Post-processing API

Post-processing is a flat set of top-level functions grouped into four namespaces. Each call takes a DataFrame (or a few), runs one calculation and returns a DataFrame or a dict — nothing is written to disk.

Namespace	Page	Covers
`AeroViz.size`	size	`psd_stats`, `psd_distributions`, `merge_psd`, and the `SizeDist` class
`AeroViz.chemistry`	chemistry	mass reconstruction, volumes and refractive index, kappa, partitioning, OC/EC, ISORROPIA
`AeroViz.optical`	optical	IMPROVE, Mie (bulk, lognormal, core-shell, angular), gas extinction, RI retrieval, brown carbon
`AeroViz.voc`	voc	`voc_potentials` (OFP / SOAP / LOH)

Every function is also importable from the top level (from AeroViz import reconstruct_mass). Worked usage lives in the guide: Post-Processing Functions and the four Examples pages.

DataProcess(...) is deprecated

The DataProcess(method=..., path_out=...) factory and its method classes (Chemistry, Optical, SizeDistr, VOC) still exist but emit a DeprecationWarning and will be removed. The migration table below maps every legacy method to its replacement.

Migrating from `DataProcess`

Legacy call	Replacement
`DataProcess('SizeDistr').basic(df)`	`psd_stats(df)` — statistics under `['other']`
`DataProcess('SizeDistr').distributions(df)`	`psd_distributions(df)`
`DataProcess('SizeDistr').merge_SMPS_APS(smps, aps)`	`merge_psd(smps, aps, version=1)`
`DataProcess('SizeDistr').merge_SMPS_APS_v2(smps, aps)`	`merge_psd(smps, aps, version=2)`
`DataProcess('SizeDistr').merge_SMPS_APS_v3(smps, aps)`	`merge_psd(smps, aps, version=3)`
`DataProcess('SizeDistr').merge_SMPS_APS_v4(smps, aps, pm25)`	`merge_psd(smps, aps, df_pm25=pm25, version=4)`
`DataProcess('SizeDistr').dry_psd(df, df_gRH)`	`SizeDist(df).to_dry(df_gRH)`
`DataProcess('SizeDistr').extinction_distribution(df, df_RI)`	`SizeDist(df).to_extinction(df_RI, method='internal')` or `mie(df, ri, distribution=True)`
`DataProcess('SizeDistr').extinction_full(df, df_RI)`	`mie(df, ri)`
`DataProcess('Chemistry').ReConstrc_basic(df)`	`reconstruct_mass(df)`
`DataProcess('Chemistry').volume_average_mixing(df_volume, df_alwc)`	`volume_ri(df_volume, df_alwc=df_alwc)`
`DataProcess('Chemistry').gRH(df_volume, df_alwc)`	`growth_factor(df_volume, df_alwc)`
`DataProcess('Chemistry').kappa(df, diameter)`	`kappa(df, diameter=...)`
`DataProcess('Chemistry').Partition(df)`	`partition_ratios(df)`
`DataProcess('Chemistry').OCEC_basic(df_lcres, df_mass)`	`split_oc_ec(df_lcres, df_mass=...)`
`DataProcess('Chemistry').ISOROPIA(df)`	`isoropia(df_ions, df_met)`
`DataProcess('Optical').basic(df_sca, df_abs, ...)`	`optical_basic(df_sca, df_abs, ...)`
`DataProcess('Optical').IMPROVE(df_mass, df_RH, method)`	`improve(df_mass, df_RH, method=...)`
`DataProcess('Optical').Mie(df_psd, df_m)`	`mie(df_psd, df_m)`
`DataProcess('Optical').gas_extinction(df_no2, df_temp)`	`gas_extinction(df_no2, df_temp)`
`DataProcess('Optical').retrieve_RI(df_optical, df_pnsd)`	`retrieve_ri(df_optical, df_pnsd)`
`DataProcess('Optical').BrC(df_abs, ...)`	`brown_carbon(df_abs, ...)`
`DataProcess('Optical').scaCoe` / `.absCoe`	done inside the readers (`sca_550`, `SAE`, `abs_*`, `AAE`, `eBC` are reader outputs)
`DataProcess('Optical').derived(...)`	no single replacement — its columns were `PG` (sca + abs + gas), `MAC`, `Ox` (NO₂ + O₃), `Vis_cal` (Koschmieder visibility), `fRH_IMPR`, `OCEC_ratio`, `PM1_PM25`; compose them from `optical_basic`, `gas_extinction`, `improve` and plain pandas
`DataProcess('VOC').VOC_basic(df)`	`voc_potentials(df)`

The legacy factory, for reference only:

AeroViz.dataProcess.DataProcess

DataProcess(method: str, path_out: Path, excel: bool = False, csv: bool = True)

Factory function (DEPRECATED) — use the top-level functions instead.

.. deprecated:: DataProcess(...) will be removed in a future release. Each method on the returned Writer instance now has a direct top-level function equivalent. Migration cheatsheet::

    DataProcess('Chemistry', ...).ReConstrc_basic(df, ...)
      → from AeroViz import reconstruct_mass
        reconstruct_mass(df, ...)

    DataProcess('Optical', ...).IMPROVE(df_mass, df_RH, method='revised')
      → from AeroViz import improve
        improve(df_mass, df_RH, method='revised')

    DataProcess('Optical', ...).Mie(df_psd, df_m)
      → from AeroViz import mie
        mie(df_psd, df_m)

    DataProcess('SizeDistr', ...).merge_SMPS_APS_v4(df_smps, df_aps, df_pm25)
      → from AeroViz import merge_psd
        merge_psd(df_smps, df_aps, df_pm25=df_pm25, version=4)

    DataProcess('VOC', ...).VOC_basic(df_voc)
      → from AeroViz import voc_potentials
        voc_potentials(df_voc)

Sub-namespaces (``AeroViz.chemistry``, ``AeroViz.optical``,
``AeroViz.size``, ``AeroViz.voc``) are also supported. The new
functions return DataFrames/dicts directly — pass ``df.to_csv(...)``
yourself if you want files written.

Parameters:

Name	Type	Description	Default
`method`	`str`	The processing method to use. Must be one of: 'Chemistry', 'Optical', 'SizeDistr', or 'VOC'.	required
`path_out`	`Path`	Path where processed output files will be saved.	required
`excel`	`bool`	Whether to save output in Excel format.	`False`
`csv`	`bool`	Whether to save output in CSV format.	`True`

Returns:

Type	Description
`object`	Instance of the selected processing class initialized with the provided parameters.

Raises:

Type	Description
`ValueError`	If the specified method name is not in the supported methods list.

Examples:

>>> from AeroViz import DataProcess
>>> from pathlib import Path
>>> processor = DataProcess(method='Optical', path_out=Path('./results'))  # deprecated

Post-processing API

Migrating from DataProcess

AeroViz.dataProcess.DataProcess

Migrating from `DataProcess`