NeuralHydrology LSTM¶

Overview¶

neuralhydrology_lstm is the first community-style hydrology baseline in the PyHazards flood roadmap.

At a Glance¶

Hazard Family

Flood

Public catalog grouping used for this model.

Maturity

Implemented

Catalog maturity label used on the index page.

Tasks

Streamflow

Benchmark Family

Flood Benchmark

Primary benchmark-family link used for compatible evaluation coverage.

Description¶

neuralhydrology_lstm is the first community-style hydrology baseline in the PyHazards flood roadmap.

The adapter consumes the shared graph-temporal streamflow batch format and produces next-step nodewise discharge predictions.

Benchmark Compatibility¶

Primary benchmark family: Flood Benchmark

Mapped benchmark ecosystems: Caravan

External References¶

Paper: Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets | Repo: Repository

Registry Name¶

Primary entrypoint: neuralhydrology_lstm

Supported Tasks¶

Streamflow

Programmatic Use¶

import torch
from pyhazards.models import build_model

model = build_model(name="neuralhydrology_lstm", task="regression", input_dim=2, out_dim=1)
preds = model({"x": torch.randn(1, 4, 6, 2)})
print(preds.shape)

Notes¶

The smoke test uses the shared synthetic streamflow dataset shape.