This week's main focus:
✅ Quarto's new features
✅ New learning resources
✅ Book of the week - Statistical Rethinking
✅ Introduction to the statsforecast library
www.linkedin.com/pulse/quarto...
#DataScience
This week's main focus:
✅ Quarto's new features
✅ New learning resources
✅ Book of the week - Statistical Rethinking
✅ Introduction to the statsforecast library
www.linkedin.com/pulse/quarto...
#DataScience
https://github.com/Nixtla/statsforecast/tree/main/experiments/m3
"a deep-learning ensemble that takes more than 14 days to run and costs around USD 11,000, outperforms a statistical ensemble that takes 6 minutes to run and costs $0.5c by … 1/3
https://github.com/Nixtla/statsforecast/tree/main/experiments/m3
"a deep-learning ensemble that takes more than 14 days to run and costs around USD 11,000, outperforms a statistical ensemble that takes 6 minutes to run and costs $0.5c by … 1/3
The statsforecast library, as the name implies, provides statistical forecasting models such as ARIMA, Holt-Winters, ETS, Theta, and many others 🎯.
#Python #DataScience #Stats #timeseries
The statsforecast library, as the name implies, provides statistical forecasting models such as ARIMA, Holt-Winters, ETS, Theta, and many others 🎯.
#Python #DataScience #Stats #timeseries
Tired of guessing the right ETS model? AutoETS(model="ZZZ") does the heavy lifting:
✅ Automatically selects the best error, trend & seasonality combo (AAN, MAM, etc.)
Tired of guessing the right ETS model? AutoETS(model="ZZZ") does the heavy lifting:
✅ Automatically selects the best error, trend & seasonality combo (AAN, MAM, etc.)
https://www.libhunt.com/r/autokitteh
Event Attributes
https://www.libhunt.com/r/autokitteh
Event Attributes
• Uso de bibliotecas modernas como sktime, statsforecast e Darts
• Uso de bibliotecas modernas como sktime, statsforecast e Darts
www.linkedin.com/pulse/foreca...
Code: github.com/RamiKrispin/...
Notebook: ramikrispin.github.io/linkedin-new...
www.linkedin.com/pulse/foreca...
Code: github.com/RamiKrispin/...
Notebook: ramikrispin.github.io/linkedin-new...
📖🔗: cienciadedatos.net/documentos/p...
📖🔗: cienciadedatos.net/documentos/p...
statsforecast is a mature, production-oriented Python library for high-performance univariate time-series forecasting using a broad suite of statistical/econometric models (AutoARIMA/ETS/CES/Theta, MSTL, TBAT...
https://gitrated.com/Nixtla/statsforecast
statsforecast is a mature, production-oriented Python library for high-performance univariate time-series forecasting using a broad suite of statistical/econometric models (AutoARIMA/ETS/CES/Theta, MSTL, TBAT...
https://gitrated.com/Nixtla/statsforecast
v0.1.2 adds:
- Orientation-correct finite-sample conformal quantile — closes the asymmetric L/R miss gap that affected v0.1.1
- CSPModel statsforecast-compatible wrapper now matches the core path to floating-point precision
v0.1.2 adds:
- Orientation-correct finite-sample conformal quantile — closes the asymmetric L/R miss gap that affected v0.1.1
- CSPModel statsforecast-compatible wrapper now matches the core path to floating-point precision
It’s about fixing a design that was never meant for modern workloads.
StatsForecast was a stepping stone.
Chronax is what happens when you actually rethink the execution model.
It’s about fixing a design that was never meant for modern workloads.
StatsForecast was a stepping stone.
Chronax is what happens when you actually rethink the execution model.
StatsForecast was dubiously marketed as “fast” — but let’s be honest about what it actually is:
👉 a fairly awkward rewrite of Hyndman-style R libraries into Python
👉 with some acceleration layered on top
StatsForecast was dubiously marketed as “fast” — but let’s be honest about what it actually is:
👉 a fairly awkward rewrite of Hyndman-style R libraries into Python
👉 with some acceleration layered on top
Try "ZZZ"—your new forecasting shortcut! 🔥
'Mastering Modern Time Series Forecasting : The Complete Guide to Statistical, Machine Learning & Deep Learning Models in Python' -> valeman.gumroad.com/...
#TimeSeries #Forecasting #Python #StatsForecast
Try "ZZZ"—your new forecasting shortcut! 🔥
'Mastering Modern Time Series Forecasting : The Complete Guide to Statistical, Machine Learning & Deep Learning Models in Python' -> valeman.gumroad.com/...
#TimeSeries #Forecasting #Python #StatsForecast
ARIMA's surprising connection to Vapnik's Statistical Learning Theory (VC dimension enters the chat!).
👨💻 Code That Cuts the Fluff: No filler.
Discover the best forecasting libraries:
StatsForecast (Nixtla) | statsmodels | pmdarima | ETNA | R | Julia
ARIMA's surprising connection to Vapnik's Statistical Learning Theory (VC dimension enters the chat!).
👨💻 Code That Cuts the Fluff: No filler.
Discover the best forecasting libraries:
StatsForecast (Nixtla) | statsmodels | pmdarima | ETNA | R | Julia