{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Making highly customizable plot of Signal2D using matplotlib"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook shows how to make a publication quality figure of a HyperSpy Signal2D object, where we use the matplotlib figure and subplot object to customize the plot."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Requires"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "HyperSpy 1.3 or later\n",
    "\n",
    "Matplotlib 2.0 or later"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Credits"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 2017/09/10 Magnus Nord. Initial version"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Making a Signal2D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import hyperspy.api as hs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Firstly, we make a 2D signal by generating a numpy array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "image_data = np.arange(10000).reshape(100, 100) + np.random.random(size=(100, 100))*5000"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "HyperSpy signals can be made by passing a numpy array to a HyperSpy signal class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "s = hs.signals.Signal2D(image_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Making the matplotlib figure"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In addition to the pyplot module, we also need to import the size bar, font manager and patheffects. As the three latter  modules are not present in the main pyplot module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.axes_grid1.anchored_artists import AnchoredSizeBar\n",
    "import matplotlib.font_manager as fm\n",
    "import matplotlib.patheffects as patheffects"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The figsize argument is used to make sure the figure has the same length in both the x- and y-directions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(6, 6))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Image data is normally not plotted with ticks on the axis, so we disable these using `set_axis_off`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "ax.set_axis_off()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting the image data itself is done by directly accessing the numpy array in the signal: `s.data`.\n",
    "\n",
    "The calibrated size of the signal is contained in s.axes_manager.signal_extent, which is needed if we want the scalebar to have the correct size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "cax = ax.imshow(s.data, extent=s.axes_manager.signal_extent, cmap='viridis')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "cax can be used to set color limits for the plotted image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "cax.set_clim(200, 11000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scalebar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In matplotlib, what is normally referred to as a scalebar, is called an AnchoredSizeBar.\n",
    "\n",
    "This AnchoredSizeBar takes a large amount of arguments, which can be seen in the docstring by running AnchoredSizeBar?\n",
    "\n",
    "To set the size of the text in the scalebar, the fontproperties object is used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fontprops = fm.FontProperties(size=30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scalebar = AnchoredSizeBar(\n",
    "        transform=ax.transData, size=30, label='30 nm',\n",
    "        loc=4, frameon=False,\n",
    "        color='black', size_vertical=3,\n",
    "        label_top=False,\n",
    "        fontproperties=fontprops)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To increase the contrast of the scalebar text, use the patheffects feature. This adds a border around the text. The capstyle argument needs to be used to get a continuous border."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scalebar.txt_label._text.set_path_effects([patheffects.withStroke(linewidth=2, foreground='white', capstyle=\"round\")])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ax.add_artist(scalebar)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The next line is needed due to a bug in the current version of matplotlib (2.0.2), which leads to the scalebar being not solid. This has been fixed in the current development version."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scalebar.size_bar.get_children()[0].fill = True"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Arrow annotation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using ax.annotate, we add an arrow point to some xy coordinate. The xytext specifies either the tail or the text of the arrow. If no text is wanted, use s=''."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "annotation = ax.annotate(s=\"An arrow\", xy=(20, 70), xytext=(40, 50), \n",
    "            arrowprops={'arrowstyle':'->'},\n",
    "            path_effects=[patheffects.withStroke(linewidth=2, foreground='white', capstyle=\"round\")])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To increase the contrast of arrow, add a border in a different color."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "annotation.arrow_patch.set_path_effects([patheffects.withStroke(linewidth=2, foreground='white', capstyle=\"round\")])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ax.text(0.05, 0.95, \"Some random data\", size=20, transform=ax.transAxes,\n",
    "        verticalalignment='top', horizontalalignment='left', color='white',\n",
    "       path_effects=[patheffects.withStroke(linewidth=3, foreground='black', capstyle=\"round\")])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Saving the figure"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For figures of image data, it is possible to make the figure borderless by using the subplot_adjust.\n",
    "\n",
    "Note, this only works well if the axis ticks are deactivated."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "fig.subplots_adjust(left=0., bottom=0., right=1., top=1.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig.savefig(\"image_data_figure.png\", dpi=300)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "hyperspy_dev",
   "language": "python",
   "name": "hyperspy_dev"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}