"""
Stacked Bar Chart
-----------------
This example shows how to make a stacked bar chart of the weather type in Seattle from 2012 to 2015 by month.
"""
import altair as alt
from vega_datasets import data
weather = data.seattle_weather()
chart = alt.Chart(weather).mark_bar().encode(
alt.Color(
'weather:N',
legend=alt.Legend(title='Weather type'),
scale=alt.Scale(
domain=['sun', 'fog', 'drizzle', 'rain', 'snow'],
range=['#e7ba42', '#c7c7c7', '#aec7e8', '#1f77b4', '#9467bd']),
),
alt.X(
'date:N',
axis=alt.Axis(title='Month of the Year'),
timeUnit='month',
),
y='count(*):Q',
)
"""
Layered Plot with Dual-Axis
---------------------------
This example shows how to combine two plots and keep their axes.
"""
import altair as alt
from vega_datasets import data
source = data.seattle_weather()
base = alt.Chart(source).encode(
alt.X('date:O',
axis=alt.Axis(format='%b'),
timeUnit='month',
scale=alt.Scale(zero=False)))
bar = base.mark_bar().encode(y='mean(precipitation)')
line = base.mark_line(color='red').encode(y='mean(temp_max)', )
chart = (bar + line).resolve_scale(y='independent')
# create new plots and share both ranges linked_plot2 = figure(width=250, height=250, x_range=linked_plot1.x_range, y_range=linked_plot1.y_range) linked_plot2.line(linked_data_x, linked_data_y) linked_plot3 = figure(width=250, height=250, x_range=linked_plot1.x_range, y_range=linked_plot1.y_range) linked_plot3.vbar(linked_data_x, top=linked_data_y, width=0.5) # the subplots in a gridplot linked_gridplot = gridplot([[linked_plot1, linked_plot2, linked_plot3]]) # show the results show(linked_gridplot) Linked Selection - Box Select, Lasso Select In [76]: # data seattle_weather = vds.seattle_weather() seattle_weather.tail() Out[76]: date precipitation temp_max temp_min wind weather 1456 2015-12-27 8.6 4.4 1.7 2.9 fog 1457 2015-12-28 1.5 5.0 1.7 1.3 fog 1458 2015-12-29 0.0 7.2 0.6 2.6 fog 1459 2015-12-30 0.0 5.6 -1.0 3.4 sun 1460 2015-12-31 0.0 5.6 -2.1 3.5 sun In [77]: from bokeh.transform import factor_cmap, factor_mark TOOLS = 'box_select, lasso_select, reset, wheel_zoom, pan' weather_types = ['drizzle', 'rain', 'sun', 'snow', 'fog'] weather_markers = ['hex', 'cross', 'triangle', 'square', 'circle_x']
sf.resample('D').min().plot()
sf.resample('M').mean().idxmin().temp.month
sf.resample('M').mean().idxmax().temp.month
extremes = sf.resample('D').agg(['min', 'max'])
extremes['diff'] = extremes.temp['max'] - extremes.temp['min']
extremes.resample('M').diff.mean().idxmax().month
plt.plot(sf.resample('D').mean())
plt.plot(sf.resample('D').min())
plt.plot(sf.resample('D').max())
plt.show()
seattle = data.seattle_weather().set_index('date').sort_index()
seattle.resample('M').precipitation.sum().idxmax()
seattle.resample('M').precipitation.sum().plot()
seattle.resample('W').wind.mean().plot()
seattle.resample('M').wind.mean().idxmax()
seattle['sunny'] = seattle['weather'] == 'sun'
seattle.resample('Y').sunny.sum().idxmax()
seattle.groupby(seattle.index.month).precipitation.sum().idxmax()
seattle['rained'] = seattle.precipitation > 0
seattle.groupby(seattle.index.month).rained.sum().idxmax()
"""
Seattle Weather Interactive
===========================
This chart provides an interactive exploration of Seattle weather over the
course of the year. It includes a one-axis brush selection to easily
see the distribution of weather types in a particular date range.
"""
# category: case studies
import altair as alt
from vega_datasets import data
source = data.seattle_weather()
scale = alt.Scale(domain=['sun', 'fog', 'drizzle', 'rain', 'snow'],
range=['#e7ba52', '#a7a7a7', '#aec7e8', '#1f77b4', '#9467bd'])
color = alt.Color('weather:N', scale=scale)
# We create two selections:
# - a brush that is active on the top panel
# - a multi-click that is active on the bottom panel
brush = alt.selection_interval(encodings=['x'])
click = alt.selection_multi(encodings=['color'])
# Top panel is scatter plot of temperature vs time
points = alt.Chart().mark_point().encode(
alt.X('monthdate(date):T', axis=alt.Axis(title='Date')),
alt.Y('temp_max:Q',
axis=alt.Axis(title='Maximum Daily Temperature (C)'),
scale=alt.Scale(domain=[-5, 40])
),
color=alt.condition(brush, color, alt.value('lightgray')),
#5 Which month has the highest average temperature?
average_temps = df.resample('M').mean()
average_temps[average_temps.temp == max(average_temps.temp)]
#6 Resample by the day and calculate the min and max temp for the day (Hint: .agg(['min','max'])).
# Use the resampled dataframe to calculate the change in temperature for the day. Which month has the highest daily temperature variability?
min_max = df.resample('D').agg(['min', 'max'])
min_max['temp_diff'] = min_max['temp']['max'] - min_max['temp']['min']
temp_var = min_max[['temp_diff']].resample('M').mean()
temp_var[temp_var.temp_diff == max(temp_var.temp_diff)]
# Use the dataset to answer the following questions:
# Which year and month combination has the highest amount of precipitation
df = data.seattle_weather().set_index('date')
rain = df.resample('M').sum()[['precipitation']]
rain[rain.precipitation == max(rain.precipitation)]
yearly_rain = df.resample('Y').sum()[['precipitation']]
yearly_rain[yearly_rain.precipitation == max(
yearly_rain.precipitation)].reset_index()['date'].dt.year[0]
# Visualize the amount of monthly precipitation over time.
df.resample('M').sum()['precipitation'].plot(color='blue')
# Visualize the amount of wind over time. Choose a time interval you think is appropriate?
df.resample('Q').mean()['wind'].plot(color='black')
wind_months = df.resample('D').max().resample('M').mean()['wind']
