René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"plotting using secondary axis with mpl_toolkits.axes_grid1.parasite_axes.SubplotHost().twin() produces very weird outputs","articleBody":"For several plots I'd like to have a secondary x axis that is a non-affine transformation of the original axis. Following a very old discussion on [nabble](http://matplotlib.1069221.n5.nabble.com/Dual-x-axes-with-transformation-td10865.html) I recognized that it is only possible, if I define my own matplotlib.transforms.Transform() class or I have to set tick locations manually according to a discussion on [stackoverflow](http://stackoverflow.com/questions/10514315/how-to-add-a-second-x-axis-in-matplotlib). I managed to make the code from the nabble link work, but I had to rename the class method transform() to transform_non_affine(). Otherwise it would not work.\nSo I now have the following code, which seemed to work fine for the first few tests.\n\n```\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport matplotlib.transforms as mtransforms\nfrom mpl_toolkits.axes_grid1.parasite_axes import SubplotHost\n\n\ndef transformed_x_axis(ax, transformation):\n\n    class MyTransform(mtransforms.Transform):\n        input_dims = 1\n        output_dims = 1\n        is_separable = False\n        has_inverse = False\n\n        def transform_non_affine(self, x):\n            return transformation(x)\n\n        def inverted(self):\n            return MyInvertedTransform()\n\n    class MyInvertedTransform(MyTransform):\n        def inverted(self):\n            return MyTransform()\n\n    aux_trans = mtransforms.BlendedGenericTransform(\n            MyTransform(), mtransforms.IdentityTransform())\n\n    ax2 = ax.twin(aux_trans)\n    ax2.set_viewlim_mode(\"transform\")\n    ax2.axis[\"right\"].toggle(ticklabels=False)\n    return ax2\n```\n\nFor convenience I defined a function 'transformed_x_axis()' that just accepts an original matplotlib.axes.Axes object and a transformation formula.\n(Actually, I think that my use case is pretty common and matplotlib should provide such kind of function that I tried to use, here.)\n\nUsing it with\n\n```\nfig = plt.figure()\nax = SubplotHost(fig, 1,1,1)\nfig.add_subplot(ax)\n\nax2 = transformed_x_axis(ax, lambda x: 1. / x)\n\nx = np.linspace(1., 2., 200)\nax.plot(x, np.sin(x))\nplt.show()\n```\n\nproduces exactly what I want -- a plot with a secondary transformed x axis:\n![01](https://cloud.githubusercontent.com/assets/7665678/19308706/53662004-9081-11e6-89dd-9b9adbc9e7bf.png)\n\nUnfortunately, for other transformations or for other data limits I see all kinds of weirdest things happening.\n\nFor other data ranges, e.g. `x = np.linspace(1., 5., 200)`, the ticker has problems and draws ticklabels on top of each other:\n![02](https://cloud.githubusercontent.com/assets/7665678/19308895/3dc33eac-9082-11e6-86e1-f35549a89cbe.png)\n\nIf the axis approaches the critical point 0.0 of my transformation, the ticks don'e even fill the axis but only a tiny part of it:\n![03](https://cloud.githubusercontent.com/assets/7665678/19309003/b6549276-9082-11e6-8a16-9fb948e96998.png)\n\nI guess that this is a result of the linear scale of the ticklabels. It seems the ticks for the axis are chosen by value and not by position in axes coordinates, which I find kind of wrong.\n\nOther transformations also produce weird things, e.g.\n\n```\nfig = plt.figure()\nax = SubplotHost(fig, 1,1,1)\nfig.add_subplot(ax)\n\nax2 = transformed_x_axis(ax, lambda x: np.exp(x))\n\nx = np.linspace(1., 5., 200)\nax.plot(x, np.sin(x))\nplt.show()\n```\n\ngives me a RuntimeError in an IPython Notebook and in a terminal it produces a secondary axis that contains nothing but the value 0.\n![04](https://cloud.githubusercontent.com/assets/7665678/19309474/408e4462-9085-11e6-942a-424feb9bc618.png)\n\nFor other data limits,\n\n```\nax2 = transformed_x_axis(ax, lambda x: np.exp(x))\nx = np.linspace(1., 2., 200)\nax.plot(x, np.sin(x))\n```\n\nI get a MemoryError in the notebook. When done in the terminal, I see no secondary x axis.\n\nThe strangest things occur for simple linear transformations:\n\n```\nfig = plt.figure()\nax = SubplotHost(fig, 1,1,1)\nfig.add_subplot(ax)\n\nax2 = transformed_x_axis(ax, lambda x: 2. * x)\n\nx = np.linspace(1., 2., 200)\nax.plot(x, np.sin(x))\nplt.show()\n```\n\nHere, in the terminal, I see, again, no secondary axis. But in the notebook, I think one can see that something goes dramatically wrong:\n![05](https://cloud.githubusercontent.com/assets/7665678/19309566/f6e4dbea-9085-11e6-8ea3-b3bbc45d93b7.png)\n\nI think, it is the ticker that does those weird things when it faces transformed axes.\n\nAnyway, the behavior is inconsistent, unexpected and very weird.\n\nCheers, Gerhard\n\n---\n\nPython 2.7.12, matplotlib 1.5.1 on Windows 64 bit\neverything installed via Anaconda\n","author":{"url":"https://github.com/griai","@type":"Person","name":"griai"},"datePublished":"2016-10-12T12:17:41.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":14},"url":"https://github.com/7258/matplotlib/issues/7258"}