search

Nixtla / nixtla

TimeGPT-1: production ready pre-trained Time Series Foundation Model for forecasting and anomaly detection. Generative pretrained transformer for time series trained on over 100B data points. It's capable of accurately predicting various domains such as retail, electricity, finance, and IoT with just a few lines of code 🚀.
https://docs.nixtla.io
Other
2.31k stars 187 forks source link

feat: support custom and integer frequencies #532

Open jmoralez opened 1 week ago

jmoralez commented 1 week ago

Adds support for providing custom pandas offsets as freq as well as integer timestamps and freq=some_int.

review-notebook-app[bot] commented 1 week ago

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

github-actions[bot] commented 1 week ago
Experiment Results ## Experiment 1: air-passengers ### Description: | variable | experiment | |:--------------|:-------------| | h | 12 | | season_length | 12 | | freq | MS | | level | None | | n_windows | 1 | ### Results: | metric | timegpt-1 | timegpt-1-long-horizon | SeasonalNaive | Naive | |:-----------|------------:|-------------------------:|----------------:|-----------:| | mae | 12.6793 | 11.0623 | 47.8333 | 76 | | mape | 0.027 | 0.0232 | 0.0999 | 0.1425 | | mse | 213.936 | 199.132 | 2571.33 | 10604.2 | | total_time | 8.4852 | 1.9724 | 0.0054 | 0.004 | ### Plot: ![](https://github.com/Nixtla/nixtla/blob/docs-figs-model-performance//action_files/models_performance/plots/plot_air-passengers_12_12_MS_None_1.png?raw=true) ## Experiment 2: air-passengers ### Description: | variable | experiment | |:--------------|:-------------| | h | 24 | | season_length | 12 | | freq | MS | | level | None | | n_windows | 1 | ### Results: | metric | timegpt-1 | timegpt-1-long-horizon | SeasonalNaive | Naive | |:-----------|------------:|-------------------------:|----------------:|-----------:| | mae | 58.1031 | 58.4587 | 71.25 | 115.25 | | mape | 0.1257 | 0.1267 | 0.1552 | 0.2358 | | mse | 4040.21 | 4110.79 | 5928.17 | 18859.2 | | total_time | 1.2967 | 1.1749 | 0.0045 | 0.0041 | ### Plot: ![](https://github.com/Nixtla/nixtla/blob/docs-figs-model-performance//action_files/models_performance/plots/plot_air-passengers_24_12_MS_None_1.png?raw=true) ## Experiment 3: electricity-multiple-series ### Description: | variable | experiment | |:--------------|:-------------| | h | 24 | | season_length | 24 | | freq | H | | level | None | | n_windows | 1 | ### Results: | metric | timegpt-1 | timegpt-1-long-horizon | SeasonalNaive | Naive | |:-----------|------------:|-------------------------:|----------------:|---------------:| | mae | 178.293 | 268.13 | 269.23 | 1331.02 | | mape | 0.0234 | 0.0311 | 0.0304 | 0.1692 | | mse | 121589 | 219485 | 213677 | 4.68961e+06 | | total_time | 1.0219 | 2.7283 | 0.0054 | 0.0049 | ### Plot: ![](https://github.com/Nixtla/nixtla/blob/docs-figs-model-performance//action_files/models_performance/plots/plot_electricity-multiple-series_24_24_H_None_1.png?raw=true) ## Experiment 4: electricity-multiple-series ### Description: | variable | experiment | |:--------------|:-------------| | h | 168 | | season_length | 24 | | freq | H | | level | None | | n_windows | 1 | ### Results: | metric | timegpt-1 | timegpt-1-long-horizon | SeasonalNaive | Naive | |:-----------|------------:|-------------------------:|----------------:|---------------:| | mae | 465.497 | 346.972 | 398.956 | 1119.26 | | mape | 0.062 | 0.0436 | 0.0512 | 0.1583 | | mse | 835021 | 403760 | 656723 | 3.17316e+06 | | total_time | 0.4739 | 0.7891 | 0.0056 | 0.0052 | ### Plot: ![](https://github.com/Nixtla/nixtla/blob/docs-figs-model-performance//action_files/models_performance/plots/plot_electricity-multiple-series_168_24_H_None_1.png?raw=true) ## Experiment 5: electricity-multiple-series ### Description: | variable | experiment | |:--------------|:-------------| | h | 336 | | season_length | 24 | | freq | H | | level | None | | n_windows | 1 | ### Results: | metric | timegpt-1 | timegpt-1-long-horizon | SeasonalNaive | Naive | |:-----------|--------------:|-------------------------:|----------------:|---------------:| | mae | 558.673 | 459.757 | 602.926 | 1340.95 | | mape | 0.0697 | 0.0565 | 0.0787 | 0.17 | | mse | 1.22723e+06 | 739114 | 1.61572e+06 | 6.04619e+06 | | total_time | 0.6664 | 0.7267 | 0.006 | 0.0056 | ### Plot: ![](https://github.com/Nixtla/nixtla/blob/docs-figs-model-performance//action_files/models_performance/plots/plot_electricity-multiple-series_336_24_H_None_1.png?raw=true)