_model(model_name)
   
   # Forecast using the selected model
   forecast_custom = custom_model.forecast(
       dataset=df,
       target='demand',
       group_by='location',
       date='timestamp',
       frequency='H',
       horizon=24,
       num_samples=50
   )
   print(forecast_custom)
   

Start Your Forecasting Journey

To embark on your forecasting journey, obtain your API key from the Sulie Dashboard and integrate it within your projects to unlock the power of Sulie's advanced forecasting capabilities.

Additional Resources

• API Documentation: Explore comprehensive documentation for detailed usage.
• Forecasting Guide: Discover the parameters for effective forecasting.
• Fine-Tuning Guide: Learn about tuning options and customizations.
• Support: Get assistance and provide feedback on the SDK.