Matplotlib Colorbar missing 1 required positional argument: 'mappable'

Question

I would like to create a scatter chart for dataframe below:

df_sample.head(10)

    duration  distance speed
0   26.299999   3.569   8.1
1   6.000000    0.739   7.4
2   25.700001   2.203   5.1
3   34.400002   2.876   5.0
4   3.000000    0.656   13.1
5   29.299999   3.704   7.6
6   10.200000   2.076   12.2
7   4.000000    0.774   11.6
8   9.200000    1.574   10.3
9   10.800000   0.782   4.3

Almost get it done with codes below. I want to add colorbar to the figure based on speed (Yellow: Slowest & Blue : Fastest) and eventually I got an error at fig.colorbar(ax=ax) at the last line. Please advise: what is mappable?

with plt.style.context('seaborn-ticks'):

    fig, ax = plt.subplots(figsize = (10, 6))

    ax.set_title('Relationship between Distance & Duration', fontdict={'fontsize': 18, 'fontweight': 'bold'}, loc='left', pad=20)
    ax.scatter(x=df_sample.duration.values, y=df_sample.distance.values, c=df_sample.speed.values, cmap=cm.YlGnBu)

    # remove top & right spines
    ax.spines['top'].set_visible(False)
    ax.spines['right'].set_visible(False)

    # equivalent to 'sns.despine(offset = 5)'
    ax.spines['left'].set_position(('outward', 5))
    ax.spines['left'].set_linewidth(1.2)
    ax.spines['bottom'].set_position(('outward', 5))
    ax.spines['bottom'].set_linewidth(1.2)
    
    # set ticks & ticklabels
    xloc = np.arange(0, 45, 5)
    ax.set_xticks(xloc)
    ax.set_xticklabels(labels=xloc, fontdict={'fontsize': 12})
    ax.set_xlabel('Minute(s)', fontdict={'fontsize': 14, 'fontweight': 'bold'})
    
    yloc = np.arange(6)
    ylab = [f"{int(num)}" for num in yloc]
    ax.set_yticks(yloc)
    ax.set_yticklabels(labels=ylab, fontdict={'fontsize': 12})
    ax.set_ylabel("Distance (KM)" , fontdict={'fontsize': 14, 'fontweight': 'bold'})
    
    fig.colorbar(ax=ax);

Answer 1

You can assign your scatterplot to a variable, for instance:

sp=ax.scatter(x=df_sample.duration.values, y=df_sample.distance.values, c=df_sample.speed.values, cmap=cm.YlGnBu)

and then pass it as a mappable object to the colorbar:

fig.colorbar(sp)