Animate a Plotly map with a sliding date bar

user_2205969 0
Viewed 3318

I'm struggling to turn this piece of code I wrote - which constructs a static heatmap - into an animated version with a date slider. 

import pandas as pd
import plotly.graph_objects as go

...

fig = go.Figure(go.Densitymapbox(lat=df_heat['lat'], lon=df_heat['lon'], z=df_heat['count'],
                                 radius=10,))
fig.update_layout(mapbox_style="carto-positron", mapbox_zoom=10, mapbox_center = {"lat": 40.7831, "lon": -73.9712},)

fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
fig.show()

The above code successfully converts a Pandas DataFrame df_heat, which looks like the following, into a Plotly heatmap.

    lat         lon         count
0   -62.884215  39.440236   1
1   -62.834226  39.408072   1
2   -62.811707  39.380462   1
3   -62.744564  39.489112   1
...

Static heatmap output:

static heatmap[1]

df_heat is itself just an aggregated view of the following DataFrame, which also includes a date. 

    date        lat         lon         count
0   2018-07-29  40.691828   -73.944609  1
1   2018-07-29  40.693601   -73.945092  1
2   2018-07-29  40.696132   -73.945178  1
3   2018-07-29  40.692726   -73.945532  1

My question is, how can I convert this DataFrame with dates into an animated plotly map, such as the ones here, here, and here, which feature a date slider as a filter.

Dummy data for testing:

df = pd.DataFrame({'datetime': {0: '2018-09-29 00:00:00', 1: '2018-07-28 00:00:00', 2: '2018-07-29 00:00:00', 3: '2018-07-29 00:00:00', 4: '2018-08-01 00:00:00', 5: '2018-08-01 00:00:00', 6: '2018-08-01 00:00:00', 7: '2018-08-05 00:00:00', 8: '2018-09-06 00:00:00', 9: '2018-09-07 00:00:00', 10: '2018-09-07 00:00:00', 11: '2018-09-08 00:00:00', 12: '2018-09-08 00:00:00', 13: '2018-09-08 00:00:00', 14: '2018-10-08 00:00:00', 15: '2018-10-10 00:00:00', 16: '2018-10-10 00:00:00', 17: '2018-10-11 00:00:00', 18: '2018-10-11 00:00:00', 19: '2018-10-11 00:00:00'},
 'lat': {0: 40.6908284, 1: 40.693601, 2: 40.6951317, 3: 40.6967261, 4: 40.697593, 5: 40.6987141, 6: 40.7186497, 7: 40.7187772, 8: 40.7196151, 9: 40.7196865, 10: 40.7187408, 11: 40.7189716, 12: 40.7214273, 13: 40.7226571, 14: 40.7236955, 15: 40.7247207, 16: 40.7221074, 17: 40.7445859, 18: 40.7476252, 19: 40.7476451},
 'lon': {0: -73.9336094, 1: -73.9350917, 2: -73.9351778, 3: -73.9355315, 4: -73.9366737, 5: -73.9393797, 6: -74.0011939, 7: -74.0010918, 8: -73.9887851, 9: -74.0035125, 10: -74.0250842, 11: -74.0299202, 12: -74.029886, 13: -74.027542, 14: -74.0290157, 15: -74.0291541, 16: -74.0220728, 17: -73.9442636, 18: -73.9641326, 19: -73.9533039},
 'count': {0: 1, 1: 2, 2: 5, 3: 1, 4: 6, 5: 1, 6: 3, 7: 2, 8: 1, 9: 7, 10: 3, 11: 3, 12: 1, 13: 2, 14: 1, 15: 1, 16: 2, 17: 1, 18: 1, 19: 1}})
pandas plotly animation python-3.x python
2 Answers

I did few changes and added the timeline animation to your code. Similarly to the solution of Teoretic I also used plotly.express which make things shorter.

Live example http://www.erangrinberg.de/plotly/map-animation.html

import pandas as pd
import plotly.express as px

df = pd.DataFrame({'Date': {0: '2018-09-29 00:00:00', 1: '2018-07-28 00:00:00', 2: '2018-07-29 00:00:00', 3: '2018-07-29 00:00:00', 4: '2018-08-01 00:00:00', 5: '2018-08-01 00:00:00', 6: '2018-08-01 00:00:00', 7: '2018-08-05 00:00:00', 8: '2018-09-06 00:00:00', 9: '2018-09-07 00:00:00', 10: '2018-09-07 00:00:00', 11: '2018-09-08 00:00:00', 12: '2018-09-08 00:00:00', 13: '2018-09-08 00:00:00', 14: '2018-10-08 00:00:00', 15: '2018-10-10 00:00:00', 16: '2018-10-10 00:00:00', 17: '2018-10-11 00:00:00', 18: '2018-10-11 00:00:00', 19: '2018-10-11 00:00:00'},
                  'lat': {0: 40.6908284, 1: 40.693601, 2: 40.6951317, 3: 40.6967261, 4: 40.697593, 5: 40.6987141, 6: 40.7186497, 7: 40.7187772, 8: 40.7196151, 9: 40.7196865, 10: 40.7187408, 11: 40.7189716, 12: 40.7214273, 13: 40.7226571, 14: 40.7236955, 15: 40.7247207, 16: 40.7221074, 17: 40.7445859, 18: 40.7476252, 19: 40.7476451},
                  'lon': {0: -73.9336094, 1: -73.9350917, 2: -73.9351778, 3: -73.9355315, 4: -73.9366737, 5: -73.9393797, 6: -74.0011939, 7: -74.0010918, 8: -73.9887851, 9: -74.0035125, 10: -74.0250842, 11: -74.0299202, 12: -74.029886, 13: -74.027542, 14: -74.0290157, 15: -74.0291541, 16: -74.0220728, 17: -73.9442636, 18: -73.9641326, 19: -73.9533039},
                  'count': {0: 1, 1: 2, 2: 5, 3: 1, 4: 6, 5: 1, 6: 3, 7: 2, 8: 1, 9: 7, 10: 3, 11: 3, 12: 1, 13: 2, 14: 1, 15: 1, 16: 2, 17: 1, 18: 1, 19: 1}})

fig = px.density_mapbox(df, lat=df['lat'], 
                            lon=df['lon'], 
                            z=df['count'],
                            radius=10,
                            animation_frame="Date"
                                )
fig.update_layout(mapbox_style="carto-positron", mapbox_zoom=10, mapbox_center = {"lat": 40.7831, "lon": -73.9712},)
fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})

fig.layout.updatemenus[0].buttons[0].args[1]["frame"]["duration"] = 600
fig.layout.updatemenus[0].buttons[0].args[1]["transition"]["duration"] = 600
fig.layout.coloraxis.showscale = True   
fig.layout.sliders[0].pad.t = 10
fig.layout.updatemenus[0].pad.t= 10             

fig.show()

Would be nice to see the final result, or maybe you can share the DataSet Source.

user_136121050

You can play with scatter_geo plot from plotly.express to get an interactive graph.
It doesn't produce the heat map, but it can make dots like on your graph.

Sample code with your dummy data:

import pandas as pd
import plotly.express as px

df = pd.DataFrame({'datetime': {0: '2018-09-29 00:00:00', 1: '2018-07-28 00:00:00', 2: '2018-07-29 00:00:00', 3: '2018-07-29 00:00:00', 4: '2018-08-01 00:00:00', 5: '2018-08-01 00:00:00', 6: '2018-08-01 00:00:00', 7: '2018-08-05 00:00:00', 8: '2018-09-06 00:00:00', 9: '2018-09-07 00:00:00', 10: '2018-09-07 00:00:00', 11: '2018-09-08 00:00:00', 12: '2018-09-08 00:00:00', 13: '2018-09-08 00:00:00', 14: '2018-10-08 00:00:00', 15: '2018-10-10 00:00:00', 16: '2018-10-10 00:00:00', 17: '2018-10-11 00:00:00', 18: '2018-10-11 00:00:00', 19: '2018-10-11 00:00:00'},
 'lat': {0: 40.6908284, 1: 40.693601, 2: 40.6951317, 3: 40.6967261, 4: 40.697593, 5: 40.6987141, 6: 40.7186497, 7: 40.7187772, 8: 40.7196151, 9: 40.7196865, 10: 40.7187408, 11: 40.7189716, 12: 40.7214273, 13: 40.7226571, 14: 40.7236955, 15: 40.7247207, 16: 40.7221074, 17: 40.7445859, 18: 40.7476252, 19: 40.7476451},
 'lon': {0: -73.9336094, 1: -73.9350917, 2: -73.9351778, 3: -73.9355315, 4: -73.9366737, 5: -73.9393797, 6: -74.0011939, 7: -74.0010918, 8: -73.9887851, 9: -74.0035125, 10: -74.0250842, 11: -74.0299202, 12: -74.029886, 13: -74.027542, 14: -74.0290157, 15: -74.0291541, 16: -74.0220728, 17: -73.9442636, 18: -73.9641326, 19: -73.9533039},
 'count': {0: 1, 1: 2, 2: 5, 3: 1, 4: 6, 5: 1, 6: 3, 7: 2, 8: 1, 9: 7, 10: 3, 11: 3, 12: 1, 13: 2, 14: 1, 15: 1, 16: 2, 17: 1, 18: 1, 19: 1}})
fig = px.scatter_geo(df, 
                     lat='lat', 
                     lon='lon', 
                     scope='usa',
                     color="count", 
                     size='count',
                     projection="albers usa", 
                     animation_frame="datetime", 
                     title='Your title')
fig.update(layout_coloraxis_showscale=False)
fig.show()

enter image description here

Also you can check this kaggle notebook for more examples of usage of this graph.

user_52999690
Related