AeroViz 0.1.3__py3-none-any.whl → 0.1.4__py3-none-any.whl

AeroViz/plot/templates/templates.py

@@ -1,398 +0,0 @@
-from typing import Literal
-
-import matplotlib.pyplot as plt
-import numpy as np
-import pandas as pd
-import seaborn as sns
-from matplotlib.pyplot import Figure, Axes
-from pandas import DataFrame
-
-from AeroViz.plot.utils import *
-
-__all__ = [
-	'pie',
-	'donuts',
-	'violin',
-	'bar',
-]
-
-
-def _auto_label_pct(pct,
-					symbol: bool = True,
-					include_pct: bool = False,
-					ignore: Literal["inner", "outer"] = 'inner',
-					value: float = 2):
-	if not symbol:
-		return ''
-	cond = pct <= value if ignore == 'inner' else pct > value
-	label = '' if cond else '{:.1f}'.format(pct)
-	return '' if label == '' else label + '%' if include_pct else label
-
-
-@set_figure(fs=8, fw='bold')
-def pie(data_set: DataFrame | dict,
-		labels: list[str],
-		unit: str,
-		style: Literal["pie", 'donut'],
-		ax: Axes | None = None,
-		symbol: bool = True,
-		**kwargs) -> tuple[Figure, Axes]:
-	"""
-	Create a pie or donut chart based on the provided data.
-
-	Parameters
-	----------
-	data_set : pd.DataFrame | dict
-		A pandas DataFrame or dictionary mapping category names to a list of species.
-		If a DataFrame is provided, the index represents the categories, and each column contains species data.
-		If a dictionary is provided, it maps category names to lists of species data.
-		It is assumed that all lists or DataFrame columns contain the same number of entries as the *labels* list.
-	labels : list of str
-		The labels for each category.
-	unit : str
-		The unit to display in the center of the donut chart.
-	style : Literal["pie", 'donut']
-		The style of the chart, either 'pie' for a standard pie chart or 'donut' for a donut chart.
-	ax : plt.Axes or None, optional
-		The Axes object to plot the chart onto. If None, a new figure and Axes will be created.
-	symbol : bool, optional
-		Whether to display values for each species in the chart.
-	**kwargs
-		Additional keyword arguments to be passed to the plotting function.
-
-	Returns
-	-------
-	matplotlib.axes.Axes
-		The Axes object containing the violin plot.
-
-	Notes
-	-----
-	- If *data_set* is a dictionary, it should contain lists of species that correspond to each category in *labels*.
-	- The length of each list in *data_set* or the number of columns in the DataFrame should match the length of the *labels* list.
-
-	Examples
-	--------
-	>>> data_set = {'Category 1': [10, 20, 30], 'Category 2': [15, 25, 35]}
-	>>> labels = ['Species 1', 'Species 2', 'Species 3']
-	>>> pie(data_set, labels, unit='kg', style='pie', symbol=True)
-	"""
-	if isinstance(data_set, DataFrame):
-		category_names = list(data_set.index)
-		data = data_set.to_numpy()
-
-		pies, species = data.shape
-
-	elif isinstance(data_set, dict):
-		category_names = list(data_set.keys())
-		data = np.array(list(data_set.values()))
-
-		pies, species = data.shape
-
-	else:
-		raise ValueError('data_set must be a DataFrame or a dictionary.')
-
-	colors = kwargs.get('colors') or (Color.colors1 if species == 6 else Color.getColor(num=species))
-
-	radius = 4
-	width = 4 if style == 'pie' else 1
-
-	text = [''] * pies if style == 'pie' else [Unit(unit) + '\n\n' + '{:.2f}'.format(x) for x in data.sum(axis=1)]
-	pct_distance = 0.6 if style == 'pie' else 0.88
-
-	fig, ax = plt.subplots(1, pies, figsize=((pies * 2) + 1, 2)) if ax is None else (ax.get_figure(), ax)
-
-	if pies == 1:
-		ax = [ax]
-
-	for i in range(pies):
-		ax[i].pie(data[i], labels=None, colors=colors, textprops=None,
-				  autopct=lambda pct: _auto_label_pct(pct, symbol=symbol, include_pct=True),
-				  pctdistance=pct_distance, radius=radius, wedgeprops=dict(width=width, edgecolor='w'))
-
-		ax[i].pie(data[i], labels=None, colors=colors, textprops=None,
-				  autopct=lambda pct: _auto_label_pct(pct, symbol=symbol, ignore='outer', include_pct=True),
-				  pctdistance=1.3, radius=radius, wedgeprops=dict(width=width, edgecolor='w'))
-		ax[i].axis('equal')
-		ax[i].text(0, 0, text[i], ha='center', va='center')
-		ax[i].set_title(category_names[i])
-
-	ax[-1].legend(labels, loc='center left', prop={'weight': 'bold'}, bbox_to_anchor=(1, 0, 1.15, 1))
-
-	# fig.savefig(f"pie_{style}_{title}")
-
-	plt.show()
-
-	return fig, ax
-
-
-@set_figure(fs=8, fw='bold')
-def donuts(data_set: DataFrame | dict,
-		   labels: list[str],
-		   unit: str,
-		   ax: Axes | None = None,
-		   symbol=True,
-		   **kwargs) -> tuple[Figure, Axes]:
-	"""
-	Plot a donut chart based on the data set.
-
-	Parameters
-	----------
-	data_set : pd.DataFrame | dict
-		A pandas DataFrame or a dictionary mapping category names to a list of species.
-		If a DataFrame is provided, the index represents the categories, and each column contains species data.
-		If a dictionary is provided, it maps category names to lists of species data.
-		It is assumed that all lists or DataFrame columns contain the same number of entries as the *labels* list.
-	labels : list of str
-		The category labels.
-	unit : str
-		The unit to be displayed in the center of the donut chart.
-	ax : matplotlib.axes.Axes, optional
-		The axes to plot on. If None, the current axes will be used (default).
-	symbol : bool, optional
-		Whether to display values for each species (default is True).
-	**kwargs : dict, optional
-		Additional keyword arguments to pass to the matplotlib pie chart function.
-
-	Returns
-	-------
-	matplotlib.axes.Axes
-		The axes containing the donut chart.
-	"""
-
-	if isinstance(data_set, DataFrame):
-		category_names = list(data_set.index)
-		data = data_set.to_numpy()
-
-		pies, species = data.shape
-
-	elif isinstance(data_set, dict):
-		category_names = list(data_set.keys())
-		data = np.array(list(data_set.values()))
-
-		pies, species = data.shape
-
-	else:
-		raise ValueError('data_set must be a DataFrame or a dictionary.')
-
-	colors1 = kwargs.get('colors') or (Color.colors1 if species == 6 else Color.getColor(num=species))
-	colors2 = Color.adjust_opacity(colors1, 0.8)
-	colors3 = Color.adjust_opacity(colors1, 0.6)
-
-	fig, ax = plt.subplots(**kwargs.get('fig_kws', {})) if ax is None else (ax.get_figure(), ax)
-
-	ax.pie(data[2], labels=None, colors=colors1, textprops=None,
-		   autopct=lambda pct: _auto_label_pct(pct, symbol=symbol, include_pct=True),
-		   pctdistance=0.9, radius=14, wedgeprops=dict(width=3, edgecolor='w'))
-
-	ax.pie(data[1], labels=None, colors=colors2, textprops=None,
-		   autopct=lambda pct: _auto_label_pct(pct, symbol=symbol, include_pct=True),
-		   pctdistance=0.85, radius=11, wedgeprops=dict(width=3, edgecolor='w'))
-
-	ax.pie(data[0], labels=None, colors=colors3, textprops=None,
-		   autopct=lambda pct: _auto_label_pct(pct, symbol=symbol, include_pct=True),
-		   pctdistance=0.80, radius=8, wedgeprops=dict(width=3, edgecolor='w'))
-
-	text = (Unit(f'{unit}') + '\n\n' +
-			'Event : ' + "{:.2f}".format(np.sum(data[2])) + '\n' +
-			'Transition : ' + "{:.2f}".format(np.sum(data[1])) + '\n' +
-			'Clean : ' + "{:.2f}".format(np.sum(data[0])))
-
-	ax.text(0, 0, text, ha='center', va='center')
-	ax.axis('equal')
-
-	ax.set_title(kwargs.get('title', ''))
-
-	ax.legend(labels, loc='center', prop={'weight': 'bold'}, title_fontproperties={'weight': 'bold'},
-			  title=f'Outer : {category_names[2]}' + '\n' + f'Middle : {category_names[1]}' + '\n' + f'Inner : {category_names[0]}',
-			  bbox_to_anchor=(0.8, 0, 0.5, 1))
-
-	# fig.savefig(f"donuts_{title}")
-
-	plt.show()
-
-	return fig, ax
-
-
-@set_figure(figsize=(5, 4))
-def bar(data_set: DataFrame | dict,
-		data_std: DataFrame | None,
-		labels: list[str],
-		unit: str,
-		style: Literal["stacked", "dispersed"] = "dispersed",
-		orientation: Literal["va", "ha"] = 'va',
-		ax: Axes | None = None,
-		symbol=True,
-		**kwargs
-		) -> tuple[Figure, Axes]:
-	"""
-	Parameters
-	----------
-	data_set : pd.DataFrame or dict
-		A mapping from category names to a list of species mean or a DataFrame with columns as categories and values as means.
-	data_std : pd.DataFrame or None
-		A DataFrame with standard deviations corresponding to data_set, or None if standard deviations are not provided.
-	labels : list of str
-		The species names.
-	unit : str
-		The unit for the values.
-	style : {'stacked', 'dispersed'}, default 'dispersed'
-		Whether to display the bars stacked or dispersed.
-	orientation : {'va', 'ha'}, default 'va'
-		The orientation of the bars, 'va' for vertical and 'ha' for horizontal.
-	ax : plt.Axes or None, default None
-		The Axes object to plot on. If None, a new figure and Axes are created.
-	symbol : bool, default True
-		Whether to display values for each bar.
-	kwargs : dict
-		Additional keyword arguments passed to the barplot function.
-
-	Returns
-	-------
-	matplotlib.Axes
-		The Axes object containing the plot.
-
-	"""
-	# data process
-	data = data_set.values
-
-	if data_std is None:
-		data_std = np.zeros(data.shape)
-	else:
-		data_std = data_std.values
-
-	groups, species = data.shape
-	groups_arr = np.arange(groups)
-	species_arr = np.arange(species)
-
-	total = np.array([data.sum(axis=1), ] * species).T
-
-	pct_data = data / total * 100
-	data_cum = pct_data.cumsum(axis=1)
-
-	# figure info
-	category_names = kwargs.get('ticks') or list(data_set.index)
-	title = kwargs.get('title', '')
-	colors = kwargs.get('colors') or (Color.colors1 if species == 6 else Color.getColor(num=species))
-
-	fig, ax = plt.subplots(**kwargs.get('fig_kws', {})) if ax is None else (ax.get_figure(), ax)
-
-	if style == "stacked":
-		for i in range(species):
-			widths = pct_data[:, i]
-			starts = data_cum[:, i] - pct_data[:, i]
-
-			if orientation == 'va':
-				_ = ax.bar(groups_arr, widths, bottom=starts, width=0.7, color=colors[i], label=labels[i],
-						   edgecolor=None, capsize=None)
-			if orientation == 'ha':
-				_ = ax.barh(groups_arr, widths, left=starts, height=0.7, color=colors[i], label=labels[i],
-							edgecolor=None, capsize=None)
-			if symbol:
-				ax.bar_label(_, fmt=_auto_label_pct, label_type='center', padding=0, fontsize=10, weight='bold')
-
-	if style == "dispersed":
-		width = 0.1
-		block = width / 4
-
-		for i in range(species):
-			val = data[:, i]
-			std = (0,) * groups, data_std[:, i]
-			if orientation == 'va':
-				_ = ax.bar(groups_arr + (i + 1) * (width + block), val, yerr=std, width=width, color=colors[i],
-						   edgecolor=None, capsize=None)
-			if orientation == 'ha':
-				_ = ax.barh(groups_arr + (i + 1) * (width + block), val, xerr=std, height=width, color=colors[i],
-							edgecolor=None, capsize=None)
-			if symbol:
-				ax.bar_label(_, fmt=_auto_label_pct, label_type='center', padding=0, fontsize=8, weight='bold')
-
-	if orientation == 'va':
-		xticks = groups_arr + (species / 2 + 0.5) * (width + block) if style == "dispersed" else groups_arr
-		ax.set_xticks(xticks, category_names, weight='bold')
-		ax.set_ylabel(Unit(unit) if style == "dispersed" else '$Contribution (\\%)$')
-		ax.set_ylim(0, None if style == "dispersed" else 100)
-		ax.legend(labels, bbox_to_anchor=(1, 1), loc='upper left', prop={'size': 12})
-
-	if orientation == 'ha':
-		ax.invert_yaxis()
-		yticks = groups_arr + 3.5 * (width + block) if style == "dispersed" else groups_arr
-		ax.set_yticks(yticks, category_names, weight='bold')
-		ax.set_xlabel(Unit(unit) if style == "dispersed" else '$Contribution (\\%)$')
-		ax.set_xlim(0, None if style == "dispersed" else 100)
-		ax.legend(labels, bbox_to_anchor=(1, 1), loc='upper left', prop={'size': 12})
-
-	# fig.savefig(f"Barplot_{title}")
-
-	plt.show()
-
-	return fig, ax
-
-
-@set_figure
-def violin(data_set: DataFrame | dict,
-		   unit: str,
-		   ax: Axes | None = None,
-		   **kwargs
-		   ) -> tuple[Figure, Axes]:
-	"""
-	Generate a violin plot for multiple data sets.
-
-	Parameters
-	----------
-	data_set : pd.DataFrame or dict
-		A mapping from category names to pandas DataFrames containing the data.
-	unit : str
-		The unit for the data being plotted.
-	ax : matplotlib.axes.Axes, optional
-		The Axes object to draw the plot onto. If not provided, a new figure will be created.
-	**kwargs : dict
-		Additional keyword arguments to be passed to the violinplot function.
-
-	Returns
-	-------
-	matplotlib.axes.Axes
-		The Axes object containing the violin plot.
-
-	"""
-	fig, ax = plt.subplots(**kwargs.get('fig_kws', {})) if ax is None else (ax.get_figure(), ax)
-
-	data = data_set.to_numpy()
-
-	data = data[~np.isnan(data).any(axis=1)]
-
-	grps = data.shape[1]
-
-	width = 0.6
-	block = width / 2
-	x_position = np.arange(grps)
-
-	plt.boxplot(data, positions=x_position, widths=0.15,
-				showfliers=False, showmeans=True, meanline=False, patch_artist=True,
-				capprops=dict(linewidth=0),
-				whiskerprops=dict(linewidth=1.5, color='k', alpha=1),
-				boxprops=dict(linewidth=1.5, color='k', facecolor='#4778D3', alpha=1),
-				meanprops=dict(marker='o', markeredgecolor='black', markerfacecolor='white', markersize=6),
-				medianprops=dict(linewidth=1.5, ls='-', color='k', alpha=1))
-
-	sns.violinplot(data=data, density_norm='area', color='#4778D3', inner=None)
-
-	for violin, alpha in zip(ax.collections[:], [0.5] * len(ax.collections[:])):
-		violin.set_alpha(alpha)
-		violin.set_edgecolor(None)
-
-	plt.scatter(x_position, data.mean(), marker='o', facecolor='white', edgecolor='k', s=10)
-
-	xlim = kwargs.get('xlim') or (x_position[0] - (width / 2 + block), x_position[-1] + (width / 2 + block))
-	ylim = kwargs.get('ylim') or (0, None)
-	xlabel = kwargs.get('xlabel') or ''
-	ylabel = kwargs.get('ylabel') or Unit(unit)
-	xticks = kwargs.get('xticks') or [x.replace('-', '\n') for x in list(data_set.keys())]
-
-	ax.set(xlim=xlim, ylim=ylim, xlabel=xlabel, ylabel=ylabel, title=kwargs.get('title'))
-	ax.set_xticks(x_position, xticks, fontweight='bold', fontsize=12)
-
-	# fig.savefig(f'Violin_{unit}')
-
-	plt.show()
-
-	return fig, ax

AeroViz/plot/utils/_decorator.py

@@ -1,74 +0,0 @@
-from functools import wraps
-
-import matplotlib.pyplot as plt
-
-__all__ = ['set_figure']
-
-
-# For more details please see https://matplotlib.org/stable/users/explain/customizing.html
-
-
-def set_figure(func=None,
-			   *,
-			   figsize: tuple | None = None,
-			   fs: int | None = None,
-			   fw: str = None,
-			   autolayout: bool = True
-			   ):
-	def decorator(_func):
-		@wraps(_func)
-		def wrapper(*args, **kwargs):
-			print(f'\t\t Plot: \033[96m{_func.__name__}\033[0m')
-
-			plt.rcParams['mathtext.fontset'] = 'custom'
-			plt.rcParams['mathtext.rm'] = 'Times New Roman'
-			plt.rcParams['mathtext.it'] = 'Times New Roman: italic'
-			plt.rcParams['mathtext.bf'] = 'Times New Roman: bold'
-			plt.rcParams['mathtext.default'] = 'regular'
-
-			# The font properties used by `text.Text`.
-			# The text, annotate, label, title, ticks, are used to create text
-			plt.rcParams['font.family'] = 'Times New Roman'
-			plt.rcParams['font.weight'] = fw or 'normal'
-			plt.rcParams['font.size'] = fs or 8
-
-			plt.rcParams['axes.titlesize'] = 'large'
-			plt.rcParams['axes.titleweight'] = 'bold'
-			plt.rcParams['axes.labelweight'] = 'bold'
-
-			# color
-			plt.rcParams['axes.prop_cycle'] = plt.cycler(color=['b', 'g', 'r', 'c', 'm', 'y', 'k'])
-
-			plt.rcParams['xtick.labelsize'] = 'medium'
-			plt.rcParams['ytick.labelsize'] = 'medium'
-
-			# matplotlib.font_manager.FontProperties ---> matplotlib.rcParams
-			plt.rcParams['legend.loc'] = 'best'
-			plt.rcParams['legend.frameon'] = False
-			plt.rcParams['legend.fontsize'] = 'small'
-			plt.rcParams['legend.title_fontsize'] = 'medium'
-			plt.rcParams['legend.handlelength'] = 1.5
-			plt.rcParams['legend.labelspacing'] = 0.7
-
-			plt.rcParams['figure.figsize'] = figsize or (5, 4)
-			plt.rcParams['figure.dpi'] = 200
-			plt.rcParams['figure.autolayout'] = autolayout
-
-			if ~autolayout:
-				plt.rcParams['figure.subplot.left'] = 0.1
-				plt.rcParams['figure.subplot.right'] = 0.875
-				plt.rcParams['figure.subplot.top'] = 0.875
-				plt.rcParams['figure.subplot.bottom'] = 0.125
-
-			# plt.rcParams['figure.constrained_layout.use'] = True
-
-			plt.rcParams['savefig.transparent'] = True
-
-			return _func(*args, **kwargs)
-
-		return wrapper
-
-	if func is None:
-		return decorator
-
-	return decorator(func)

AeroViz/rawDataReader/script/IGAC_TH.py

@@ -1,104 +0,0 @@
-# read meteorological data from google sheet
-
-
-import numpy as np
-from pandas import read_csv, concat, to_datetime
-
-from AeroViz.rawDataReader.core import AbstractReader
-
-
-class Reader(AbstractReader):
-	nam = 'IGAC_TH'
-
-	def _raw_reader(self, _file):
-
-		self.meta['freq'] = self._oth_set.get('data_freq') or self.meta['freq']
-
-		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', format=self._oth_set.get('date_format') or 'mixed')
-			_df.index.name = 'time'
-
-			_df.columns = _df.keys().str.strip(' ')
-
-			_df = _df.loc[_df.index.dropna()].copy()
-
-		return _df.loc[~_df.index.duplicated()]
-
-	## QC data
-	def _QC(self, _df):
-
-		## QC parameter, function (MDL SE LE)
-		_mdl = {
-			'Na+': 0.05,
-			'NH4+': 0.05,
-			'K+': 0.05,
-			'Mg2+': 0.05,
-			'Ca2+': 0.05,
-			'Cl-': 0.05,
-			'NO2-': 0.05,
-			'NO3-': 0.05,
-			'SO42-': 0.05,
-		}
-
-		def _se_le(_df_, _log=False):
-			_df_ = np.log10(_df_) if _log else _df_
-
-			_df_qua = _df_.quantile([.25, .75])
-			_df_q1, _df_q3 = _df_qua.loc[.25].copy(), _df_qua.loc[.75].copy()
-			_df_iqr = _df_q3 - _df_q1
-
-			_se = concat([_df_q1 - 1.5 * _df_iqr] * len(_df_), axis=1).T.set_index(_df_.index)
-			_le = concat([_df_q3 + 1.5 * _df_iqr] * len(_df_), axis=1).T.set_index(_df_.index)
-
-			if _log:
-				return 10 ** _se, 10 ** _le
-			return _se, _le
-
-		_cation, _anion, _main = ['Na+', 'NH4+', 'K+', 'Mg2+', 'Ca2+'], ['Cl-', 'NO2-', 'NO3-', 'SO42-', ], ['SO42-',
-																											 'NO3-',
-																											 'NH4+']
-
-		_df_salt = _df[_mdl.keys()].copy()
-		_df_pm = _df['PM2.5'].copy()
-
-		## lower than PM2.5
-		## conc. of main salt should be present at the same time (NH4+, SO42-, NO3-)
-		_df_salt = _df_salt.mask(_df_salt.sum(axis=1, min_count=1) > _df_pm).dropna(subset=_main).copy()
-
-		## mdl
-		for (_key, _df_col), _mdl_val in zip(_df_salt.items(), _mdl.values()):
-			_df_salt[_key] = _df_col.mask(_df_col < _mdl_val, _mdl_val / 2)
-
-		## group by time (per month)
-		_df_salt['tm'] = _df_salt.index.strftime('%Y-%m')
-
-		_df_lst = []
-		for _ky, _df_grp in _df_salt.groupby('tm'):
-			_df_grp = _df_grp[_mdl.keys()].copy()
-
-			## calculate SE LE
-			## salt < LE
-			_se, _le = _se_le(_df_grp, _log=True)
-			_df_grp = _df_grp.mask(_df_grp > _le).copy()
-
-			## C/A, A/C
-			_rat_CA = (_df_grp[_cation].sum(axis=1) / _df_grp[_anion].sum(axis=1)).to_frame()
-			_rat_AC = (1 / _rat_CA).copy()
-
-			_se, _le = _se_le(_rat_CA, )
-			_cond_CA = (_rat_CA < _le) & (_rat_CA > 0)
-
-			_se, _le = _se_le(_rat_AC, )
-			_cond_AC = (_rat_AC < _le) & (_rat_AC > 0)
-
-			_df_grp = _df_grp.where((_cond_CA * _cond_AC)[0]).copy()
-
-			## conc. of main salt > SE
-			_se, _le = _se_le(_df_grp[_main], _log=True)
-			_df_grp[_main] = _df_grp[_main].mask(_df_grp[_main] < _se).copy()
-
-			_df_lst.append(_df_grp)
-
-		return concat(_df_lst).reindex(_df.index)

AeroViz/rawDataReader/script/IGAC_ZM.py

@@ -1,90 +0,0 @@
-# read meteorological data from google sheet
-
-
-import numpy as np
-from pandas import read_csv, concat, to_numeric
-
-from AeroViz.rawDataReader.core import AbstractReader
-
-
-class Reader(AbstractReader):
-	nam = 'IGAC_ZM'
-
-	def _raw_reader(self, _file):
-
-		with (_file).open('r', encoding='utf-8-sig', errors='ignore') as f:
-			_df = read_csv(f, parse_dates=[0], index_col=[0], na_values=['-']).apply(to_numeric, errors='coerce')
-
-			_df.columns = _df.keys().str.strip(' ')
-			_df.index.name = 'time'
-
-		return _df.loc[_df.index.dropna()].loc[~_df.index.duplicated()]
-
-	## QC data
-	def _QC(self, _df):
-
-		## QC parameter, function (MDL SE LE)
-		_mdl = {
-			'Na+': 0.06,
-			'NH4+': 0.05,
-			'K+': 0.05,
-			'Mg2+': 0.12,
-			'Ca2+': 0.07,
-			'Cl-': 0.07,
-			'NO2-': 0.05,
-			'NO3-': 0.11,
-			'SO42-': 0.08,
-		}
-		_mdl.update(self._oth_set.get('mdl', {}))
-
-		def _se_le(_df_, _log=False):
-			_df_ = np.log10(_df_) if _log else _df_
-
-			_df_qua = _df_.quantile([.25, .75])
-			_df_q1, _df_q3 = _df_qua.loc[.25].copy(), _df_qua.loc[.75].copy()
-			_df_iqr = _df_q3 - _df_q1
-
-			_se = concat([_df_q1 - 1.5 * _df_iqr] * len(_df_), axis=1).T.set_index(_df_.index)
-			_le = concat([_df_q3 + 1.5 * _df_iqr] * len(_df_), axis=1).T.set_index(_df_.index)
-
-			if _log:
-				return 10 ** _se, 10 ** _le
-			return _se, _le
-
-		_cation, _anion, _main = ['Na+', 'NH4+', 'K+', 'Mg2+', 'Ca2+'], ['Cl-', 'NO2-', 'NO3-', 'SO42-', ], ['SO42-',
-																											 'NO3-',
-																											 'NH4+']
-
-		_df_salt = _df[_mdl.keys()].copy()
-		_df_pm = _df['PM2.5'].copy()
-
-		## lower than PM2.5
-		## conc. of main salt should be present at the same time (NH4+, SO42-, NO3-)
-		_df_salt = _df_salt.mask(_df_salt.sum(axis=1, min_count=1) > _df_pm).dropna(subset=_main).copy()
-
-		## mdl
-		for (_key, _df_col), _mdl_val in zip(_df_salt.items(), _mdl.values()):
-			_df_salt[_key] = _df_col.mask(_df_col < _mdl_val, _mdl_val / 2)
-
-		## calculate SE LE
-		## salt < LE
-		_se, _le = _se_le(_df_salt, _log=True)
-		_df_salt = _df_salt.mask(_df_salt > _le).copy()
-
-		## C/A, A/C
-		_rat_CA = (_df_salt[_cation].sum(axis=1) / _df_salt[_anion].sum(axis=1)).to_frame()
-		_rat_AC = (1 / _rat_CA).copy()
-
-		_se, _le = _se_le(_rat_CA, )
-		_cond_CA = (_rat_CA < _le) & (_rat_CA > 0)
-
-		_se, _le = _se_le(_rat_AC, )
-		_cond_AC = (_rat_AC < _le) & (_rat_AC > 0)
-
-		_df_salt = _df_salt.where((_cond_CA * _cond_AC)[0]).copy()
-
-		## conc. of main salt > SE
-		_se, _le = _se_le(_df_salt[_main], _log=True)
-		_df_salt[_main] = _df_salt[_main].mask(_df_salt[_main] < _se).copy()
-
-		return _df_salt.reindex(_df.index)

AeroViz/rawDataReader/script/OCEC_LCRES.py

@@ -1,34 +0,0 @@
-from pandas import to_datetime, read_csv
-
-from AeroViz.rawDataReader.core import AbstractReader
-
-
-class Reader(AbstractReader):
-	nam = 'OCEC_LCRES'
-
-	def _raw_reader(self, _file):
-		with open(_file, 'r', encoding='utf-8', errors='ignore') as f:
-			_df = read_csv(f, skiprows=3)
-
-			_col = {'Thermal/Optical OC (ugC/LCm^3)': 'Thermal_OC',
-					'Thermal/Optical EC (ugC/LCm^3)': 'Thermal_EC',
-					'OC=TC-BC (ugC/LCm^3)': 'Optical_OC',
-					'BC (ugC/LCm^3)': 'Optical_EC',
-					'Sample Volume Local Condition Actual m^3': 'Sample_Volume',
-					'TC (ugC/LCm^3)': 'TC', }
-
-			_tm_idx = to_datetime(_df['Start Date/Time'], errors='coerce')
-			_df['time'] = _tm_idx
-
-			_df = _df.dropna(subset='time').loc[~_tm_idx.duplicated()].set_index('time')
-
-		return _df[_col.keys()].rename(columns=_col)
-
-	## QC data
-	def _QC(self, _df):
-		_df[['Thermal_OC', 'Optical_OC']] = _df[['Thermal_OC', 'Optical_OC']].where(
-			_df[['Thermal_OC', 'Optical_OC']] > 0.3).copy()
-		_df[['Thermal_EC', 'Optical_EC']] = _df[['Thermal_EC', 'Optical_EC']].where(
-			_df[['Thermal_EC', 'Optical_EC']] > .015).copy()
-
-		return _df