.. _deepc-available-models:

Available trained models
========================

gempa :cite:`gempa` provides many models trained on and applied to different
data sets at different scales of epicentral distances, source depths and source
types. They are listed in table :ref:`deepc-models`.
Use these model names for configuration of `modelWeights` which exists for each
sub-picker:

* :confval:`picker.deepc.deepcEC.modelWeights`
* :confval:`picker.deepc.deepcPC.modelWeights`
* :confval:`picker.deepc.deepcPR.modelWeights`
* :confval:`spicker.deepcSR.modelWeights`

The corresponding DL models are installed with the *dlmodels* package. When
installing *deepc* by gsm, *dlmodels* is installed automatically.
A typical installation directory is :file:`/home/data/dlmodels`.


Event classification models
---------------------------

Currently, gempa doesn't provide a model for event classification.

.. _deepc-event-classification-models:

.. csv-table::
   :header: model name, event types, applications, training data
   :widths: 15, 30, 30, 25
   :delim: ;

   pncx-if-3; earthquake, anthropogenic event, mining; signals from earthquakes and machinery in mining environment at epicentral distance of tens of meters up to 10 km; mine
   pncx-if-10; earthquake, anthropogenic event, mining; signals from earthquakes and machinery in mining environment at epicentral distance of tens of meters up to 10 km; mine


Phase classification models
---------------------------

.. _deepc-phase-classification-models:

.. csv-table::
   :header: model name, specifics, applications, training data
   :widths: 15, 30, 30, 25
   :delim: ;

   pnc-cb-1_4; P, S, Noise; (re-) classify phase hints for picks; Combined (INSTANCE :cite:`seisbench-data-in`, SCEDC :cite:`seisbench-data-sc` and STEAD :cite:`seisbench-data-st`)


Joint time prediction and phase classification models
-----------------------------------------------------

The Seismology Benchmark collection, SeisBench :cite:`M_nchmeyer_2022`, has
published many pre-trained picker models.
|gempa| provides converted versions of some of these models and also in-house
trained ones.
These models predict probabilities for P, S, and Noise for each sample of a
trace.
|deepc| post-processes their predictions to phase classification and onset
refinement of a given trigger.


EqTransformer
^^^^^^^^^^^^^

The following models are trained on the SeisBench implementation of
EqTransformer :cite:`Mousavi_2020`.

.. _deepc-phase-classification-timepred-models-eqt:

.. csv-table::
   :header: model name, specifics, applications, training data
   :widths: 15, 30, 30, 25
   :delim: ;

   eq-et; 60 s, 100 Hz; regional data; `ETHZ` :cite:`seisbench-data-et`
   eq-ge; 60 s, 100 Hz; teleseismic data; `GEOFON` :cite:`seisbench-data-ge`
   eq-in; 60 s, 100 Hz; regional data; `INSTANCE` :cite:`seisbench-data-in`
   eq-iq; 60 s, 100 Hz; regional data; `Iquique` :cite:`seisbench-data-iq`
   eq-ne; 60 s, 100 Hz; regional data; `NEIC` :cite:`seisbench-data-ne`
   eq-or; 60 s, 100 Hz; regional data; original EqTransformer weights :cite:`Mousavi_2020`
   eq-sc; 60 s, 100 Hz; regional data; `SCEDC` :cite:`seisbench-data-sc`
   eq-st; 60 s, 100 Hz; regional data; `STEAD` :cite:`seisbench-data-st`


PhaseNet
^^^^^^^^

The following models are trained on the SeisBench adaptions of PhaseNet
:cite:`Zhu_2018`.

.. _deepc-phase-classification-timepred-models-pn:

.. csv-table::
   :header: model name, specifics, applications, training data
   :widths: 15, 30, 30, 25
   :delim: ;

   pn-et; 30.01 s, 100 Hz; regional data; `ETHZ` :cite:`seisbench-data-et`
   pn-ge; 30.01 s, 100 Hz; teleseismic data; `GEOFON` :cite:`seisbench-data-ge`
   pn-in; 30.01 s, 100 Hz; regional data; `INSTANCE` :cite:`seisbench-data-in`
   pn-iq; 30.01 s, 100 Hz; regional data; `Iquique` :cite:`seisbench-data-iq`
   pn-ne; 30.01 s, 100 Hz; regional data; `NEIC` :cite:`seisbench-data-ne`
   pn-or; 30.01 s, 100 Hz; regional data; original PhaseNet weights :cite:`Zhu_2018`
   pn-sc; 30.01 s, 100 Hz; regional data; `SCEDC` :cite:`seisbench-data-sc`
   pn-st; 30.01 s, 100 Hz; regional data; `STEAD` :cite:`seisbench-data-st`
   pn-in_vl-74; 30.01 s, 100 Hz; regional data; `INSTANCE` :cite:`seisbench-data-in`, Vogtland
   pn-in_vl-97; 30.01 s, 100 Hz, all-comps-normalized; regional data; `INSTANCE` :cite:`seisbench-data-in`, Vogtland
   pn-vl-91; 30.01 s, 100 Hz, all-comps-normalized; regional data; Vogtland
   pn-vp; 30.01 s, 100 Hz; volcanic data; `VCSEIS` :cite:`seisbench-data-vc`
   pnv-bq-0; 60 s, 100 Hz; volcanic data?; BQCombined
   pnv-bq-10; 30 s, 100 Hz; volcanic data?; BQCombined
   pnv-il_vc-2; 15 s, 100 Hz; volcanic data?; Iceland, `VCSEIS_hawaii` :cite:`seisbench-data-vc`
   pnv_15s-et-0; 15 s, 100 Hz; regional; `ETHZ` :cite:`seisbench-data-et`
   pnv_15s-il-2; 15 s, 100 Hz; regional; Iceland
   pnv_30s-et-sb; 30 s, 100 Hz; regional; Iceland
   pnv_60s-et-0; 60 s, 100 Hz; regional; Iceland