docs: improve forecasting model search metadata - #1594
Merged
Conversation
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
|
|
marcopeix
approved these changes
Aug 10, 2026
mergenthaler
left a comment
mergenthaler
left a comment
Member
There was a problem hiding this comment.
Minor comments for you @marcopeix. (Also, the changes are still not deployed to nixtlaverse.nixtla.io)
| --- | ||
| description: >- | ||
| Informer: Efficient Transformer with ProbSparse attention for long-sequence time series forecasting. Reduces O(L^2) complexity for scalable predictions. | ||
| Build long horizon forecasts with Informer in NeuralForecast. Learn ProbSparse attention, architecture, parameters, and a complete Python workflow. |
Member
There was a problem hiding this comment.
@marcopeix, I don't think this description is accurate (no complete worklow) and it looses some of the important details.
| --- | ||
| description: >- | ||
| iTransformer: Inverted Transformer architecture for multivariate time series forecasting with attention on time points and feed-forward on series dimensions. | ||
| Build multivariate forecasts with iTransformer in NeuralForecast. Learn its inverted attention architecture, parameters, and Python workflow. |
| --- | ||
| description: >- | ||
| KAN: Kolmogorov-Arnold Networks for time series forecasting. MLP alternative using learnable activation functions for improved non-linear pattern modeling. | ||
| Build time series forecasts with Kolmogorov Arnold Networks in NeuralForecast. Compare KAN with MLP models and run a complete Python example. |
Member
There was a problem hiding this comment.
Maybe I missed something but I did not see any complete python example or comparison with MLP models.
| --- | ||
| description: >- | ||
| PatchTST: Efficient Transformer model for multivariate forecasting using patched time series and channel-independence for scalable long-term predictions. | ||
| Build long horizon forecasts with PatchTST in NeuralForecast. Learn how temporal patches and channel independence work, then run a complete Python example. |
Member
There was a problem hiding this comment.
IMO the original description is correct.
| --- | ||
| description: >- | ||
| TFT: Temporal Fusion Transformer with interpretable multi-horizon forecasting. LSTM encoder, multi-head attention, variable selection for complex time series. | ||
| Build interpretable forecasts with Temporal Fusion Transformer in NeuralForecast using static, historic, and future variables in Python. |
Member
There was a problem hiding this comment.
Some important info is lost on this new description
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Why
These pages already receive meaningful search visibility, but terse titles do not clearly communicate forecasting intent, Python applicability, or the practical value of each reference.
How
Each title names the model and forecasting task. Each description states the implementation value without making performance claims.
Testing
jq.jupyter nbconvertto validate notebook structure.git diff --check.Checklist