|
1 | 1 | """ |
2 | | -======== |
3 | | -Log Demo |
4 | | -======== |
| 2 | +========= |
| 3 | +Log scale |
| 4 | +========= |
5 | 5 |
|
6 | 6 | Examples of plots with logarithmic axes. |
| 7 | +
|
| 8 | +You can set the x/y axes to be logarithmic by passing "log" to `~.Axes.set_xscale` / |
| 9 | +`~.Axes.set_yscale`. |
| 10 | +
|
| 11 | +Since plotting data on semi-logarithmic or double-logarithmic scales is very common, |
| 12 | +the functions `~.Axes.semilogx`, `~.Axes.semilogy`, and `~.Axes.loglog` are shortcuts |
| 13 | +for setting the scale and plotting data; e.g. ``ax.semilogx(x, y)`` is equivalent to |
| 14 | +``ax.set_xscale('log'); ax.plot(x, y)``. |
7 | 15 | """ |
8 | 16 |
|
9 | 17 | import matplotlib.pyplot as plt |
10 | 18 | import numpy as np |
11 | 19 |
|
12 | | -# Data for plotting |
13 | | -t = np.arange(0.01, 20.0, 0.01) |
14 | | - |
15 | | -# Create figure |
16 | | -fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2) |
17 | | - |
18 | | -# log y axis |
19 | | -ax1.semilogy(t, np.exp(-t / 5.0)) |
20 | | -ax1.set(title='semilogy') |
| 20 | +fig, (ax1, ax2, ax3) = plt.subplots(1, 3, layout='constrained', figsize=(7, 7/3)) |
| 21 | +# log x axis |
| 22 | +t = np.arange(0.01, 10.0, 0.01) |
| 23 | +ax1.semilogx(t, np.sin(2 * np.pi * t)) |
| 24 | +ax1.set(title='semilogx') |
21 | 25 | ax1.grid() |
| 26 | +ax1.grid(which="minor", color="0.9") |
22 | 27 |
|
23 | | -# log x axis |
24 | | -ax2.semilogx(t, np.sin(2 * np.pi * t)) |
25 | | -ax2.set(title='semilogx') |
| 28 | +# log y axis |
| 29 | +x = np.arange(4) |
| 30 | +ax2.semilogy(4*x, 10**x, 'o--') |
| 31 | +ax2.set(title='semilogy') |
26 | 32 | ax2.grid() |
| 33 | +ax2.grid(which="minor", color="0.9") |
27 | 34 |
|
28 | 35 | # log x and y axis |
29 | | -ax3.loglog(t, 20 * np.exp(-t / 10.0)) |
30 | | -ax3.set_xscale('log', base=2) |
31 | | -ax3.set(title='loglog base 2 on x') |
| 36 | +x = np.array([1, 10, 100, 1000]) |
| 37 | +ax3.loglog(x, 5 * x, 'o--') |
| 38 | +ax3.set(title='loglog') |
32 | 39 | ax3.grid() |
33 | | - |
34 | | -# With errorbars: clip non-positive values |
35 | | -# Use new data for plotting |
36 | | -x = 10.0**np.linspace(0.0, 2.0, 20) |
37 | | -y = x**2.0 |
38 | | - |
39 | | -ax4.set_xscale("log", nonpositive='clip') |
40 | | -ax4.set_yscale("log", nonpositive='clip') |
41 | | -ax4.set(title='Errorbars go negative') |
42 | | -ax4.errorbar(x, y, xerr=0.1 * x, yerr=5.0 + 0.75 * y) |
43 | | -# ylim must be set after errorbar to allow errorbar to autoscale limits |
44 | | -ax4.set_ylim(bottom=0.1) |
45 | | - |
46 | | -fig.tight_layout() |
| 40 | +ax3.grid(which="minor", color="0.9") |
| 41 | + |
| 42 | +# %% |
| 43 | +# Using the *base* parameter, one can set the base of the logarithm |
| 44 | +fig, ax = plt.subplots() |
| 45 | +ax.bar(["L1 cache", "L2 cache", "L3 cache", "RAM", "SSD"], |
| 46 | + [32, 1_000, 32_000, 16_000_000, 512_000_000]) |
| 47 | +ax.set_yscale('log', base=2) |
| 48 | +ax.set_yticks([1, 2**10, 2**20, 2**30], labels=['kB', 'MB', 'GB', 'TB']) |
| 49 | +ax.set_title("Typical memory sizes") |
| 50 | +ax.yaxis.grid() |
47 | 51 | plt.show() |
0 commit comments