布局#

cuxfilter 提供了预设和自定义布局选项。请参见下面的示例了解如何使用这两种选项。

下载数据集#

[1]:

from cuxfilter.sampledata import datasets_check

[ ]:

DATA_DIR = './data'
! curl https://data.rapids.ai/viz-data/146M_predictions_v2.arrow.gz --create-dirs -o $DATA_DIR/146M_predictions_v2.arrow.gz
datasets_check('mortgage', base_dir=DATA_DIR)

导入并设置图表#

[ ]:

from cuxfilter import charts
import cuxfilter
from bokeh import palettes
import panel as pn

cux_df = cuxfilter.DataFrame.from_arrow('./data/146M_predictions_v2.arrow')

chart0 = charts.choropleth(x='zip', color_column='delinquency_12_prediction', color_aggregate_fn='mean',
                                  geo_color_palette=palettes.Purples9,
                                  geoJSONSource = 'https://raw.githubusercontent.com/rapidsai/cuxfilter/GTC-2018-mortgage-visualization/javascript/demos/GTC%20demo/public/data/zip3-ms-rhs-lessprops.json',
                                  nan_color='white')
chart1 = charts.bar('dti')
chart2 = charts.bar('delinquency_12_prediction',data_points=50)
chart3 = charts.bar('borrower_credit_score', step_size=1)
chart4 = charts.bar('seller_name')
chart5 = charts.scatter(x='loan_id',y='current_actual_upb')
chart6 = charts.scatter('zip', 'dti')
chart7 = charts.heatmap('dti','borrower_credit_score', aggregate_col='delinquency_12_prediction', aggregate_fn="mean")
chart8 = charts.line('loan_id','borrower_credit_score')

#create a list of charts
charts_list = [chart0, chart3, chart1, chart2, chart4, chart5, chart6, chart7, chart8]

widgets = [charts.multi_select('dti'), charts.card(pn.pane.Markdown("""
## Sample Palette Legend

- ![#A932FF](https://via.placeholder.com/15/A932FF/000000?text=+) `#A932FF`: Purple 1
- ![#8E44AD](https://via.placeholder.com/15/8E44AD/000000?text=+) `#8E44AD`: Purple 2
- ![#6C3483](https://via.placeholder.com/15/6C3483/000000?text=+) `#6C3483`: Purple 3
- ![#512E5F](https://via.placeholder.com/15/512E5F/000000?text=+) `#512E5F`: Purple 4
- ![#341C4E](https://via.placeholder.com/15/341C4E/000000?text=+) `#341C4E`: Purple 5
""")) ]

用户自定义布局#

Layout_array#

自定义布局通过传递给 .dashboard() API 的输入参数 layout_array 来应用。

布局数组是一个列表的列表，表示一个二维布局页面。每个列表对应布局中的一整行。一个列表包含图表编号（从 1 到 n），表示它们在页面上的确切位置。输入的数组会自动缩放以适应整个屏幕。

示例 1：#

layout_array = [[1]]

将导致单个图表占据整个页面。

3d7f3b659d0e44e3b0fa18aac4321b53

示例 2：#

layout_array = [[1], [1], [2]]

将导致图表 1 占据前两行，图表 2 占据最后一行，大致将两图表布局按 66%-33% 的比例划分。

f588dc5e26c54724849cfe88433d6abe

示例 3：#

[ ]:

d = cux_df.dashboard(charts_list, sidebar=widgets, layout_array=[
    [1, 1, 2, 2],
    [1, 1, 3, 4]
], theme=cuxfilter.themes.rapids_dark, title="Layout - Custom")

showcase-custom

预设布局#

预设布局通过传递给 .dashboard() API 的输入参数 layout 来应用。

单特征

f839cc9688d64050bc968218607a1819

[ ]:

d = cux_df.dashboard(charts_list, sidebar=widgets, layout=cuxfilter.layouts.single_feature, theme=cuxfilter.themes.rapids_dark, title="Layout - single feature")

showcase-single-feature

特征和基础

7a1f5e09320a4160b2757fd95f7f1a8f

[ ]:

d = cux_df.dashboard(charts_list, sidebar=widgets, layout=cuxfilter.layouts.feature_and_base, theme=cuxfilter.themes.rapids_dark, title="Layout - feature and base")

showcase-feature-and-base

双特征

611548a938324d32a07dfc60d21513b8

[ ]:

d = cux_df.dashboard([chart0, chart1], sidebar=widgets, layout=cuxfilter.layouts.double_feature, theme=cuxfilter.themes.rapids_dark, title="Layout - double feature")

showcase-double-feature

左特征右双列

def3779cc7644b93bf3ad17a6f11881d

[ ]:

d = cux_df.dashboard(charts_list[:4], sidebar=widgets, layout=cuxfilter.layouts.left_feature_right_double, theme=cuxfilter.themes.rapids_dark, title="Layout - left feature right double")

showcase-left-feature-right-double

三特征

0d7f7a5ad05a406eba103d02bfc42581

[ ]:

d = cux_df.dashboard([chart1, chart2, chart3], sidebar=widgets, layout=cuxfilter.layouts.triple_feature, theme=cuxfilter.themes.rapids_dark, title="Layout - triple feature")

showcase-triple-feature

特征和双基础

96eb21ddaf1d4d13aefd1fb6379b4ee1

[ ]:

d = cux_df.dashboard([chart0, chart2, chart3], sidebar=widgets, layout=cuxfilter.layouts.feature_and_double_base, theme=cuxfilter.themes.rapids_dark, title="Layout - feature and double base")

showcase-feature-and-double-base

两行两列

34ebfba4c90f4e3ca1b80f50e62a4f29

[ ]:

d = cux_df.dashboard([chart0, chart2, chart3, chart4], sidebar=widgets, layout=cuxfilter.layouts.two_by_two, theme=cuxfilter.themes.rapids_dark, title="Layout - two by two")

showcase-two-by-two

特征和三基础

581b46e0735f4ca2a45d15e1e94d547b

[ ]:

d = cux_df.dashboard(charts_list, sidebar=widgets, layout=cuxfilter.layouts.feature_and_triple_base, theme=cuxfilter.themes.rapids_dark, title="Layout - feature and triple base")

showcase-feature-and-triple-base

特征和四基础

a6d9da6c0ff841f1b7e6c56d83377d67

[ ]:

d = cux_df.dashboard(charts_list, sidebar=widgets, layout=cuxfilter.layouts.feature_and_quad_base, theme=cuxfilter.themes.rapids_dark, title="Layout - feature and quad base")

showcase-feature-and-quad-base

特征和五边缘

3752862e1344499686fd83901bffb26a

[ ]:

d = cux_df.dashboard(charts_list, sidebar=widgets, layout=cuxfilter.layouts.feature_and_five_edge, theme=cuxfilter.themes.rapids_dark, title="Layout - feature and five edge")

showcase-feature-and-five-base

两行三列

f8f41e269e704ef5aa7e865693280f0f

[ ]:

d = cux_df.dashboard([chart3, chart1, chart2, chart4, chart5, chart6], sidebar=widgets, layout=cuxfilter.layouts.two_by_three, theme=cuxfilter.themes.rapids_dark, title="Layout - two by three")

showcase-two-by-three

双特征四基础

e265db790e614c87aa9e7f7f901bdaad

[ ]:

d = cux_df.dashboard(charts_list, sidebar=widgets, layout=cuxfilter.layouts.double_feature_quad_base,
                     theme=cuxfilter.themes.rapids_dark, title="Layout - double feature quad base")

showcase-double-feature-quad-base

三行三列

ad3715fac6514bfa8c62d446562de175

[ ]:

d = cux_df.dashboard(charts_list, sidebar=widgets, layout=cuxfilter.layouts.three_by_three,
                     theme=cuxfilter.themes.rapids_dark, title="Layout - three by three")

showcase-three-by-three

[ ]:

布局#