René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"`MaxNLocator` changes the scientific notation exponent with different number of tick labels","articleBody":"### Bug report\r\n\r\n**Bug summary**\r\n\r\n`MaxNLocator` changes the scientific notation with different number of tick labels.\r\n\r\n**Code for reproduction**\r\n\r\nI encountered this unexpected behavior when I changed the y-ticklabel size. Going from size 13 to 14 changes the scientific notation from 1e9 to 1e10, and adds an decimal offset to all the ticklabels.\r\n\r\n```python\r\nimport matplotlib.pyplot as plt\r\n\r\nsci_nums = [10**9, 8*10**9]\r\nplt.rcParams['ytick.labelsize'] = 13\r\nfig, ax = plt.subplots()\r\nax.scatter(range(len(sci_nums)), sci_nums)\r\n```\r\n\r\n![image](https://user-images.githubusercontent.com/4560057/45266103-bf3b7500-b423-11e8-8203-da7c77a97e81.png)\r\n\r\n```python\r\nplt.rcParams['ytick.labelsize'] = 14\r\nfig, ax = plt.subplots()\r\nax.scatter(range(len(sci_nums)), sci_nums)\r\n```\r\n![image](https://user-images.githubusercontent.com/4560057/45266108-c9f60a00-b423-11e8-829b-2e08ce0b1dfa.png)\r\n\r\nThis is related to that the larger font size reduces the number of tickslabels. `MaxNLocator` apparently adjusts the scientific notation depending on the number of ticklabels.\r\n\r\n```python\r\nplt.rcParams['ytick.labelsize'] = 13\r\nfig, ax = plt.subplots()\r\nax.scatter(range(len(sci_nums)), sci_nums)\r\n\r\nimport matplotlib.ticker as ticker\r\nax.yaxis.set_major_locator(ticker.MaxNLocator(4))\r\n```\r\n![image](https://user-images.githubusercontent.com/4560057/45266138-3d981700-b424-11e8-932c-af3a74e29e8e.png)\r\n\r\nI dont' quite understand the rule that `MaxNLocator` uses. Setting 1, 2, 4 or 5 ticklabels results in `1e10`. Setting 3, 6, 7, 8, or 9 ticklabels results in `1e9`.\r\n\r\n**Expected outcome**\r\nI would expect that changing the number of ticklabels (either directly or via another parameter such as the font size) would not result in a change in the formatting of the scientific notation. In this case, I would expect it to remain `1e9` for all the examples above.\r\n\r\n**Matplotlib version**\r\n\r\n  * Operating system: `Linux`\r\n  * Matplotlib version: `2.2.2`\r\n  * Matplotlib backend (`print(matplotlib.get_backend())`): `module://ipykernel.pylab.backend_inline`\r\n  * Python version: `3.6.6` (conda-forge)\r\n  * Jupyter version (if applicable): `jupyterlab=0.34.3`, jupyter_core=`4.4.0`\r\n  * Other libraries: \r\n","author":{"url":"https://github.com/joelostblom","@type":"Person","name":"joelostblom"},"datePublished":"2018-09-09T15:46:57.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":8},"url":"https://github.com/12072/matplotlib/issues/12072"}