Floatview Stats

Floatview

A floatview output widget for JupyterLab and a data explorer for glue/iplotly

Installation

If you use jupyterlab 2.X:

pip install floatview
jupyter labextension install @jupyter-widgets/[email protected]
jupyter labextension install [email protected]
jupyter labextension install [email protected]
jupyter labextension install [email protected]

If you use jupyterlab 1.X:

pip install floatview==0.2.0
jupyter labextension install @jupyter-widgets/[email protected]
jupyter labextension install [email protected]
jupyter labextension install [email protected]
jupyter labextension install [email protected]
jupyter labextension install [email protected]

older versions:

pip install floatview==0.1.18
jupyter labextension install @jupyterlab/[email protected]
jupyter labextension install [email protected]
jupyter labextension install @jupyter-widgets/[email protected]
jupyter labextension install [email protected]

Usage

The floatview widget is used as a context manager, just like ipywidgets' output widget.

from floatview import Floatview
from ipywidgets import IntSlider

sc = Floatview(title='Floatview Output', mode='tab-after', active=True)
sl = IntSlider(description='Some slider')
with sc:
    display(sl)

When a single output is displayed in a Floatview, it is allowed to occupy all of the vertical space available. If more content is displayed, the natural height is used instead.

The gluemanagerwidget is used as a data/visualization manager for a glue dataset.

from floatview import GlueManagerWidget
from pandas import read_csv

data = read_csv('your_data.csv', index_col=False, usecols=cols)
gmw = GlueManagerWidget(data, modal=True, label="Data")

Available Visualizations

#Histogram (supports 1 component)
view = gmw.gluemanager.newView(
    "histogram",
    ["PULocationID"],
    "Histogram"
)

#Scatter (supports 2-n components)
view = gmw.gluemanager.newView(
    "scatter",
    ["PULocationID", "DOLocationID"],
    "Scatter"
)
view = gmw.gluemanager.newView(
    "composed_scatter",
    ["trip_distance", "payment_type", 'passenger_count'],
    "Composed Scatter"
)

#ErrorBar (supports 2-n components)
view = gmw.gluemanager.newView(
    "errorbar",
    ["trip_distance", "total_amount"],
    "Error"
)
view = gmw.gluemanager.newView(
    "composed_errorbar",
    ["trip_distance", "payment_type", 'passenger_count'],
    "Composed Error"
)

#Polynomial Fitting 2-n degree (supports n components)
view = gmw.gluemanager.newView(
    "composed_polyfit_3d",
    ["trip_distance", "total_amount"],
    "Polyfit"
);

#scatter 3D (supports 3 components)
view = gmw.gluemanager.newView(
    "scatter3D",
    ["trip_distance", "total_amount", "passenger_count"],
    "Scatter3D"
)

#Contours 2D (supports 2 components)
view = gmw.gluemanager.newView(
    "contour",
    ["trip_distance", "total_amount"],
    "Contour"
);

#Table (supports n components)
view = gmw.gluemanager.newView(
    "table",
    ['passenger_count', 'trip_distance', 'total_amount', 'payment_type'],
    "Table"
);

#Parallel coordinatess (supports n components)
view = gmw.gluemanager.newView(
    "parallels",
    ['passenger_count', 'trip_distance', 'total_amount', 'payment_type'],
    "Parallels"
);

#Parallel categories (supports n components)
view = gmw.gluemanager.newView(
    "parallelscat",
    ['passenger_count', 'trip_distance', 'total_amount', 'payment_type'],
    "Parallels Categ"
 );

#Sankey (supports n components)
view = gmw.gluemanager.newView(
    "sankey",
    ['passenger_count', 'trip_distance', 'total_amount', 'payment_type'],
    "Sankey"
);

#Sunburst (supports n components)
view = gmw.gluemanager.newView(
    "sunburst",
    ['passenger_count', 'trip_distance', 'total_amount', 'payment_type'],
    "Sunburst"
);

#Sankey Tree (supports n components)
view = gmw.gluemanager.newView(
    "sankeytree",
    ['total_amount', 'payment_type', 'passenger_count', ],
    "Sankey Tree"
);

#Scatter Matrix (supports n components)
view = gmw.gluemanager.newView(
    "scattermatrix",
    ['passenger_count', 'trip_distance', 'total_amount', 'payment_type'],
    "scatter Matrix"
);

#Correlation Matrix (supports n components)
view = gmw.gluemanager.newView(
    "corrcoef",
    ['passenger_count', 'trip_distance', 'total_amount', 'payment_type'],
    "Correlation Matrix"
);

#Principal components (supports n components)
view = gmw.gluemanager.newView(
    "pca",
    ['passenger_count', 'trip_distance', 'total_amount', 'payment_type'],
    "Principal components"
);

#Network (supports 2 components)
view = gmw.gluemanager.newView(
    "network",
    ['trip_distance', 'total_amount'],
    "Network"
);

#Image (supports 3 components)
view = gmw.gluemanager.newView(
    "image",
    ["trip_distance", "total_amount", 'passenger_count'],
    "Image"
);

#Lines (supports n components)
view = gmw.gluemanager.newView(
    "composed_lines",
    ["trip_distance", "payment_type", 'passenger_count'],
    "Lines"
);