René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Unexpected plot behavior when running stock code for confusion matrix","articleBody":"### Bug report\r\n\r\n**Bug summary**\r\n\r\nI'm fairly new to matplotlib and I was trying to create a confusion matrix following the tutorial here: https://scikit-learn.org/stable/auto_examples/model_selection/plot_confusion_matrix.html. I downloaded the example code and when I ran it, the plots looked slightly cropped on the top and bottom edges.\r\n\r\n**Code for reproduction**\r\n\r\n```python\r\nprint(__doc__)\r\n\r\nimport numpy as np\r\nimport matplotlib.pyplot as plt\r\n\r\nfrom sklearn import svm, datasets\r\nfrom sklearn.model_selection import train_test_split\r\nfrom sklearn.metrics import confusion_matrix\r\nfrom sklearn.utils.multiclass import unique_labels\r\n\r\n# import some data to play with\r\niris = datasets.load_iris()\r\nX = iris.data\r\ny = iris.target\r\nclass_names = iris.target_names\r\n\r\n# Split the data into a training set and a test set\r\nX_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\r\n\r\n# Run classifier, using a model that is too regularized (C too low) to see\r\n# the impact on the results\r\nclassifier = svm.SVC(kernel='linear', C=0.01)\r\ny_pred = classifier.fit(X_train, y_train).predict(X_test)\r\n\r\n\r\ndef plot_confusion_matrix(y_true, y_pred, classes,\r\n                          normalize=False,\r\n                          title=None,\r\n                          cmap=plt.cm.Blues):\r\n    \"\"\"\r\n    This function prints and plots the confusion matrix.\r\n    Normalization can be applied by setting `normalize=True`.\r\n    \"\"\"\r\n    if not title:\r\n        if normalize:\r\n            title = 'Normalized confusion matrix'\r\n        else:\r\n            title = 'Confusion matrix, without normalization'\r\n\r\n    # Compute confusion matrix\r\n    cm = confusion_matrix(y_true, y_pred)\r\n    # Only use the labels that appear in the data\r\n    classes = classes[unique_labels(y_true, y_pred)]\r\n    if normalize:\r\n        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\r\n        print(\"Normalized confusion matrix\")\r\n    else:\r\n        print('Confusion matrix, without normalization')\r\n\r\n    print(cm)\r\n\r\n    fig, ax = plt.subplots()\r\n    im = ax.imshow(cm, interpolation='nearest', cmap=cmap)\r\n    ax.figure.colorbar(im, ax=ax)\r\n    # We want to show all ticks...\r\n    ax.set(xticks=np.arange(cm.shape[1]),\r\n           yticks=np.arange(cm.shape[0]),\r\n           # ... and label them with the respective list entries\r\n           xticklabels=classes, yticklabels=classes,\r\n           title=title,\r\n           ylabel='True label',\r\n           xlabel='Predicted label')\r\n\r\n    # Rotate the tick labels and set their alignment.\r\n    plt.setp(ax.get_xticklabels(), rotation=45, ha=\"right\",\r\n             rotation_mode=\"anchor\")\r\n\r\n    # Loop over data dimensions and create text annotations.\r\n    fmt = '.2f' if normalize else 'd'\r\n    thresh = cm.max() / 2.\r\n    for i in range(cm.shape[0]):\r\n        for j in range(cm.shape[1]):\r\n            ax.text(j, i, format(cm[i, j], fmt),\r\n                    ha=\"center\", va=\"center\",\r\n                    color=\"white\" if cm[i, j] \u003e thresh else \"black\")\r\n    fig.tight_layout()\r\n    return ax\r\n\r\n\r\nnp.set_printoptions(precision=2)\r\n\r\n# Plot non-normalized confusion matrix\r\nplot_confusion_matrix(y_test, y_pred, classes=class_names,\r\n                      title='Confusion matrix, without normalization')\r\n\r\n# Plot normalized confusion matrix\r\nplot_confusion_matrix(y_test, y_pred, classes=class_names, normalize=True,\r\n                      title='Normalized confusion matrix')\r\n\r\nplt.show()\r\n```\r\n\r\n**Actual outcome**\r\n\r\n![Screenshot from 2019-10-03 10-12-56](https://user-images.githubusercontent.com/56123251/66134690-b2f1ff00-e5c6-11e9-9c34-f96644eaf220.png)\r\n\r\n**Expected outcome**\r\n\r\nThe plots were expected to not look cropped.\r\n\r\n**Matplotlib version**\r\n\r\n  * Operating system: Ubunto 18.04\r\n  * Matplotlib version: 3.1.1\r\n  * Matplotlib backend (`print(matplotlib.get_backend())`):\r\n  * Python version: 3.7\r\n  * Jupyter version (if applicable):\r\n  * Other libraries: \r\n\r\nmatplotlib installed via pip.","author":{"url":"https://github.com/vkng12","@type":"Person","name":"vkng12"},"datePublished":"2019-10-03T14:20:34.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":1},"url":"https://github.com/15368/matplotlib/issues/15368"}