Time Series Forecasting
Chronos
Safetensors
t5
time series
forecasting
pretrained models
foundation models
time series foundation models
time-series
Instructions to use amazon/chronos-bolt-small with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Chronos
How to use amazon/chronos-bolt-small with Chronos:
pip install chronos-forecasting
import pandas as pd from chronos import BaseChronosPipeline pipeline = BaseChronosPipeline.from_pretrained("amazon/chronos-bolt-small", device_map="cuda") # Load historical data context_df = pd.read_csv("https://autogluon.s3.us-west-2.amazonaws.com/datasets/timeseries/misc/AirPassengers.csv") # Generate predictions pred_df = pipeline.predict_df( context_df, prediction_length=36, # Number of steps to forecast quantile_levels=[0.1, 0.5, 0.9], # Quantiles for probabilistic forecast id_column="item_id", # Column identifying different time series timestamp_column="Month", # Column with datetime information target="#Passengers", # Column(s) with time series values to predict ) - Notebooks
- Google Colab
- Kaggle
Enhance model card with library, paper link, and GitHub reference
#1
by nielsr HF Staff - opened
This PR enriches the model card for chronos-bolt-small by:
- Adding the
library_name: chronos-forecastingto enable the interactive "How to use" widget on the Hub, providing a seamless experience for users. This choice is based on the explicit usage examples provided in the model card and GitHub README, which demonstrate inference withchronos.BaseChronosPipelinefrom thechronos-forecastingpackage. - Linking directly to the official Hugging Face paper page for Chronos-2: From Univariate to Universal Forecasting as the primary associated paper, and including its BibTeX citation in the "Citation" section.
- Adding a direct link to the GitHub repository (https://github.com/amazon-science/chronos-forecasting) for easy access to the source code and examples.
These updates improve discoverability and provide clearer guidance for users.