Prevent axes from being in scientific notation (powers of 10) using matplotlib in Python on semilogy plot

Question

I've read here (How to prevent numbers being changed to exponential form in Python matplotlib figure) and here (Matplotlib: disable powers of ten in log plot) and tried their solutions to no avail.

How can I convert my y-axis to display normal decimal numbers instead of scientific notation? Note this is Python 3.5.2.

Here's my code:

#Imports:
import matplotlib.pyplot as plt

possible_chars = 94
max_length = 8
pw_possibilities = [] 
for num_chars in range(1, max_length+1):
    pw_possibilities.append(possible_chars**num_chars)

x = range(1, max_length+1)
y = pw_possibilities 

#plot 
plt.figure()
plt.semilogy(x, y, 'o-')
plt.xlabel("num chars in password")
plt.ylabel("number of password possibilities")
plt.title("password (PW) possibilities verses # chars in PW")
plt.show()

Highwayman

Answer 1

How do you want to display 10^15? As 1000000000000000 ?! The other answer applies to the default formatter, when you switch to log scale a LogFormatter is used which has a different set of rules. You can switch back to ScalarFormatter and disable the offset

import matplotlib.pyplot as plt
import matplotlib.ticker as mticker
plt.ion()

possible_chars = 94
max_length = 8
pw_possibilities = [] 
for num_chars in range(1, max_length+1):
    pw_possibilities.append(possible_chars**num_chars)

x = range(1, max_length+1)
y = pw_possibilities 

#plot 
fig, ax = plt.subplots()
ax.semilogy(x, y, 'o-')
ax.set_xlabel("num chars in password")
ax.set_ylabel("number of password possibilities")
ax.set_title("password (PW) possibilities verses # chars in PW")
ax.yaxis.set_major_formatter(mticker.ScalarFormatter())
ax.yaxis.get_major_formatter().set_scientific(False)
ax.yaxis.get_major_formatter().set_useOffset(False)
fig.tight_layout()
plt.show()

See http://matplotlib.org/api/ticker_api.html for all of the available Formatter classes.

(this image is generated off of the 2.x branch, but should work on all recent version of mpl)

Cirri

Looks great! I'm showing this to a middle-schooler, so yes, although 10^15 is unreadable as 1000000000000000, it has more of an impact to them with all of the zeros. Is it possible to force a format that puts commas every 3 zeros? – Highwayman

See edit, you probably want to use matplotlib.org/api/… or matplotlib.org/api/… – Cirri

It still works! I just used this again! The secret is to put the 3 ax.yaxis. formatting lines after I set up the rest of the plot and the log-y axes! Otherwise, those changes apparently get internally overwritten when you set the semilogy and other plot parameters. – Highwayman