René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"cbook._reshape_2D flattens ndarray with 2 dims (rectangular ndarray)","articleBody":"### Bug report\r\n\r\nboxplot throwing an error (below) when x is an ndarray with len(x.shape)==2 (I.E. when x is rectangular).\r\n\r\n**`ValueError: List of boxplot statistics and 'positions' values must have same the length`**\r\n\r\n**Code for reproduction**\r\n\r\n```python\r\nimport numpy as np\r\nimport matplotlib.pyplot as plt\r\n\r\na = np.array([np.array([1, 2, 3, 4]), np.array([3, 2, 7, 4]), np.array([3, 9, 3, 1, 6])]) \r\nb = np.array([np.array([1, 2, 3, 4]), np.array([3, 2, 7, 4]), np.array([3, 9, 3, 1])]) \r\n\r\n# Ragged ndarray works correctly\r\nplt.boxplot(x=a, positions=range(len(a)))\r\n\r\n# Rectangular ndarray throws error (above)\r\nplt.boxplot(x=b, positions=range(len(b)))\r\n```\r\n\r\n**Matplotlib version**\r\n\r\n- matplotlib 1.5.3 np111py35_0\r\n- python 3.5.2\r\n- Windows Server 2012 R2 Standard\r\n- conda installation\r\n\r\n\r\n**Possible cause**\r\n\r\nI believe the issue is in cbook._reshape_2D in the line below. I am not sure why that logic is in place but assume it is for good reason.\r\n```python \r\nX = [X[:, i] for i in xrange(ncols)]\r\n```\r\n\r\nGot there by looking in:\r\n\r\n1. _axes.boxplot\r\n2. _axes.bxp\r\n3. cbook.boxplot_stats\r\n\r\nreference _reshape_2D:\r\n\r\n```python\r\ndef _reshape_2D(X):\r\n    \"\"\"\r\n    Converts a non-empty list or an ndarray of two or fewer dimensions\r\n    into a list of iterable objects so that in\r\n\r\n        for v in _reshape_2D(X):\r\n\r\n    v is iterable and can be used to instantiate a 1D array.\r\n    \"\"\"\r\n    if hasattr(X, 'shape'):\r\n        # one item\r\n        if len(X.shape) == 1:\r\n            if hasattr(X[0], 'shape'):\r\n                X = list(X)\r\n            else:\r\n                X = [X, ]\r\n\r\n        # several items\r\n        elif len(X.shape) == 2:\r\n            nrows, ncols = X.shape\r\n            if nrows == 1:\r\n                X = [X]\r\n            elif ncols == 1:\r\n                X = [X.ravel()]\r\n            else:\r\n                X = [X[:, i] for i in xrange(ncols)]\r\n        else:\r\n            raise ValueError(\"input `X` must have 2 or fewer dimensions\")\r\n\r\n    if not hasattr(X[0], '__len__'):\r\n        X = [X]\r\n    else:\r\n        X = [np.ravel(x) for x in X]\r\n\r\n    return X\r\n```\r\n\r\n**Current Workaround**\r\nConverting the ndarray to a list of lists. The necessity for this workaround doesn't really make sense especially since the plotting works fine with a ragged/non-rectangular ndarray but does not work with a rectangular ndarray.","author":{"url":"https://github.com/lcapalleja","@type":"Person","name":"lcapalleja"},"datePublished":"2017-02-16T23:49:42.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":19},"url":"https://github.com/8092/matplotlib/issues/8092"}