bio-timesfm-forecasting

What problem does it solve?

This Skill automates the process of forecasting univariate time series data, eliminating the need for custom model training and complex statistical tuning.

Core Features & Use Cases

• Zero-Shot Forecasting: Predict future values for any univariate time series (sales, sensor data, energy, etc.) using Google's TimesFM foundation model without retraining.
• Probabilistic Forecasts: Generates point forecasts along with prediction intervals (quantiles) for uncertainty estimation.
• Flexible Input: Supports CSV, DataFrame, or array inputs.
• Use Case: Predict next month's sales figures based on the last three years of historical monthly sales data, including an 80% confidence interval for the forecast.

Quick Start

Use the bio-timesfm-forecasting skill to forecast the next 12 months of data from the attached file 'monthly_sales.csv'.