Anyone ever tried pycaret timeseries?
Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
31–40 of 91 posts
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#32For time series, classical methods (ARIMA etc) still continue to perform very well for most problems.
I don't think they necessarily perform well, its just that time-series forecasting is notoriously hard and we haven't seemed to progress that far from ARIMA type models (to the best of my knowledge)
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#33State of the art is 1D convnets, bleeding edge is transformers.
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#34Earlier quoted context omitted.
Can you substantiate your comment? What would you like to see in a python tool that R does uniquely well?
Not parent, but easy: https://cran.r-project.org/web/packages/forecast/index.html /e: I'm a bit out of this game for 2-3 years now, but Python had nothing comparable. Prophet is suboptimal at best. Some things are implemented here or there, but it's all over the place. So I agree that R really shines here.
Wonderful, I'm gonna go check this out in great depth!
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#35Earlier quoted context omitted.
Not parent, but easy: https://cran.r-project.org/web/packages/forecast/index.html /e: I'm a bit out of this game for 2-3 years now, but Python had nothing comparable. Prophet is suboptimal at best. Some things are implemented here or there, but it's all over the place. So I agree that R really shines here.
Wonderful, I'm gonna go check this out in great depth!
I can also recommend the book and blog by Rob Hyndman! https://robjhyndman.com/hyndsight/
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#36Former Reuters Research Director here.
When modeling time series, you will want a model that is sensitive both to short term and longer term movements. In other words, a Long Term Short Term Memory (LSTM).
Sepp Hochreiter invented this concept in his Master's thesis supervised by Jürgen Schmidhuber in Munich in the 1990s; today, it's the most-cited type of neural network.
Here are papers describing it: https://people.idsia.ch/~juergen/rnn.html
In Python, you can use TensorFlow's LSTMCell class: https://www.datacamp.com/tutorial/lstm-python-stock-market
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#37Earlier quoted context omitted.
I've had someone in a team implement feature engineering using tsfresh. It lead to a malignantly under-performing, complicated heap of spaghetti that was a nightmare to get into production. Weird API, slow code, little added value over simple features found in a day of manual exploration. Person doing the implementation wasn't a rock star coder so we couldn't fix the performance and complexity issues in time; it was…
I've tried it a number of times, and had a similar experience. The whole stack is orders of magnitudes slower to compute compared to "simple" features (i.e. rolling averages), without showing real predictive improvements.
Thanks for the elaboration. Good to know someone had a similar experience.
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#38Earlier quoted context omitted.
Can you substantiate your comment? What would you like to see in a python tool that R does uniquely well?
Not parent, but easy: https://cran.r-project.org/web/packages/forecast/index.html /e: I'm a bit out of this game for 2-3 years now, but Python had nothing comparable. Prophet is suboptimal at best. Some things are implemented here or there, but it's all over the place. So I agree that R really shines here.
To be fair, Darts looks pretty good relative to forecast: https://github.com/unit8co/darts
I would generally prefer R for this kind of stuff as the experts generally write the code, but Darts seems OK and is well-tested, at the very least (haven't had a chance to use it in anger yet).
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#39Stuff I've used:
- Prophet - seems to be the current 'standard' choice
- ARIMA - Classical choice
- Exponential Moving Average - dead simple to implement, works well for stuff that's a time series but not very seasonal
- Kalman/Statespace model - used by Splunk's predict[1] command (pretty sure I always used LLP5)
- 2-3 anomalies in a row
- high deviation from expected
- multiple models all detected anomaly
[1] :https://docs.splunk.com/Documentation/Splunk/9.0.1/SearchRef...
Re: Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
#40Stuff I've used: - Prophet - seems to be the current 'standard' choice - ARIMA - Classical choice - Exponential Moving Average - dead simple to implement, works well for stuff that's a time series but not very seasonal - Kalman/Statespace model - used by Splunk's predict[1] command (pretty sure I always used LLP5) I did some anomaly detection work, in business transactions, and found the best way was to create a sort…
I don't know what you've been using Prophet for but I found it to be very brittle.