Dockerfile

FROM ghcr.io/kbredies/gratopy:pocl-latest

ARG NB_UID=1000
ARG NB_USER=gratopy

ENV USER ${NB_USER}
ENV NB_UID ${NB_UID}
ENV HOME /home/${NB_USER}

RUN adduser --disabled-password \
    --gecos "Default user" \
    --uid ${NB_UID} \
    ${NB_USER}

RUN pip3 install notebook

USER ${NB_USER}
COPY . ${HOME}
USER root
RUN chown -R ${NB_UID} ${HOME}

USER ${NB_USER}
ENV PATH="${PATH}:${HOME}/.local/bin"
WORKDIR ${HOME}

examples.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "094bd1fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pyopencl as cl\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import gratopy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0cb61f49",
   "metadata": {},
   "outputs": [],
   "source": [
    "# check that gratopy has been imported correctly\n",
    "gratopy.VERSION"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9afef88b",
   "metadata": {},
   "source": [
    "# Example 1\n",
    "\n",
    "... from https://gratopy.readthedocs.io/en/latest/getting_started.html#first-example-radon-transform"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5db64b23",
   "metadata": {},
   "outputs": [],
   "source": [
    "# discretization parameters\n",
    "number_angles = 60\n",
    "number_detectors = 300\n",
    "Nx = 300\n",
    "# Alternatively to number_angles one could give as angle input\n",
    "# angles = np.linspace(0, np.pi, number_angles+1)[:-1]\n",
    "\n",
    "# create pyopencl context\n",
    "ctx = cl.create_some_context()\n",
    "queue = cl.CommandQueue(ctx)\n",
    "\n",
    "# create phantom as test image (a pyopencl.array.Array of dimensions (Nx, Nx))\n",
    "phantom = gratopy.phantom(queue,Nx)\n",
    "\n",
    "# create suitable projectionsettings\n",
    "PS = gratopy.ProjectionSettings(queue, gratopy.RADON, phantom.shape,\n",
    "                                number_angles, number_detectors)\n",
    "\n",
    "# compute forward projection and backprojection of created sinogram\n",
    "# results are pyopencl arrays\n",
    "sino = gratopy.forwardprojection(phantom, PS)\n",
    "backproj = gratopy.backprojection(sino, PS)\n",
    "\n",
    "# plot results\n",
    "plt.figure()\n",
    "plt.title(\"Generated Phantom\")\n",
    "plt.imshow(phantom.get(), cmap=\"gray\")\n",
    "\n",
    "plt.figure()\n",
    "plt.title(\"Sinogram\")\n",
    "plt.imshow(sino.get(), cmap=\"gray\")\n",
    "\n",
    "plt.figure()\n",
    "plt.title(\"Backprojection\")\n",
    "plt.imshow(backproj.get(), cmap=\"gray\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e99940b",
   "metadata": {},
   "source": [
    "# Example 2\n",
    "... from https://gratopy.readthedocs.io/en/latest/getting_started.html#second-example-fanbeam-transform"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4d064136",
   "metadata": {},
   "outputs": [],
   "source": [
    "# discretization parameters\n",
    "number_angles = 60\n",
    "number_detectors = 300\n",
    "image_shape = (500, 500)\n",
    "\n",
    "# create pyopencl context\n",
    "ctx = cl.create_some_context()\n",
    "queue = cl.CommandQueue(ctx)\n",
    "\n",
    "# physical parameters\n",
    "my_detector_width = 120\n",
    "my_R = 200\n",
    "my_RE = 100\n",
    "\n",
    "# fanbeam setting with automatic image_width\n",
    "PS1 = gratopy.ProjectionSettings(queue, gratopy.FANBEAM,\n",
    "                                 img_shape=image_shape,\n",
    "                                 angles=number_angles,\n",
    "                                 n_detectors=number_detectors,\n",
    "                                 detector_width=my_detector_width,\n",
    "                                 R=my_R, RE=my_RE)\n",
    "\n",
    "\n",
    "print(\"image_width chosen by gratopy: {:.2f}\".format((PS1.image_width)))\n",
    "\n",
    "# fanbeam setting with set image_width\n",
    "my_image_width = 80.0\n",
    "PS2 = gratopy.ProjectionSettings(queue, gratopy.FANBEAM,\n",
    "                                 img_shape=image_shape,\n",
    "                                 angles=number_angles,\n",
    "                                 n_detectors=number_detectors,\n",
    "                                 detector_width=my_detector_width,\n",
    "                                 R=my_R, RE=my_RE,\n",
    "                                 image_width=my_image_width)\n",
    "\n",
    "# plot geometries associated to these projectionsettings\n",
    "fig, (axes1, axes2) = plt.subplots(1,2)\n",
    "PS1.show_geometry(np.pi/4, figure=fig, axes=axes1, show=False)\n",
    "PS2.show_geometry(np.pi/4, figure=fig, axes=axes2, show=False)\n",
    "axes1.set_title(\"Geometry chosen by gratopy as: {:.2f}\".format((PS1.image_width)))\n",
    "axes2.set_title(\"Geometry for manually-chosen image_width as: {:.2f}\"\n",
    "                .format((my_image_width)))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8245d535",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}