René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[ENH]: Add split feature for violin plots","articleBody":"### Problem\r\n\r\nRecently, I wanted to get split violin plots at defined x-positions (see image).\r\nMy problem is that while it's possible to specify the x-position in `pyplot.violinplot()`, there is no option for split violins.\r\n~~`seaborn.violinplot()` on the other hand allows split violins but seems to enforce discrete x-positions with fixed intervals.~~ This is not true anymore since v0.13\r\n\r\nThis example shows two data sets (black and red/blue) in four categories defined by $A=0,1$ and $B=0,1$.\r\n![example_violins](https://github.com/matplotlib/matplotlib/assets/15108686/90142ce9-ac66-4915-aa15-41dc5d2eab64)\r\n\r\nThe code that produces the plot based on [this answer on stackoverflow](https://stackoverflow.com/a/29781988).\r\n```python\r\nimport matplotlib.pyplot as plt\r\nimport numpy as np\r\n\r\nnp.random.seed(0)\r\n\r\ncols = [\"firebrick\",\"dodgerblue\"]\r\nleg = {\"patches\": [], \"labels\": []}\r\n\r\n# Loop over categories A and B\r\nfor A in range(0,2):\r\n    xpos   = []\r\n    dLeft  = []\r\n    dRight = []\r\n\r\n    for B in range(0,2):\r\n        # Create data for the left and right violins in the current category\r\n        mu = np.random.randint(10,100)\r\n        dLeft.append(np.random.poisson(mu,1000))\r\n        dRight.append(np.random.normal(mu+10,np.sqrt(mu+10),1000))\r\n\r\n        # Create x-positions for the violins\r\n        xpos.append(B+(A-0.5)*0.15)\r\n\r\n        # Create right violins (colors)\r\n        vRight = plt.violinplot(dRight,positions=xpos,widths=0.1,showextrema=False,showmedians=False)\r\n        for vr in vRight[\"bodies\"]:\r\n            # Get the center\r\n            m = np.mean(vr.get_paths()[0].vertices[:, 0])\r\n            # Get the rightmost point\r\n            r = np.max(vr.get_paths()[0].vertices[:, 0])\r\n            # Only allow values right of the center\r\n            vr.get_paths()[0].vertices[:, 0] = np.clip(vr.get_paths()[0].vertices[:, 0], m, r)\r\n            # Set category A color\r\n            vr.set_color(cols[A])\r\n            vr.set_alpha(1.)\r\n\r\n        # Create left violins (black)\r\n        vLeft = plt.violinplot(dLeft,positions=xpos,widths=0.15,showextrema=False,showmedians=False)\r\n        for vl in vLeft[\"bodies\"]:\r\n            # Get the center\r\n            m = np.mean(vl.get_paths()[0].vertices[:, 0])\r\n            # Get the leftmost point\r\n            l = np.min(vl.get_paths()[0].vertices[:, 0])\r\n            # Only allow values left of the center\r\n            vl.get_paths()[0].vertices[:, 0] = np.clip(vl.get_paths()[0].vertices[:, 0], l, m)\r\n            # Set color\r\n            vl.set_color(\"k\")\r\n            vl.set_alpha(1.)\r\n\r\n    # Add legend entry\r\n    leg[\"patches\"].append(vRight[\"bodies\"][0])\r\n    leg[\"labels\"].append(fr\"$A={A}$\")\r\n\r\nplt.legend(leg[\"patches\"],leg[\"labels\"])\r\nplt.xlabel(r\"$B$\")\r\nplt.xticks([0,1],[\"0\",\"1\"])\r\nplt.xlim(-0.5,1.5)\r\nplt.tight_layout()\r\nplt.savefig(\"example_violins.png\")\r\n```\r\n\r\n### Proposed solution\r\n\r\nAn easy-to-use solution would be to add an argument in the `violinplot()` function through which the user can define which side to plot. Something like `asymetric=None, \"below\", \"above\"`, where `\"below\"`/`\"above\"` only plots the side below/above the values given to `positions`.\r\n\r\nThis would mean that calling\r\n```python\r\nvRight = plt.violinplot(dRight,positions=xpos,widths=0.1,showextrema=False,showmedians=False,asymetric=\"above\")\r\n```\r\ninternally executes this loop\r\n```python\r\nfor vr in vRight[\"bodies\"]:\r\n    # Get the center\r\n    m = np.mean(vr.get_paths()[0].vertices[:, 0])\r\n    # Get the rightmost point\r\n    r = np.max(vr.get_paths()[0].vertices[:, 0])\r\n    # Only allow values right of the center\r\n    vr.get_paths()[0].vertices[:, 0] = np.clip(vr.get_paths()[0].vertices[:, 0], m, r)\r\n```","author":{"url":"https://github.com/anjabeck","@type":"Person","name":"anjabeck"},"datePublished":"2024-02-22T14:35:55.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":4},"url":"https://github.com/27812/matplotlib/issues/27812"}