nixtla Profile picture
Feb 21 โ€ข 8 tweets โ€ข 3 min read
๐ŸŽ‰ We are thrilled to announce the release of the latest version of mlforecast a #Python library for Scalable #machinelearning ๐Ÿค– for #timeseries #forecasting

๐Ÿš€ This version comes with exciting new features that are sure to make forecasting even more efficient and accurate

๐Ÿงต
๐Ÿ”ฎ Conformal Prediction: We've added the ability to generate probabilistic forecasts using conformal prediction. You can choose the levels of your intervals and the number of conformity scores to use.
โšก๏ธ In addition, mlforecast includes the capability to fit multiple time series with global models. That means you donโ€™t need to calculate sequential conformal intervals for each series and period. I.e., conformal prediction for time series just got a lot faster and easier. ๐Ÿคฏ
๐Ÿค– Models: We've expanded our model selection to include some of your favorite sklearn methods, including LinearRegression, KNeighborsRegressor, MLPRegressor, and tree-based models like LightGBM and XGBoost.
๐Ÿ“Š Static and Exogenous Variables: Our models can now take static and exogenous variables, giving you even more flexibility in your forecasting tasks.
๐Ÿง  Training Improvements: You can now train one model per horizon or generate recursive forecasts for the whole horizon.
๐Ÿ“š New Tutorials and Documentation: We've added resources to help you master probabilistic forecasting (conformal prediction), cross-validation, transfer learning, and detecting demand peaks.
Thank you for your support, and happy forecasting! ๐ŸŒŸ
โญ๏ธ Repo: github.com/nixtla/mlforecโ€ฆ

โ€ข โ€ข โ€ข

Missing some Tweet in this thread? You can try to force a refresh
ใ€€

Keep Current with nixtla

nixtla Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @nixtlainc

Feb 20
๐ŸŽ‰ We are very excited to release the new features of NeuralForecast! ๐Ÿฅณ๐Ÿš€

With this release time series forecasting with neural models is even more accessible and powerful. Here are some of the highlights

๐Ÿงต

#timeseries #python #deeplearning #forecasting
๐Ÿ“ˆ New Models:
- Temporal Convolution Network
- AutoNBEATSx
- AutoTFT (Transformers)
๐Ÿงฎ Recurrent models (RNN, LSTM, GRU, DilatedRNN) can now take static, historical, and future exogenous variables.
Read 9 tweets
Dec 5, 2022
Today we are glad to announce these exciting features! ๐ŸŽ‰

* Improved StatsForecast class
* Selection of best-performing models
* New documentation
* Plotting functionalities

Stay tuned for an exciting polemic comparison tomorrow!

#reinventforecasting
StatsForecast class:

With the new StatsForecast class, you can fit many models for many time series in less than 15 lines of code

You can include @dask_dev, @raydistributed, and #spark by changing just the backend parameter Image
Models: StatsForecast 27 models.

They can be included and imported in a few lines of code. Image
Read 7 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(