PhaseNet¶

Overview¶

phasenet is the first earthquake picking baseline in the staged PyHazards roadmap and is paired with the synthetic waveform dataset for smoke validation.

At a Glance¶

Hazard Family

Earthquake

Public catalog grouping used for this model.

Maturity

Implemented

Catalog maturity label used on the index page.

Tasks

Phase Picking

Benchmark Family

Earthquake Benchmark

Primary benchmark-family link used for compatible evaluation coverage.

Description¶

phasenet is the first earthquake picking baseline in the staged PyHazards roadmap and is paired with the synthetic waveform dataset for smoke validation.

This initial adapter focuses on the shared waveform-to-pick interface and does not claim exact reproduction of the original PhaseNet training stack.

Benchmark Compatibility¶

Primary benchmark family: Earthquake Benchmark

Mapped benchmark ecosystems: SeisBench

External References¶

Paper: PhaseNet: A Deep-Neural-Network-Based Seismic Arrival Time Picking Method | Repo: Repository

Registry Name¶

Primary entrypoint: phasenet

Supported Tasks¶

Phase Picking

Programmatic Use¶

import torch
from pyhazards.models import build_model

model = build_model(
    name="phasenet",
    task="regression",
    in_channels=3,
)

picks = model(torch.randn(4, 3, 256))
print(picks.shape)

Notes¶

Outputs are P- and S-arrival sample indices in the current smoke-test adapter.