AeroViz 0.1.2__py3-none-any.whl → 0.1.3b0__py3-none-any.whl

AeroViz/plot/bar.py

@@ -0,0 +1,126 @@
+from typing import Literal
+
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+from matplotlib.pyplot import Figure, Axes
+from pandas import DataFrame
+
+from AeroViz.plot.utils import *
+
+__all__ = ['bar']
+
+
+@set_figure(fw='bold')
+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)
+			else:
+				_ = 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=8, 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)
+			else:
+				_ = 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': 8})
+
+	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': 8})
+
+	# fig.savefig(f"Barplot_{title}")
+
+	plt.show()
+
+	return fig, ax

AeroViz/plot/box.py

@@ -0,0 +1,68 @@
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+from matplotlib.pyplot import Figure, Axes
+
+from AeroViz.plot.utils import *
+
+__all__ = ['box']
+
+
+@set_figure
+def box(df: pd.DataFrame,
+        x: str,
+        y: str,
+        x_bins: list | np.ndarray = None,
+        add_scatter: bool = True,
+        ax: Axes | None = None,
+        **kwargs) -> tuple[Figure, Axes]:
+    fig, ax = plt.subplots(**kwargs.get('fig_kws', {})) if ax is None else (ax.get_figure(), ax)
+
+    df = df.dropna(subset=[x, y]).copy()
+    x_data, y_data = df[x].to_numpy(), df[y].to_numpy()
+
+    bins = np.array(x_bins)
+    bins = np.round(bins)
+    wid = (bins + (bins[1] - bins[0]) / 2)[0:-1]
+
+    df[x + '_bin'] = pd.cut(x=x_data, bins=bins, labels=wid)
+
+    group = x + '_bin'
+    column = y
+    grouped = df.groupby(group, observed=False)
+
+    names, vals = [], []
+
+    for i, (name, subdf) in enumerate(grouped):
+        names.append('{:.0f}'.format(name))
+        vals.append(subdf[column].dropna().values)
+
+    plt.boxplot(vals, labels=names, positions=wid, widths=(bins[1] - bins[0]) / 3,
+                showfliers=False, showmeans=True, meanline=True, patch_artist=True,
+                boxprops=dict(facecolor='#f2c872', alpha=.7),
+                meanprops=dict(color='#000000', ls='none'),
+                medianprops=dict(ls='-', color='#000000'))
+
+    ax.set(xlim=kwargs.get('xlim', (x_data.min(), x_data.max())),
+           ylim=kwargs.get('ylim', (y_data.min(), y_data.max())),
+           xlabel=kwargs.get('xlabel', Unit(x)),
+           ylabel=kwargs.get('ylabel', Unit(y)),
+           title=kwargs.get('title', ''))
+
+    ax.set_xticks(bins, labels=bins.astype(int))
+
+    if add_scatter:
+        for i, (name, subdf) in enumerate(grouped):
+            jitter = np.random.normal(0, 0.5, len(subdf))
+            ax.scatter([name] * len(subdf) + jitter, subdf[column], s=10, c='gray', alpha=0.5)
+
+    plt.show()
+
+    return fig, ax
+
+
+if __name__ == '__main__':
+    from AeroViz import DataBase
+
+    df = DataBase(load_data=True)
+    box(df, x='PM25', y='Extinction', x_bins=np.arange(0, 120, 10))