AeroViz 0.1.6__py3-none-any.whl → 0.1.7__py3-none-any.whl

AeroViz/process/script/IMPROVE.py

@@ -1,161 +0,0 @@
-from pathlib import Path
-from typing import Literal
-
-from pandas import read_csv, concat, read_json
-
-from AeroViz.process.core import DataProc
-from AeroViz.tools.datareader import DataReader
-
-
-class ImproveProc(DataProc):
-    """
-    A class for process improved chemical data.
-
-    Parameters:
-    -----------
-    reset : bool, optional
-        If True, resets the process. Default is False.
-    filename : str, optional
-        The name of the file to process. Default is None.
-    version : str, optional
-        The version of the data process. Should be one of 'revised' or 'modified'.
-        Default is None.
-
-    Methods:
-    --------
-    revised(_df):
-        Calculate revised version of particle contribution.
-
-    modified(_df):
-        Calculate modified version of particle contribution.
-
-    gas(_df):
-        Calculate gas contribution.
-
-    frh(_RH, version=None):
-        Helper function to get frh values based on relative humidity (RH) and version.
-
-    process_data():
-        Process data and save the result.
-
-    Attributes:
-    -----------
-    DEFAULT_PATH : Path
-        The default path for data files.
-
-    Examples:
-    ---------
-    >>> df = ImproveProc(reset=True, filename='revised_IMPROVE.csv', version='revised').process_data()
-
-    """
-
-    def __init__(self, file_paths: list[Path | str] = None):
-        super().__init__()
-        self.file_paths = [Path(fp) for fp in file_paths]
-
-    @staticmethod
-    def frh(_RH):
-        _frh = read_json(Path(__file__).parent.parent.parent / 'plot' / 'utils' / 'fRH.json')
-        if _RH is not None:
-            if _RH > 95:
-                _RH = 95
-            _RH = round(_RH)
-            return _frh.loc[_RH].values.T
-
-        return 1, 1, 1, 1
-
-    def revised(self, _df):
-        def mode(Mass):
-            L_mode = Mass ** 2 / 20 if Mass < 20 else Mass
-            S_mode = Mass - L_mode if Mass < 20 else 0
-
-            return L_mode, S_mode
-
-        _frh, _frhss, _frhs, _frhl = self.frh(_df['RH'])
-
-        L_AS, S_AS = mode(_df['AS'])
-        L_AN, S_AN = mode(_df['AN'])
-        L_OM, S_OM = mode(_df['OM'])
-
-        _df['AS_ext_dry'] = 2.2 * 1 * S_AS + 4.8 * 1 * L_AS
-        _df['AN_ext_dry'] = 2.4 * 1 * S_AN + 5.1 * 1 * L_AN
-        _df['OM_ext_dry'] = 2.8 * S_OM + 6.1 * L_OM
-        _df['Soil_ext_dry'] = 1 * _df['Soil']
-        _df['SS_ext_dry'] = 1.7 * 1 * _df['SS']
-        _df['EC_ext_dry'] = 10 * _df['EC']
-        _df['total_ext_dry'] = sum(_df['AS_ext_dry':'EC_ext_dry'])
-
-        _df['AS_ext'] = (2.2 * _frhs * S_AS) + (4.8 * _frhl * L_AS)
-        _df['AN_ext'] = (2.4 * _frhs * S_AN) + (5.1 * _frhl * L_AN)
-        _df['OM_ext'] = (2.8 * S_OM) + (6.1 * L_OM)
-        _df['Soil_ext'] = (1 * _df['Soil'])
-        _df['SS_ext'] = (1.7 * _frhss * _df['SS'])
-        _df['EC_ext'] = (10 * _df['EC'])
-        _df['total_ext'] = sum(_df['AS_ext':'EC_ext'])
-
-        _df['ALWC_AS_ext'] = _df['AS_ext'] - _df['AS_ext_dry']
-        _df['ALWC_AN_ext'] = _df['AN_ext'] - _df['AN_ext_dry']
-        _df['ALWC_SS_ext'] = _df['SS_ext'] - _df['SS_ext_dry']
-        _df['ALWC_ext'] = _df['total_ext'] - _df['total_ext_dry']
-
-        _df['fRH_IMPR'] = _df['total_ext'] / _df['total_ext_dry']
-
-        return _df['AS_ext_dry':]
-
-    def modified(self, _df):
-        _frh, _frhss, _frhs, _frhl = self.frh(_df['RH'])
-
-        _df['AS_ext_dry'] = 3 * 1 * _df['AS']
-        _df['AN_ext_dry'] = 3 * 1 * _df['AN']
-        _df['OM_ext_dry'] = 4 * _df['OM']
-        _df['Soil_ext_dry'] = 1 * _df['Soil']
-        _df['SS_ext_dry'] = 1.7 * 1 * _df['SS']
-        _df['EC_ext_dry'] = 10 * _df['EC']
-        _df['total_ext_dry'] = sum(_df['AS_ext_dry':'EC_ext_dry'])
-
-        _df['AS_ext'] = (3 * _frh * _df['AS'])
-        _df['AN_ext'] = (3 * _frh * _df['AN'])
-        _df['OM_ext'] = (4 * _df['OM'])
-        _df['Soil_ext'] = (1 * _df['Soil'])
-        _df['SS_ext'] = (1.7 * _frhss * _df['SS'])
-        _df['EC_ext'] = (10 * _df['EC'])
-        _df['total_ext'] = sum(_df['AS_ext':'EC_ext'])
-
-        _df['ALWC_AS_ext'] = _df['AS_ext'] - _df['AS_ext_dry']
-        _df['ALWC_AN_ext'] = _df['AN_ext'] - _df['AN_ext_dry']
-        _df['ALWC_SS_ext'] = _df['SS_ext'] - _df['SS_ext_dry']
-        _df['ALWC_ext'] = _df['total_ext'] - _df['total_ext_dry']
-
-        _df['fRH_IMPR'] = _df['total_ext'] / _df['total_ext_dry']
-
-        return _df['AS_ext_dry':]
-
-    @staticmethod
-    def gas(_df):
-        _df['ScatteringByGas'] = (11.4 * 293 / (273 + _df['AT']))
-        _df['AbsorptionByGas'] = (0.33 * _df['NO2'])
-        _df['ExtinctionByGas'] = _df['ScatteringByGas'] + _df['AbsorptionByGas']
-        return _df['ScatteringByGas':]
-
-    def process_data(self, reset: bool = False, save_file: Path | str = None,
-                     version: Literal["revised", "modified"] = "revised"):
-        save_file = Path(save_file)
-        if save_file.exists() and not reset:
-            return read_csv(save_file, parse_dates=['Time'], index_col='Time')
-        else:
-            # data_files = ['EPB.csv', 'IMPACT.csv', 'chemical.csv']
-            df = concat([DataReader(file) for file in self.file_paths], axis=1)
-
-            # particle contribution '銨不足不納入計算'
-            improve_input_df = df.loc[df['NH4_status'] != 'Deficiency', ['AS', 'AN', 'OM', 'Soil', 'SS', 'EC', 'RH']]
-
-            df_improve = improve_input_df.dropna().copy().apply(self.revised if version == 'revised' else self.modified,
-                                                                axis=1)
-
-            # gas contribution
-            df_ext_gas = df[['NO2', 'AT']].dropna().copy().apply(self.gas, axis=1)
-
-            _df = concat([df_improve, df_ext_gas], axis=1).reindex(df.index.copy())
-            _df.to_csv(save_file)
-
-            return _df

AeroViz/process/script/Others.py

@@ -1,65 +0,0 @@
-from pathlib import Path
-
-import numpy as np
-from pandas import read_csv, concat, DataFrame
-
-from AeroViz.process.core import DataProc
-from AeroViz.tools.datareader import DataReader
-
-
-class OthersProc(DataProc):
-    """
-    A class for process impact data.
-
-    Parameters:
-    -----------
-    reset : bool, optional
-        If True, resets the process. Default is False.
-    filename : str, optional
-        The name of the file to process. Default is None.
-
-    Methods:
-    --------
-    process_data():
-        Process data and save the result.
-
-    Attributes:
-    -----------
-    DEFAULT_PATH : Path
-        The default path for data files.
-
-    Examples:
-    ---------
-    >>> df = OthersProc().process_data(reset=True, filename=None)
-
-    """
-
-    def __init__(self, file_paths: Path | list[Path | str] = None):
-        super().__init__()
-        self.file_paths = [Path(fp) for fp in file_paths]
-
-    def process_data(self, reset: bool = False, save_file: Path | str = None) -> DataFrame:
-        save_file = Path(save_file)
-        if save_file.exists() and not reset:
-            return read_csv(save_file, parse_dates=['Time'], index_col='Time')
-        else:
-            df = concat([DataReader(file) for file in self.file_paths], axis=1)
-
-            results = DataFrame(index=df.index)
-
-            results['PG'] = df[
-                ['Scattering', 'Absorption', 'ScatteringByGas', 'AbsorptionByGas']].dropna().copy().apply(np.sum,
-                                                                                                          axis=1)
-            results['MAC'] = df['Absorption'] / df['T_EC']
-            results['Ox'] = df['NO2'] + df['O3']
-            results['N2O5_tracer'] = df['NO2'] * df['O3']
-            results['Vis_cal'] = 1096 / df['Extinction']
-            # results['fRH_Mix'] = df['Bext'] / df['Extinction']
-            # results['fRH_PNSD'] = df['Bext_internal'] / df['Bext_dry']
-            results['fRH_IMPR'] = df['total_ext'] / df['total_ext_dry']
-            results['OCEC_ratio'] = df['O_OC'] / df['O_EC']
-            results['PM1/PM25'] = np.where(df['PM1'] / df['PM2.5'] < 1, df['PM1'] / df['PM2.5'], np.nan)
-            # results['MEE_PNSD'] = df['Bext_internal'] / df['PM25']
-            # results['MEE_dry_PNSD'] = df['Bext_dry'] / df['PM25']
-
-            return results

AeroViz/process/script/PSD.py

@@ -1,103 +0,0 @@
-from pathlib import Path
-
-from pandas import concat, read_csv, DataFrame
-
-from AeroViz.process.core import DataProc
-from AeroViz.process.core.SizeDist import SizeDist
-from AeroViz.process.script.AbstractDistCalc import DistributionCalculator
-
-
-class ParticleSizeDistProc(DataProc):
-	"""
-    A class for process particle size distribution (PSD) data.
-
-    Parameters
-    ----------
-    filename : str, optional
-        The name of the PSD data file.
-        Defaults to 'PNSD_dNdlogdp.csv' in the default path.
-
-    Attributes
-    ----------
-    file_path : Path
-        The directory path where the PSD data file is located.
-
-    psd : SizeDist
-        The SizeDist object.
-
-    Methods
-    -------
-    process_data(filename='PSD.csv')
-        Process and save overall PSD properties.
-
-    Examples
-    --------
-    Example 1: Use default path and filename
-    >>> psd_data = ParticleSizeDistProc(filename='PNSD_dNdlogdp.csv').process_data(reset=True)
-    """
-
-	def __init__(self, file_path: Path | str = None):
-		super().__init__()
-		self.file_path = Path(file_path)
-
-		self.psd = SizeDist(read_csv(file_path, parse_dates=['Time'], index_col='Time'))
-
-	def process_data(self, reset: bool = False, save_file: Path | str = None) -> DataFrame:
-		save_file = Path(save_file)
-		if save_file.exists() and not reset:
-			return read_csv(save_file, parse_dates=['Time'], index_col='Time')
-
-		number = DistributionCalculator('number', self.psd).useApply()
-		surface = DistributionCalculator('surface', self.psd).useApply()
-		volume = DistributionCalculator('volume', self.psd).useApply()
-
-		surface.to_csv(save_file.parent / 'PSSD_dSdlogdp.csv')
-		volume.to_csv(save_file.parent / 'PVSD_dVdlogdp.csv')
-
-		result_df = concat(
-			[DistributionCalculator('property', SizeDist(data=number, weighting='n')).useApply(),
-			 DistributionCalculator('property', SizeDist(data=surface, weighting='s')).useApply(),
-			 DistributionCalculator('property', SizeDist(data=volume, weighting='v')).useApply()
-			 ], axis=1)
-
-		result_df.to_csv(save_file)
-		return result_df
-
-
-class ExtinctionDistProc(DataProc):
-
-	def __init__(self, file_path: Path | str = 'PNSD_dNdlogdp.csv', file_path_chem: Path | str = 'chemical.csv'):
-		super().__init__()
-		self.file_path = Path(file_path)
-		self.file_path_chem = Path(file_path_chem)
-
-		self.psd = SizeDist(read_csv(file_path, parse_dates=['Time'], index_col='Time'))
-		self.RI = read_csv(file_path_chem, parse_dates=['Time'], index_col='Time')[['n_dry', 'n_amb', 'k_dry', 'k_amb',
-																					'AS_volume_ratio',
-																					'AN_volume_ratio',
-																					'OM_volume_ratio',
-																					'Soil_volume_ratio',
-																					'SS_volume_ratio',
-																					'EC_volume_ratio',
-																					'ALWC_volume_ratio']]
-
-	def process_data(self, reset: bool = False, save_file: Path | str = 'PESD.csv'):
-		save_file = Path(save_file)
-		if save_file.exists() and not reset:
-			return read_csv(save_file, parse_dates=['Time']).set_index('Time')
-
-		ext_internal = DistributionCalculator('extinction', self.psd, self.RI, method='internal',
-											  result_type='extinction').useApply()
-		ext_external = DistributionCalculator('extinction', self.psd, self.RI, method='external',
-											  result_type='extinction').useApply()
-
-		ext_internal.to_csv(save_file.parent / 'PESD_dextdlogdp_internal.csv')
-		ext_external.to_csv(save_file.parent / 'PESD_dextdlogdp_external.csv')
-
-		result_df = concat([
-			DistributionCalculator('property', SizeDist(data=ext_internal, weighting='ext_in')).useApply(),
-			DistributionCalculator('property', SizeDist(data=ext_internal, weighting='ext_ex')).useApply(),
-		], axis=1)
-
-		result_df.to_csv(save_file)
-		return result_df

AeroViz/process/script/PSD_dry.py

@@ -1,93 +0,0 @@
-from pathlib import Path
-
-import numpy as np
-from pandas import DataFrame, read_csv, concat
-
-from AeroViz.process.core import DataProc
-from AeroViz.process.core.SizeDist import SizeDist
-
-
-class DryPSDProc(DataProc):
-    """
-    A class for process impact data.
-
-    Parameters
-    ----------
-    reset : bool, optional
-        If True, resets the process. Default is False.
-    filename : str, optional
-        The name of the file to process. Default is None.
-
-    Methods
-    -------
-    process_data():
-        Process data and save the result.
-
-    Attributes
-    ----------
-    DEFAULT_PATH : Path
-        The default path for data files.
-
-
-    Examples
-    --------
-    >>> df = DryPSDProc(reset=True, filename='PNSD_dNdlogdp_dry.csv').process_data()
-    """
-
-    def __init__(self, file_path: Path | str = 'PNSD_dNdlogdp.csv', file_path_chem: Path | str = 'chemical.csv'):
-        super().__init__()
-        self.file_path = Path(file_path)
-        self.file_path_chem = Path(file_path_chem)
-
-        self.psd = SizeDist(read_csv(file_path, parse_dates=['Time'], index_col='Time'))
-        self.RI = read_csv(file_path_chem, parse_dates=['Time'], index_col='Time')[['n_dry', 'n_amb', 'k_dry', 'k_amb',
-                                                                                    'AS_volume_ratio',
-                                                                                    'AN_volume_ratio',
-                                                                                    'OM_volume_ratio',
-                                                                                    'Soil_volume_ratio',
-                                                                                    'SS_volume_ratio',
-                                                                                    'EC_volume_ratio',
-                                                                                    'ALWC_volume_ratio']]
-
-    def process_data(self, reset: bool = False, save_filename: Path | str = None) -> DataFrame:
-        save_filename = Path(save_filename)
-        if save_filename.exists() and not reset:
-            return read_csv(save_filename, parse_dates=['Time']).set_index('Time')
-        _df = concat([self.psd, self.RI], axis=1)
-        _df.to_csv(save_filename)
-        return _df
-
-
-def dry_PNSD_process(dist, dp, **kwargs):
-    ndp = np.array(dist[:np.size(dp)])
-    gRH = resolved_gRH(dp, dist['gRH'], uniform=True)
-
-    dry_dp = dp / gRH
-    belong_which_ibin = np.digitize(dry_dp, dp) - 1
-
-    result = {}
-    for i, (ibin, dn) in enumerate(zip(belong_which_ibin, ndp)):
-        if dp[ibin] not in result:
-            result[dp[ibin]] = []
-        result[dp[ibin]].append(ndp[i])
-
-    dry_ndp = []
-    for key, val in result.items():
-        dry_ndp.append(sum(val) / len(val))
-
-    return np.array(dry_ndp)
-
-
-def resolved_gRH(dp, gRH=1.31, uniform=True):
-    if uniform:
-        return np.array([gRH] * dp.size)
-
-    else:
-        lognorm_dist = lambda x, geoMean, geoStd: (gRH / (np.log10(geoStd) * np.sqrt(2 * np.pi))) * np.exp(
-            -(x - np.log10(geoMean)) ** 2 / (2 * np.log10(geoStd) ** 2))
-        abc = lognorm_dist(np.log10(dp), 200, 2.0)
-        return np.where(abc < 1, 1, abc)
-
-
-if __name__ == '__main__':
-    pass

AeroViz/process/script/__init__.py

@@ -1,5 +0,0 @@
-from .Chemical import ChemicalProc
-from .IMPACT import ImpactProc
-from .IMPROVE import ImproveProc
-from .Others import OthersProc
-from .PSD import ParticleSizeDistProc, ExtinctionDistProc

AeroViz/process/script/retrieve_RI.py

@@ -1,69 +0,0 @@
-import numpy as np
-from pandas import DataFrame
-
-from AeroViz.process.core.SizeDist import SizeDist
-from AeroViz.process.method import Mie_PESD
-
-
-def retrieve_RI(_df: DataFrame,
-                _PNSD: DataFrame,
-                nMin: float = 1.33,
-                nMax: float = 1.60,
-                kMin: float = 0.00,
-                kMax: float = 0.60,
-                spaceSize: int = 31,
-                dlogdp: float = 0.014
-                ) -> DataFrame:
-    nRange = np.linspace(nMin, nMax, num=spaceSize)
-    kRange = np.linspace(kMin, kMax, spaceSize)
-    Delta_array = np.zeros((spaceSize, spaceSize))
-    # 同一時間除了折射率其餘數據皆相同 因此在折射率的迴圈外
-    bext_mea, bsca_mea, babs_mea = _df['Extinction'], _df['Scattering'], _df['Absorption']
-
-    dp = SizeDist(data=_PNSD).dp
-    for ki, k in enumerate(kRange):
-        for ni, n in enumerate(nRange):
-            m = n + (1j * k)
-            ndp = np.array(_df[3:])
-
-            ext_dist, sca_dist, abs_dist = Mie_PESD(m, 550, dp, ndp)
-
-            bext_cal = sum(ext_dist) * dlogdp
-            bsca_cal = sum(sca_dist) * dlogdp
-            babs_cal = sum(abs_dist) * dlogdp
-
-            Delta_array[ni][ki] = ((babs_mea - babs_cal) / 18.23) ** 2 + ((bsca_mea - bsca_cal) / 83.67) ** 2
-
-    min_delta = Delta_array.argmin()
-    next_n = nRange[(min_delta // spaceSize)]
-    next_k = kRange[(min_delta % spaceSize)]
-
-    # 將網格變小
-    nMin_small = next_n - 0.02 if next_n > 1.33 else 1.33
-    nMax_small = next_n + 0.02
-    kMin_small = next_k - 0.04 if next_k > 0.04 else 0
-    kMax_small = next_k + 0.04
-    spaceSize_small = 41
-
-    nRange_small = np.linspace(nMin_small, nMax_small, spaceSize_small)
-    kRange_small = np.linspace(kMin_small, kMax_small, spaceSize_small)
-    Delta_array_small = np.zeros((spaceSize_small, spaceSize_small))
-    # 所有數據與大網格一致所以使用上方便數即可
-    for ki, k in enumerate(kRange_small):
-        for ni, n in enumerate(nRange_small):
-            m = n + (1j * k)
-            ndp = np.array(_df[3:])
-            ext_dist, sca_dist, abs_dist = Mie_PESD(m, 550, dp, ndp)
-
-            bext_cal = sum(ext_dist) * dlogdp
-            bsca_cal = sum(sca_dist) * dlogdp
-            babs_cal = sum(abs_dist) * dlogdp
-
-            Delta_array_small[ni][ki] = ((bext_mea - bext_cal) / 18.23) ** 2 + ((bsca_mea - bsca_cal) / 83.67) ** 2
-
-    min_delta_small = Delta_array_small.argmin()
-    _df['re_real'] = (nRange_small[(min_delta_small // spaceSize_small)])
-    _df['re_imaginary'] = (kRange_small[(min_delta_small % spaceSize_small)])
-
-    print(f'\t\tReal part:{_df['re_real']}\tIm part:{_df['re_imaginary']}', end='')
-    return _df['re_real':]

AeroViz/rawDataReader/script/EPA_vertical.py

@@ -1,46 +0,0 @@
-import numpy as np
-from pandas import read_csv, to_numeric
-
-from AeroViz.rawDataReader.core import AbstractReader
-
-
-class Reader(AbstractReader):
-    nam = 'EPA_vertical'
-
-    def _raw_reader(self, file):
-        with file.open('r', encoding='ascii', errors='ignore') as f:
-            # 有、無輸出有效值都可以
-            # read 查詢小時值(測項).csv
-            df = read_csv(f, encoding='ascii', encoding_errors='ignore', index_col=0, parse_dates=True,
-                          usecols=lambda col: col != 'Unnamed: 1')
-
-            df.index.name = 'Time'
-            df.rename(columns={'AMB_TEMP': 'AT', 'WIND_SPEED': 'WS', 'WIND_DIREC': 'WD'}, inplace=True)
-
-            # 欄位排序
-            desired_order = ['SO2', 'NO', 'NOx', 'NO2', 'CO', 'O3', 'THC', 'NMHC', 'CH4', 'PM10', 'PM2.5', 'WS', 'WD',
-                             'AT', 'RH']
-
-            missing_columns = []
-
-            for col in desired_order:
-                if col not in df.columns:
-                    df[col] = np.nan
-                    missing_columns.append(col)
-
-            if missing_columns:
-                self.logger.info(f"{'=' * 60}")
-                self.logger.info(f"Missing columns: {missing_columns}")
-                self.logger.info(f"{'=' * 60}")
-                print(f"Missing columns: {missing_columns}")
-
-            df = df[desired_order]
-
-            # 如果沒有將無效值拿掉就輸出 請將包含 #、L、O 的字串替換成 *
-            df.replace(to_replace=r'\d*[#LO]\b', value='*', regex=True, inplace=True)
-            df = df.apply(to_numeric, errors='coerce')
-
-        return df
-
-    def _QC(self, _df):
-        return _df

AeroViz/rawDataReader/script/Table.py

@@ -1,27 +0,0 @@
-# read meteorological data from google sheet
-
-from pandas import read_csv, to_datetime
-
-from AeroViz.rawDataReader.core import AbstractReader
-
-
-class Reader(AbstractReader):
-    nam = 'Table'
-
-    def _raw_reader(self, file):
-        with file.open('r', encoding='utf-8-sig', errors='ignore') as f:
-            _df = read_csv(f, low_memory=False, index_col=0)
-
-            _df.index = to_datetime(_df.index, errors='coerce')
-            _df.index.name = 'time'
-
-            _df.columns = _df.keys().str.strip(' ')
-
-        return _df.loc[~_df.index.duplicated() & _df.index.notna()]
-
-    def _QC(self, _df):
-        # remove negative value
-        _df = _df.mask((_df < 0).copy())
-
-        # QC data in 6h
-        return _df.resample('6h').apply(self.basic_QC).resample(self.meta.get("freq")).mean()