如何使用时间滑块 plot 和 map 并使用 plotly 在 Z23EEEB4347BDD756BFC6B7EE9A 中放大城市

Question

I want to create an interactive map that shows the evolution of a variable (number of bikes) on different points of a city.

Why not by using plotly.

I would like to do something like this: https://amaral.northwestern.edu/blog/step-step-how-plot-map-slider-represent-time-evolu with a slider.

However, I don't achieve to reproduce it with focusing on a city, I can't choose a scope more precise than "europe".

Do you know how to do it with a zoom?

Answer 1

• you have not provided any code or sample data. Hence have used this http://api.citybik.es/v2/ https://github.com/eskerda/pybikes project to source some bike data
• data is just latest, so build up data in a pickle file for evolution
• simple case of using Plotly Express and animation_frame argument

data sources

import requests
import pandas as pd
from pathlib import Path
import plotly.express as px

# avalaible data sources...
pd.json_normalize(
    requests.get("http://api.citybik.es/v2/networks").json()["networks"]
).loc[lambda d: d["location.city"].eq("London")]

bike stand sourcing and plotting

df = pd.json_normalize(
    requests.get("http://api.citybik.es/v2/networks/santander-cycles").json()[
        "network"
    ]["stations"]
)

# build up some data over time
df["timestamp"] = pd.to_datetime(df["timestamp"]).round("15min")
f = Path.cwd().joinpath("bikes.pickle")
if not f.exists():
    df.to_pickle(f)
else:
    df = pd.concat([pd.read_pickle(f), df])
    df = df.groupby(["timestamp","id"], as_index=False).first()
    df.to_pickle(f)

# now just plot it on a map with evolution by time
df["ts_str"] = df["timestamp"].dt.strftime("%d-%b %H:%M")
px.scatter_mapbox(
    df,
    lat="latitude",
    lon="longitude",
    size="free_bikes",
    hover_data=["name"],
    animation_frame="ts_str",
).update_layout(
    mapbox={"style": "carto-positron", "zoom":11}, margin={"l": 0, "r": 0, "t": 0, "b": 0}
)