我正在尝试使用python在jupyter笔记本中构建文件/数据选择器。我的想法是,我使用multipleSelect小部件选择文件中的一些文件和数据通道,然后通过按钮返回一个dataFrame。
如何访问df_object?
#stack example
from ipywidgets import widgets
from IPython.display import display
from IPython.display import clear_output
import pandas as pd
import numpy as np
filenames = ["file1", "file2"]
file_dict = {
"file1":pd.DataFrame(np.arange(5)),
"file2":pd.DataFrame(np.arange(10,15))
}
def data_selection():
sel_file = widgets.SelectMultiple(description="Files",
options=filenames)
display(sel_file)
button = widgets.Button(description="OK")
display(button)
def on_button_clicked(button):
clear_output(wait=True) #clears the previous output
display(sel_file) #displays new selection window
display(button) #displays new button
for f in sel_file.value:
print (f)
display (file_dict[f])
#global df_object #would be a solution but not recommended for sure
df_object = file_dict[f]
return df_object #doesn't work
button.on_click(on_button_clicked)
data_selection()
答案 0 :(得分:0)
您确实应该为此使用一个类,然后将所有函数定义为对该类的实例起作用。并非所有人都需要公开访问。您还可以将df_objects
存储在单独的属性(如字典)中,并使用单独的函数访问字典。看看下面的代码:
class foo(object):
def __init__(self, file1, file2):
self.filenames = [file1, file2]
self.file_dict = {
file1:pd.DataFrame(np.arange(5)),
file2:pd.DataFrame(np.arange(10,15))
}
def _create_widgets(self):
self.sel_file = widgets.SelectMultiple(description='Files',
options=self.filenames,
value=[self.filenames[0]],
)
self.button = widgets.Button(description="OK")
self.button.on_click(self._on_button_clicked)
def _on_button_clicked(self, change):
self.out.clear_output()
self.df_objects = {}
with self.out:
for f in self.sel_file.value:
print(f)
display(self.file_dict[f])
self.df_objects[f] = self.file_dict[f]
def display_widgets(self):
self._create_widgets()
self.out = widgets.Output() # this is the output widget in which the df is displayed
display(widgets.VBox(
[
self.sel_file,
self.button,
self.out
]
)
)
def get_df_objects(self):
return self.df_objects
然后,您可以创建实例并显示小部件,如下所示:
something = foo('a', 'b')
something.display_widgets()
something.get_df_objects()
将返回包含必需的“ file:dataframe_of_file
”键值对的字典。
希望这会有所帮助:)