Question

我已经建立了一个Jupyter笔记本来分析和对比历史数据。因为数据库很大，所以我使用了search + observe小部件来允许用户选择他们想要的实验，而不是滚动浏览所有数据。现在，所有实验的所有复选框均已生成，但这既太大又无法放入框中，并且会占用大量时间。

是否可以默认显示仅10个实验，并且仅在搜索后生成所需的复选框？

i基于此gist编写代码。我已经尝试过将复选框小部件移到else循环内，但这没有用。另外，全局/局部变量情况在我的代码中似乎很混乱。

import ipywidgets as widgets

#Generate a dummy list
Allfileslist = ["Experiment {}".format(x) for x in range(600)]
#Search box + generate ALL the checkboxes, this seems to be the bottleneck
search_widget = widgets.Text(placeholder='Type for older experiments',                 description= 'Search:')
experiments = {eachfilename: widgets.Checkbox(description=eachfilename, value=False) for eachfilename in Allfileslist}
options = [experiments[eachfilename] for eachfilename in Allfileslist]
options_widget = widgets.VBox(options, layout={'overflow': 'auto'})


def whentextischanged (change):
    """Dynamically update the widget experiments"""
    search_input = change['new']
    if search_input == '':
        # Reset search field, default to last 9 experiments
        new_options = [experiments[eachfilename] for eachfilename in Allfileslist[-10:]]
    else:
        # Filter by search
        close_matches = [x for x in Allfileslist if search_input.upper() in x]
        ### magic code that generates only the match boxes ###
        new_options = [experiments[eachfilename] for eachfilename in close_matches]

    options_widget.children = new_options

#Generate the vbox, search
multi_select = widgets.VBox([search_widget, options_widget])
search_widget.observe(whentextischanged, names='value')
#serve it up
multi_select

有什么方法可以只生成加快速度所需的复选框吗？

更新我将ac24的答案进行了一些修改，并且几乎完成了所需的工作。唯一没有做的就是当搜索框开头为空时显示默认值。然后，清除搜索框将显示正确的行为。

import ipywidgets as widgets

#Generate a dummy list
Allfileslist = ["{}".format(x) for x in range(600)]
#Search box + generate some checboxes
search_widget = widgets.Text(placeholder='Type for older experiments', description= 'Search:',value='')
experiments = {}
options_widget = widgets.VBox(layout={'overflow': 'auto'})
default_options = [widgets.Checkbox(description=eachfilename, value=False) for eachfilename in Allfileslist[-10:]]

def whentextischanged (change):
    """Dynamically update the widget experiments"""
    search_input = change['new']
    if search_input == '':
        # Reset search field, default to last 9 experiments
        new_options = default_options
    else:
        # Filter by search
        close_matches = [x for x in Allfileslist if search_input.lower() in x.lower()][:10]
        for name in close_matches:
            if name not in experiments:
                experiments[name] = widgets.Checkbox(description=name, value=False)
        new_options = [experiments[eachfilename] for eachfilename in close_matches]

    options_widget.children = new_options

#Generate the vbox, search
multi_select = widgets.VBox([search_widget, options_widget])
search_widget.observe(whentextischanged, names='value')
multi_select

Answer 1

尝试先创建一个空字典，然后在更接近所需实验选择时填充复选框。

import ipywidgets as widgets

#Generate a dummy list
Allfileslist = ["{}".format(x) for x in range(600)]
#Search box + generate ALL the checkboxes, this seems to be the bottleneck
search_widget = widgets.Text(placeholder='Type for older experiments',                 description= 'Search:')
# experiments = {eachfilename: widgets.Checkbox(description=eachfilename, value=False) for eachfilename in Allfileslist}
experiments = {}
options_widget = widgets.VBox(layout={'overflow': 'auto'})


def whentextischanged (change):
    """Dynamically update the widget experiments"""
    search_input = change['new']
    if search_input == '':
        # Reset search field, default to last 9 experiments
        new_options = [experiments[eachfilename] for eachfilename in Allfileslist[-10:]]
    else:
        # Filter by search
        close_matches = [x for x in Allfileslist if search_input.lower() in x.lower()][:10]
        for name in close_matches:
            if name not in experiments:
                experiments[name] = widgets.Checkbox(description=name, value=False)

        ### magic code that generates only the match boxes ###
        new_options = [experiments[eachfilename] for eachfilename in close_matches]

    options_widget.children = new_options

#Generate the vbox, search
multi_select = widgets.VBox([search_widget, options_widget])
search_widget.observe(whentextischanged, names='value')
#serve it up
multi_select

然后，选择了几个选项后，调用[key for key, exp_check in exp_checkboxes.items() if exp_check.value]以查看所选的值。

使用jupyter窗口小部件动态生成与搜索相结合的复选框

1 个答案: