CT_reconstruction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CT Reconstruction\n",
    "\n",
    "This notebook should only be used if re-creation of the full environment `environment.yml` failed. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # You can run this code on Google Colab.\n",
    "# # Install conda\n",
    "# !pip install -q condacolab\n",
    "# import condacolab\n",
    "# condacolab.install()\n",
    "# # Install astra-toolbox\n",
    "# !conda install -c astra-toolbox astra-toolbox\n",
    "# # Install odl\n",
    "# !pip3 install https://github.com/odlgroup/odl/archive/master.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No GPU support available\n",
      "ASTRA Toolbox v2.1.2\n",
      "Testing basic CPU 2D functionality... Ok\n"
     ]
    }
   ],
   "source": [
    "# import libraries for CT and deep learning\n",
    "import odl\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import astra\n",
    "astra.test()\n",
    "from skimage.metrics import peak_signal_noise_ratio as compare_psnr\n",
    "from skimage.metrics import structural_similarity as compare_ssim\n",
    "from time import time\n",
    "import torch.nn as nn\n",
    "import argparse\n",
    "import os\n",
    "\n",
    "import torch\n",
    "import odl\n",
    "import odl.contrib.torch as odl_torch\n",
    "\n",
    "parser = argparse.ArgumentParser()\n",
    "parser.add_argument('--output_dir', default='./figures', type=str, help='Path to save the plot')\n",
    "args = parser.parse_args('--output_dir ./figures'.split(' '))\n",
    "os.makedirs(args.output_dir, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FBP reconstruction time = 0.005 seconds\n",
      "sinogram size = (30, 365)\n",
      "ADMM reconstruction time = 4.809 seconds\n",
      "number of model parameters = 50740\n",
      "Training LGD network...\n",
      "epoch = 0, loss = 0.04429273307323456\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[3], line 182\u001b[0m\n\u001b[1;32m    180\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m'''################## YOUR CODE HERE #################### '''\u001b[39;00m\n\u001b[1;32m    181\u001b[0m \u001b[38;5;66;03m#### You need to compute the reconstructed image by applying a forward pass through lgd_net, compute the loss, call the backward function on loss, and update the parameters.\u001b[39;00m\n\u001b[0;32m--> 182\u001b[0m recon \u001b[38;5;241m=\u001b[39m \u001b[43mlgd_net\u001b[49m\u001b[43m(\u001b[49m\u001b[43my\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx_init\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    183\u001b[0m loss \u001b[38;5;241m=\u001b[39m mse_loss(recon, ground_truth)\n\u001b[1;32m    184\u001b[0m loss\u001b[38;5;241m.\u001b[39mbackward()\n",
      "File \u001b[0;32m~/miniconda3/envs/ia/lib/python3.12/site-packages/torch/nn/modules/module.py:1532\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1530\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_compiled_call_impl(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)  \u001b[38;5;66;03m# type: ignore[misc]\u001b[39;00m\n\u001b[1;32m   1531\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 1532\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_impl\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/ia/lib/python3.12/site-packages/torch/nn/modules/module.py:1541\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1536\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1537\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1538\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1539\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1540\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1541\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1543\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m   1544\u001b[0m     result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n",
      "Cell \u001b[0;32mIn[3], line 149\u001b[0m, in \u001b[0;36mLGD_net.forward\u001b[0;34m(self, y, x_init)\u001b[0m\n\u001b[1;32m    147\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m k \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mniter):\n\u001b[1;32m    148\u001b[0m     f_grad \u001b[38;5;241m=\u001b[39m adj_op(fwd_op(x) \u001b[38;5;241m-\u001b[39m y)\n\u001b[0;32m--> 149\u001b[0m     x \u001b[38;5;241m=\u001b[39m x \u001b[38;5;241m-\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstep_size[k] \u001b[38;5;241m*\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mprox\u001b[49m\u001b[43m[\u001b[49m\u001b[43mk\u001b[49m\u001b[43m]\u001b[49m\u001b[43m(\u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\u001b[43mf_grad\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    151\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m x\n",
      "File \u001b[0;32m~/miniconda3/envs/ia/lib/python3.12/site-packages/torch/nn/modules/module.py:1532\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1530\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_compiled_call_impl(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)  \u001b[38;5;66;03m# type: ignore[misc]\u001b[39;00m\n\u001b[1;32m   1531\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 1532\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_impl\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/ia/lib/python3.12/site-packages/torch/nn/modules/module.py:1541\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1536\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1537\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1538\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1539\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1540\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1541\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1543\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m   1544\u001b[0m     result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n",
      "Cell \u001b[0;32mIn[3], line 123\u001b[0m, in \u001b[0;36mprox_net.forward\u001b[0;34m(self, x, u)\u001b[0m\n\u001b[1;32m    121\u001b[0m inpt \u001b[38;5;241m=\u001b[39m torch\u001b[38;5;241m.\u001b[39mcat((x,u),dim\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0\u001b[39m)\n\u001b[1;32m    122\u001b[0m out \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mact1(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mconv1(inpt))\n\u001b[0;32m--> 123\u001b[0m out \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mact2(\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mconv2\u001b[49m\u001b[43m(\u001b[49m\u001b[43mout\u001b[49m\u001b[43m)\u001b[49m)\n\u001b[1;32m    124\u001b[0m out \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mconv3(out)\n\u001b[1;32m    126\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m out\n",
      "File \u001b[0;32m~/miniconda3/envs/ia/lib/python3.12/site-packages/torch/nn/modules/module.py:1532\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1530\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_compiled_call_impl(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)  \u001b[38;5;66;03m# type: ignore[misc]\u001b[39;00m\n\u001b[1;32m   1531\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 1532\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_impl\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/ia/lib/python3.12/site-packages/torch/nn/modules/module.py:1541\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1536\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1537\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1538\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1539\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1540\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1541\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1543\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m   1544\u001b[0m     result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/envs/ia/lib/python3.12/site-packages/torch/nn/modules/conv.py:460\u001b[0m, in \u001b[0;36mConv2d.forward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    459\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mforward\u001b[39m(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;28minput\u001b[39m: Tensor) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Tensor:\n\u001b[0;32m--> 460\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_conv_forward\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mweight\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbias\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/ia/lib/python3.12/site-packages/torch/nn/modules/conv.py:456\u001b[0m, in \u001b[0;36mConv2d._conv_forward\u001b[0;34m(self, input, weight, bias)\u001b[0m\n\u001b[1;32m    452\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mpadding_mode \u001b[38;5;241m!=\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mzeros\u001b[39m\u001b[38;5;124m'\u001b[39m:\n\u001b[1;32m    453\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m F\u001b[38;5;241m.\u001b[39mconv2d(F\u001b[38;5;241m.\u001b[39mpad(\u001b[38;5;28minput\u001b[39m, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_reversed_padding_repeated_twice, mode\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mpadding_mode),\n\u001b[1;32m    454\u001b[0m                     weight, bias, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstride,\n\u001b[1;32m    455\u001b[0m                     _pair(\u001b[38;5;241m0\u001b[39m), \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdilation, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mgroups)\n\u001b[0;32m--> 456\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mF\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mconv2d\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mweight\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbias\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mstride\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    457\u001b[0m \u001b[43m                \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpadding\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdilation\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mgroups\u001b[49m\u001b[43m)\u001b[49m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# --- Set up the forward operator (ray transform) in ODL --- #\n",
    "# Reconstruction space: functions on the rectangle [-20, 20]^2\n",
    "img_size = 256 # discretized with 256 samples per dimension\n",
    "reco_space = odl.uniform_discr(min_pt=[-20, -20], max_pt=[20, 20],\n",
    "                               shape=[img_size, img_size], dtype='float32')\n",
    "# Make a parallel beam geometry with flat detector, using number of angles = num_angles\n",
    "num_angles = 30\n",
    "geometry = odl.tomo.parallel_beam_geometry(reco_space, num_angles=num_angles)\n",
    "\n",
    "# Create the forward operator, adjoint operator, and the FBO operator in ODL\n",
    "fwd_op_odl = odl.tomo.RayTransform(reco_space, geometry)\n",
    "fbp_op_odl = odl.tomo.fbp_op(fwd_op_odl,filter_type='Ram-Lak',\n",
    "                             frequency_scaling=0.6)\n",
    "adj_op_odl = fwd_op_odl.adjoint\n",
    "\n",
    "# Create phantom and noisy projection data in ODL\n",
    "phantom_odl = odl.phantom.shepp_logan(reco_space, modified=True) # ground-truth image\n",
    "data_odl = fwd_op_odl(phantom_odl) # sinogram\n",
    "data_odl += odl.phantom.white_noise(fwd_op_odl.range) * np.mean(data_odl) * 0.1 # noisy sinogram\n",
    "\n",
    "# Reconstruct the image using FBP\n",
    "t0 = time()\n",
    "fbp_odl = fbp_op_odl(data_odl)\n",
    "print('FBP reconstruction time = {:.3f} seconds'.format(time()-t0))\n",
    "\n",
    "# convert the image and the sinogram to numpy arrays\n",
    "phantom_np = phantom_odl.__array__()\n",
    "fbp_np = fbp_odl.__array__()\n",
    "data_np = data_odl.__array__()\n",
    "print('sinogram size = {}'.format(data_np.shape))\n",
    "\n",
    "data_range=np.max(phantom_np)-np.min(phantom_np)\n",
    "psnr_fbp = compare_psnr(phantom_np,fbp_np,data_range=data_range)\n",
    "ssim_fbp = compare_ssim(phantom_np,fbp_np,data_range=data_range)\n",
    "\n",
    "# Display Shepp-Logan phantom and noisy sinogram\n",
    "fig, axes = plt.subplots(1, 2, figsize=(12, 6))\n",
    "\n",
    "axes[0].imshow(phantom_np.transpose(), cmap='bone')\n",
    "axes[0].text(0, 1.05, '(a) Shepp-Logan phantom', fontsize=15, transform=axes[0].transAxes)\n",
    "axes[0].axis('off')\n",
    "\n",
    "axes[1].imshow(data_np, cmap='bone')\n",
    "axes[1].text(0, 1.5, '(b) Noisy sinogram', fontsize=15, transform=axes[1].transAxes)\n",
    "axes[1].axis('off')\n",
    "\n",
    "fig.savefig(args.output_dir + '/phantom_sinogram.png', bbox_inches='tight', dpi=300)\n",
    "\n",
    "# Let's solve the TV reconstruction problem using the linearized ADMM algorithm (implemented in ODL).\n",
    "\n",
    "# In this example we solve the optimization problem: min_x  f(x) + g(Lx) = ||A(x) - y||_2^2 + lam * ||grad(x)||_1, Where ``A`` is a parallel beam ray transform, ``grad`` is the spatial gradient and ``y`` given noisy data.\n",
    "\n",
    "# The problem is rewritten in decoupled form as: min_x g(L(x)) with a separable sum ``g`` of functionals and the stacked operator ``L``:\n",
    "\n",
    "#     g(z) = ||z_1 - g||_2^2 + lam * ||z_2||_1,\n",
    "\n",
    "#                ( A(x)    )\n",
    "#     z = L(x) = ( grad(x) ).\n",
    "\n",
    "# Gradient operator for the TV part\n",
    "grad = odl.Gradient(reco_space)\n",
    "\n",
    "# Stacking of the two operators\n",
    "L = odl.BroadcastOperator(fwd_op_odl, grad)\n",
    "\n",
    "# Data matching and regularization functionals\n",
    "data_fit = odl.solvers.L2NormSquared(fwd_op_odl.range).translated(data_odl)\n",
    "lam = 0.015\n",
    "reg_func = lam * odl.solvers.L1Norm(grad.range)\n",
    "g = odl.solvers.SeparableSum(data_fit, reg_func)\n",
    "\n",
    "# We don't use the f functional, setting it to zero\n",
    "f = odl.solvers.ZeroFunctional(L.domain)\n",
    "\n",
    "# --- Select parameters and solve using ADMM --- #\n",
    "\n",
    "# Estimated operator norm, add 10 percent for some safety margin\n",
    "op_norm = 1.1 * odl.power_method_opnorm(L, maxiter=20)\n",
    "\n",
    "niter = 200  # Number of iterations\n",
    "sigma = 2.0  # Step size for g.proximal\n",
    "tau = sigma / op_norm ** 2  # Step size for f.proximal\n",
    "\n",
    "# Optionally pass a callback to the solver to display intermediate results\n",
    "callback = (odl.solvers.CallbackPrintIteration(step=10) &\n",
    "            odl.solvers.CallbackShow(step=10))\n",
    "\n",
    "# Choose a starting point\n",
    "x_admm_odl = L.domain.zero()\n",
    "\n",
    "# Run the algorithm\n",
    "t0 = time()\n",
    "odl.solvers.admm_linearized(x_admm_odl, f, g, L, tau, sigma,\n",
    "                            niter, callback=None)\n",
    "x_admm_np = x_admm_odl.__array__()\n",
    "print('ADMM reconstruction time = {:.3f} seconds'.format(time() - t0))\n",
    "\n",
    "# Compute PSNR and SSIM\n",
    "psnr_tv = compare_psnr(phantom_np,x_admm_np,data_range=data_range)\n",
    "ssim_tv = compare_ssim(phantom_np,x_admm_np,data_range=data_range)\n",
    "\n",
    "# Now, we will \"train\" learned gradient descent (LGD) network on this specific image to show the potential of data-driven reconstruction.\n",
    "# Keep in mind that this example is unrealistic, in the sense that in practice you will train your learned reconstruction network on a dataset containing\n",
    "# many images, and then test the performance on a new test image (that was not a part of your training dataset).\n",
    "\n",
    "### first, let's create a CNN that replaces the prox operator in PGD.\n",
    "class prox_net(nn.Module):\n",
    "    def __init__(self, n_in_channels=2, n_out_channels = 1, n_filters=32, kernel_size=3):\n",
    "        super(prox_net, self).__init__()\n",
    "        self.pad = (kernel_size-1)//2\n",
    "        self.conv1 = nn.Conv2d(n_in_channels, out_channels=n_filters, kernel_size=kernel_size, stride=1, padding=self.pad, bias=True)\n",
    "        self.conv2 = nn.Conv2d(n_filters, n_filters, kernel_size=kernel_size, stride=1, padding=self.pad, bias=True)\n",
    "        self.conv3 = nn.Conv2d(n_filters, out_channels=1, kernel_size=kernel_size, stride=1, padding=self.pad, bias=True)\n",
    "\n",
    "        self.act1 = nn.PReLU(num_parameters=1, init=0.25)\n",
    "        self.act2 = nn.PReLU(num_parameters=1, init=0.25)\n",
    "\n",
    "    def forward(self, x, u):\n",
    "        ''' ################## YOUR CODE HERE #################### '''\n",
    "        ### Note: here the two inputs denote the current iterate and the gradient\n",
    "        inpt = torch.cat((x,u),dim=0)\n",
    "        out = self.act1(self.conv1(inpt))\n",
    "        out = self.act2(self.conv2(out))\n",
    "        out = self.conv3(out)\n",
    "\n",
    "        return out\n",
    "\n",
    "# Let's compute a reasonable initial value for the step-size as step_size = 1/L, where L is the spectral norm of the forward operator.\n",
    "op_norm = 1.1*odl.power_method_opnorm(fwd_op_odl)\n",
    "step_size = 1/op_norm\n",
    "\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "fbp_op = odl_torch.OperatorModule(fbp_op_odl).to(device)\n",
    "fwd_op = odl_torch.OperatorModule(fwd_op_odl).to(device)\n",
    "adj_op = odl_torch.OperatorModule(adj_op_odl).to(device)\n",
    "\n",
    "class LGD_net(nn.Module):\n",
    "    def __init__(self, niter=5,step_size=step_size):\n",
    "        super(LGD_net, self).__init__()\n",
    "        self.niter = niter\n",
    "        self.prox = nn.ModuleList([prox_net().to(device) for i in range(self.niter)])\n",
    "        self.step_size = nn.Parameter(step_size * torch.ones(self.niter).to(device))\n",
    "    def forward(self, y, x_init):\n",
    "        x = x_init\n",
    "        ''' ################## YOUR CODE HERE #################### '''\n",
    "        #### Note: the gradient at a given x is A^T(Ax-y).\n",
    "        for k in range(self.niter):\n",
    "            f_grad = adj_op(fwd_op(x) - y)\n",
    "            x = x - self.step_size[k] * self.prox[k](x,f_grad)\n",
    "\n",
    "        return x\n",
    "\n",
    "# Now, we can use the LGD network to reconstruct the image.\n",
    "lgd_net = LGD_net().to(device) # realize the network and export it to GPU\n",
    "num_learnable_params = sum(p.numel() for p in lgd_net.parameters() if p.requires_grad)\n",
    "print('number of model parameters = {}'.format(num_learnable_params))\n",
    "y = torch.from_numpy(data_np).to(device).unsqueeze(0) # noisy sinogram data as a torch tensor\n",
    "\n",
    "# from odl.tomo import fbp_op\n",
    "# initialization for the LGD net. Note that the input to a torch 2D CNN must be of size (num_batches x height x width).\n",
    "# Initialise y as a RayTransform object\n",
    "\n",
    "x_init = fbp_op(y)\n",
    "\n",
    "# print(x_init.shape, y.shape)\n",
    "\n",
    "ground_truth = torch.from_numpy(phantom_np).to(device).unsqueeze(0) # target ground-truth as a torch tensor\n",
    "\n",
    "# define the loss and the optimizer\n",
    "mse_loss = torch.nn.MSELoss()\n",
    "optimizer = torch.optim.Adam(lgd_net.parameters(),lr = 1e-4)\n",
    "num_epochs = 2000\n",
    "\n",
    "##### training loop\n",
    "t0 = time()\n",
    "losses = []\n",
    "print('Training LGD network...')\n",
    "for epoch in range(0, num_epochs):\n",
    "    optimizer.zero_grad()\n",
    "    '''################## YOUR CODE HERE #################### '''\n",
    "    #### You need to compute the reconstructed image by applying a forward pass through lgd_net, compute the loss, call the backward function on loss, and update the parameters.\n",
    "    recon = lgd_net(y, x_init)\n",
    "    loss = mse_loss(recon, ground_truth)\n",
    "    loss.backward()\n",
    "    optimizer.step()\n",
    "    losses.append(loss.item())\n",
    "\n",
    "    if(epoch%100 == 0):\n",
    "        print('epoch = {}, loss = {}'.format(epoch, loss.item()))\n",
    "print('Training time = {:.3f} seconds'.format(time()-t0))\n",
    "\n",
    "# Time inference\n",
    "t0 = time()\n",
    "recon = lgd_net(y, x_init)\n",
    "print('LGD Inference time = {:.3f} seconds'.format(time()-t0))\n",
    "\n",
    "lgd_recon_np = recon.detach().cpu().numpy().squeeze() # convert the LGD reconstruction to numpy format\n",
    "\n",
    "psnr_lgd = compare_psnr(phantom_np,lgd_recon_np,data_range=data_range)\n",
    "ssim_lgd = compare_ssim(phantom_np,lgd_recon_np,data_range=data_range)\n",
    "\n",
    "# Plot the reconstructions\n",
    "fig, axes = plt.subplots(1, 4, figsize=(12, 6))\n",
    "\n",
    "axes[0].imshow(phantom_np.transpose(), cmap='bone')\n",
    "axes[0].text(0, 1.05, '(a) Ground-truth', fontsize=15, transform=axes[0].transAxes)\n",
    "\n",
    "axes[1].imshow(fbp_np.transpose(), cmap='bone')\n",
    "axes[1].text(0, 1.05, '(b) FBP', fontsize=15, transform=axes[1].transAxes)\n",
    "axes[1].set_xlabel('PSNR: {:.2f} dB, SSIM: {:.2f}'.format(psnr_fbp, ssim_fbp))\n",
    "\n",
    "axes[2].imshow(x_admm_np.transpose(), cmap='bone')\n",
    "axes[2].text(0, 1.05, '(c) TV', fontsize=15, transform=axes[2].transAxes)\n",
    "axes[2].set_xlabel('PSNR: {:.2f} dB, SSIM: {:.2f}'.format(psnr_tv, ssim_tv))\n",
    "\n",
    "axes[3].imshow(lgd_recon_np.transpose(), cmap='bone')\n",
    "axes[3].text(0, 1.05, '(d) LGD', fontsize=15, transform=axes[3].transAxes)\n",
    "axes[3].set_xlabel('PSNR: {:.2f} dB, SSIM: {:.2f}'.format(psnr_lgd, ssim_lgd))\n",
    "\n",
    "for ax in axes:\n",
    "    ax.set_xticks([])\n",
    "    ax.set_yticks([])\n",
    "\n",
    "fig.savefig(args.output_dir + '/ct_reconstructions.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ia_6",
   "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": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}