Caravan
Open Source
The open global large-sample hydrology dataset — thousands of catchments with forcing, attributes and streamflow.
NeuralHydrology
ActiveDeep-learning rainfall-runoff modelling — the LSTM framework behind much of ML hydrology research.
- Pricing
- Open Source
- License
- BSD-3-Clause
- Commercial use
- Permitted
- Ecosystem
- Python, PyTorch
- Platform
- Windows, macOS, Linux
Install
pip install neuralhydrologyAbout NeuralHydrology
NeuralHydrology is the PyTorch framework from Frederik Kratzert and collaborators that made LSTM-based rainfall-runoff modelling reproducible. It handles the full experimental loop — CAMELS-style dataset loading, training, hyperparameter configs, multi-basin regional models, fine-tuning and evaluation — and underlies a large share of the published work showing deep learning beating conceptual models at ungauged and gauged prediction alike.
For practitioners it is the most direct path to a state-of-the-art data-driven baseline: if you have forcing and discharge data, a regional LSTM is a few config files away. Research-grade software, but unusually well documented for it.
Where it fits in the workflow
TerrainCatchmentRainfallRunoffRoutingHydraulicsMappingReporting
Alternatives to NeuralHydrology
Ranked by the topics, categories and workflow stages they share with NeuralHydrology.
OpenFloodHub
Open Source
An open-source take on Google Flood Hub — per-gauge neural network streamflow forecasting you can self-host.
Raven
Open Source
A flexible hydrological modelling framework — emulate GR4J, HBV, HMETS or build your own model structure.
StormHub
Open Source
Dewberry's open tool for building STAC catalogs of storms and gauge data — inputs for stochastic storm transposition.
ASCAT Soil Moisture (H SAF)
Gov-Public
Near-real-time scatterometer soil moisture from EUMETSAT — hours of latency, useful for operational monitoring.
Delft-FEWS
Free
Deltares' operational forecasting platform — the shell running national flood-forecast centres worldwide.