Simple Dashboarding using pandas, dash, and plotly express.
Pass a function that returns a dataframe/series, and optional keyoword arguments, to a chart. Put these tpgether in a list of lists (row x column order) to create the dashboard layout.
Two main chart interfaces:
Can use a simple chart, which takes a function that returns regular pandas dataframe/series:
BarChart(title='Data', pd_obj=some_func, pd_obj_kwargs=some_func_kwargs)Or can use a px chart which takes a function that returns a 'tidy'/normalized dataframe and has the same interface as plotly epress charts (https://www.plotly.express/plotly_express/). Pass title, chart type, func, func_kwargs, then any kwargs the plotly express function takes:
PXChart(title='Global Population', chart_type='area', pd_obj=px.data.gapminder, x="year", y="pop", color="continent", line_group="country")Can use make_px_tidy_df to turn an 'untidy' (un-normalized) dataframe into a tidy one. Can also used make_return_tidy_df as a decorator to make a function return a tidy dataframe.
Optional dark theme.
from pandex import *
import pandas as pd
import numpy as np
import plotly_express as px
import dash_core_components as dcc
def rand_df(rows=5, cols=3, cumsum=True):
if cumsum == 'False':
return pd.DataFrame(np.random.randn(rows, cols))
return pd.DataFrame(np.random.randn(rows, cols)).cumsum()
dbaord = SimpleDashboard(
title='Example Pandex Dashboard',
charts=[
[
BarChart('Bar', rand_df), LineChart('Line', rand_df, {'rows': 200})
],
[
ScatterChart('Scatter', rand_df, {'rows': 20, 'cumsum': False})
],
]
)
if __name__ == '__main__':
dbaord.run(debug=True, port=8055)
from pandex import *
import pandas as pd
import numpy as np
import plotly_express as px
def rand_df(rows=5, cols=3, cumsum=True):
if cumsum == 'False':
return pd.DataFrame(np.random.randn(rows, cols))
return pd.DataFrame(np.random.randn(rows, cols)).cumsum()
# decorator from pandex to make rand_df return a tidy/normalized dataframe
rand_df_tidy = make_return_px_tidy_df(rand_df)
dbaord = SimpleDashboard(
title='Example Pandex Dashboard',
dark_theme=True,
charts=[
[
section('SimpleChart Interface')
],
[
LineChart('Line', rand_df, {'rows': 200}),
BarChart('Bar', rand_df),
ScatterChart('Scatter', rand_df, {'rows': 20, 'cumsum': False})
],
[
BarChart('4 Columns', rand_df, {'cols':1}, html_col_count=4),
AreaChart('2 Columns', rand_df, {'rows': 200}, html_col_count=8)
],
[
section('PXChart Interface')
],
[
PXChart(title='Global Population', chart_type='area', pd_obj=px.data.gapminder, x="year", y="pop", color="continent", line_group="country"),
PXChart(title='Iris Heatmap', chart_type='density_heatmap', pd_obj=px.data.iris, x="sepal_width", y="sepal_length")
],
[
PXChart(title='Scatter', chart_type='scatter', pd_obj=px.data.iris, x="sepal_width", y="sepal_length"),
PXChart(title='Scatter With Categories', chart_type='scatter', pd_obj=px.data.iris, x="sepal_width", y="sepal_length", color="species"),
PXChart(title='Scatter Marginal', chart_type='scatter', pd_obj=px.data.iris, x="sepal_width", y="sepal_length", color="species", marginal_y="rug", marginal_x="histogram"),
],
[
PXChart(title='Horizontal Legend', chart_type='line', pd_obj=rand_df_tidy, pd_obj_kwargs=dict(rows=300, cumsum=True), x='index', y='value', color='variable', layout=dict(legend_orientation='h')),
PXTable(title='Table', chart_type=None, pd_obj=px.data.iris),
],
]
)
if __name__ == '__main__':
dbaord.run(debug=True, port=8055)
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent