René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[Bug]: hexbin cannot always accept np.max like functions as reduce_C_function","articleBody":"### Bug summary\r\n\r\nThe default reduce_C_function (np.mean) returns np.nan for size 0 sequence and hexbin does not cause Error. However, the default mincnt (=0) and reduce_C_function=np.max can cause Error for some gridsize \u0026 extent settings due to zero-sized sequence. \r\n\r\n### Code for reproduction\r\n\r\n```python\r\nimport numpy as np\r\nimport matplotlib.pyplot as plt\r\nx=np.array([0. , 1. , 2. , 0.5, 1.5, 2.5, 0. , 1. , 2. ])\r\ny=np.array([0.        , 0.        , 0.        , 0.8660254 , 0.8660254 ,\r\n       0.8660254 , 1.73205081, 1.73205081, 1.73205081])\r\nextent=(0,2.5,0,np.sqrt(3))\r\nC=np.arange(x.size)\r\ngridsize=(3,1)\r\nplt.hexbin(x,y,C, extent=extent, reduce_C_function=np.max, gridsize=gridsize)\r\n# ValueError: zero-size array to reduction operation maximum which has no identity\r\n```\r\n\r\n\r\n### Actual outcome\r\n\r\nValueError will be raised by numpy max(), which cannot accept zero sized sequence.\r\n\r\n### Expected outcome\r\n\r\nNo error is raised. \r\n\r\n### Additional information\r\n\r\nThis bug/new specification has been introduced by \r\n\r\nhttps://github.com/matplotlib/matplotlib/commit/fcda8c8121ba9007cc57a24fa4ff7bc5a17b6a1c\r\n\r\nApparently, the matplotlib versions\u003e=3.8.0 would have the behavior but versions\u003c3.8.0 do not.\r\n\r\nThe default reduce_C_function (np.mean) returns np.nan for size 0 sequence and it does not cause Error. However, the default mincnt (None and mincnt=0 is assigned for None) and reduce_C_function=np.max can cause Error due to zero-sized sequence. The old versions use 'len(acc) \u003e mincnt' comparison and it kicks size 0 sequence by default. However, the new version uses 'len(acc) \u003e= mincnt' comparison but the default mincnt is still 0. Hence, 0 sized input is not kicked and numpy max() like functions raise Error. It can be fixed by mincnt=0.5 (1 is not compatible with the old matplotlib versions).\r\n\r\nplt.hexbin(x,y,C, extent=extent, reduce_C_function=np.max, gridsize=gridsize, mincnt=0.5)\r\n\r\nIf this is a desired (planed) behavior, it would be nice if the documents (release notes) describe the behavior explicitly. At least my program won't work before fixing the problem, and I have to read the codes and their commit history to find the root cause. \r\n\r\nBest regards\r\n\r\n### Operating system\r\n\r\nWindows and Ubuntu\r\n\r\n### Matplotlib Version\r\n\r\n3.8.0\r\n\r\n### Matplotlib Backend\r\n\r\nQt5Agg or Agg\r\n\r\n### Python version\r\n\r\n3.11.4\r\n\r\n### Jupyter version\r\n\r\nNA\r\n\r\n### Installation\r\n\r\npip","author":{"url":"https://github.com/materia9","@type":"Person","name":"materia9"},"datePublished":"2023-10-16T10:42:19.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":2},"url":"https://github.com/27103/matplotlib/issues/27103"}