-
Notifications
You must be signed in to change notification settings - Fork 848
Open
Description
Description
The documentation homepage in docs/source/index.rst provides a good overview of PyTorch Forecasting, but it could be improved to make onboarding easier for new users.
Currently, the page explains the library but does not include a quick usage example or a clear feature overview. Adding these sections would help users understand the library faster and improve the overall documentation experience.
Proposed Improvements
-
Add a Quick Example section
- Include a minimal code snippet showing how to create a dataset and train a model.
-
Add a Feature Highlights section
-
Clearly summarize key capabilities such as:
- TimeSeriesDataSet abstraction
- Built-in neural forecasting models
- Hyperparameter tuning with Optuna
- Visualization and interpretation tools.
-
-
Minor wording improvements
- Replace occurrences of
timeserieswithtime seriesfor consistency with standard terminology.
- Replace occurrences of
-
Add a short model overview
- Mention major models supported by the library (e.g., Temporal Fusion Transformer, NHiTS, NBeats).
Benefits
- Improves onboarding for new users
- Makes documentation more informative
- Aligns the homepage with common documentation standards used in ML libraries
File Location
docs/source/index.rst
I would be happy to work on this improvement and submit a PR if maintainers agree with the proposal.
Please assign these task to me so that i can start conributions. @phoeenniixx
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels