demo_splitting_loss.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Self-supervised learning with measurement splitting\n",
    "\n",
    "We demonstrate self-supervised learning with measurement splitting, to train a denoiser network on the MNIST dataset.\n",
    "\n",
    "Measurement splitting constructs a ground-truth free loss $\\frac{m}{m_2}\\| y_2 - A_2 \\inversef{y_1}{A_1}\\|^2$ by splitting the measurement and the forward operator using a randomly generated mask.\n",
    "\n",
    "See :class:`deepinv.loss.SplittingLoss` for full details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\s2558406\\Documents\\Repos\\deepinv\\venv\\Lib\\site-packages\\tqdm\\auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import deepinv as dinv\n",
    "from torch.utils.data import DataLoader\n",
    "import torch\n",
    "from torchvision import transforms, datasets\n",
    "from deepinv.models.utils import get_weights_url"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.manual_seed(0)\n",
    "device = dinv.utils.get_freer_gpu() if torch.cuda.is_available() else \"cpu\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define loss\n",
    "\n",
    "Our implementation has multiple optional parameters that control how the splitting is to be achieved. For example, you can:\n",
    "\n",
    "- Use `split_ratio` to set the ratio of pixels used in the forward pass vs the loss;\n",
    "- Define custom masking methods using a `mask_generator` such as :class:`deepinv.physics.generator.BernoulliSplittingMaskGenerator`;\n",
    "- Use `MC_samples` to set how many realisations of the random mask is used at evaluation time;\n",
    "- Optionally disable measurement splitting at evaluation time using `eval_split_input`.\n",
    "- Average over both input and output masks at evaluation time using `eval_split_output`.\n",
    "\n",
    "Note that after the model has been defined, the loss must also \"adapt\" the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss = dinv.loss.SplittingLoss(split_ratio=0.6, eval_split_input=True, eval_split_output=True, MC_samples=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prepare data\n",
    "\n",
    "We use the `torchvision` MNIST dataset, and use Poisson noise physics for the forward operator.\n",
    "\n",
    ".. note::\n",
    "\n",
    "      We use a subset of the whole training set to reduce the computational load of the example.\n",
    "      We recommend to use the whole set by setting ``train_datapoints=test_datapoints=None`` to get the best results.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "transform = transforms.Compose([transforms.ToTensor()])\n",
    "\n",
    "train_dataset = datasets.MNIST(\n",
    "    root=\".\", train=True, transform=transform, download=True\n",
    ")\n",
    "test_dataset = datasets.MNIST(\n",
    "    root=\".\", train=False, transform=transform, download=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset has been saved in MNIST\n"
     ]
    }
   ],
   "source": [
    "physics = dinv.physics.Denoising(dinv.physics.PoissonNoise(0.1))\n",
    "\n",
    "deepinv_datasets_path = dinv.datasets.generate_dataset(\n",
    "    train_dataset=train_dataset,\n",
    "    test_dataset=test_dataset,\n",
    "    physics=physics,\n",
    "    device=device,\n",
    "    save_dir=\"MNIST\",\n",
    "    train_datapoints=100,\n",
    "    test_datapoints=10,\n",
    ")\n",
    "\n",
    "train_dataset = dinv.datasets.HDF5Dataset(path=deepinv_datasets_path, train=True)\n",
    "test_dataset = dinv.datasets.HDF5Dataset(path=deepinv_datasets_path, train=False)\n",
    "\n",
    "train_dataloader = DataLoader(train_dataset, shuffle=True)\n",
    "test_dataloader = DataLoader(test_dataset,  shuffle=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define model\n",
    "\n",
    "We use a simple U-Net architecture with 2 scales as the denoiser network. \n",
    "\n",
    "To reduce training time, we use a pretrained model. Here we demonstrate training with 100 images for 1 epoch, after having loaded a pretrained model trained that was with 1000 images for 100 epochs.\n",
    "\n",
    ".. note::\n",
    "\n",
    "      When using the splitting loss, the model must be \"adapted\" by the loss, as its forward pass takes only a subset of the pixels, not the full image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = dinv.models.ArtifactRemoval(\n",
    "    dinv.models.UNet(in_channels=1, out_channels=1, scales=2).to(device)\n",
    ")\n",
    "model = loss.adapt_model(model)\n",
    "\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=1e-3, weight_decay=1e-8)\n",
    "\n",
    "# Load pretrained model\n",
    "file_name = \"demo_measplit_mnist_denoising.pth\"\n",
    "url = get_weights_url(model_name=\"measplit\", file_name=file_name)\n",
    "ckpt = torch.hub.load_state_dict_from_url(\n",
    "    url, map_location=lambda storage, loc: storage, file_name=file_name\n",
    ")\n",
    "\n",
    "model.load_state_dict(ckpt[\"state_dict\"])\n",
    "optimizer.load_state_dict(ckpt[\"optimizer\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train and test network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The model has 444737 trainable parameters\n",
      "Train epoch 0: TotalLoss=0.017, PSNR=11.59\n"
     ]
    }
   ],
   "source": [
    "trainer = dinv.Trainer(\n",
    "    model=model,\n",
    "    physics=physics,\n",
    "    epochs=1,\n",
    "    losses=loss,\n",
    "    optimizer=optimizer,\n",
    "    device=device,\n",
    "    train_dataloader=train_dataloader,\n",
    "    plot_images=False,\n",
    "    save_path=None,\n",
    "    verbose=True,\n",
    "    show_progress_bar=False,\n",
    "    wandb_vis=False,\n",
    ")\n",
    "\n",
    "model = trainer.train()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test and visualise the model outputs using a small test set. We set the output to average over 50 iterations of random mask realisations. The trained model improves on the no-learning reconstruction by ~3dB."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x200 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test PSNR: No learning rec.: 19.356+-1.523 | Model: 23.150+-1.996. \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(23.150262069702148,\n",
       " 1.9961603806415074,\n",
       " 19.355849266052246,\n",
       " 1.5233685706910973)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trainer.plot_images = True\n",
    "model.MC_samples = 50\n",
    "trainer.test(test_dataloader)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since this is a denoising example, above, we have set `eval_split_output` to True (see :class:`deepinv.loss.SplittingLoss` for details). Alternatively, we get worse results when we set `eval_split_output` to False:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x200 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test PSNR: No learning rec.: 19.356+-1.523 | Model: 14.441+-1.520. \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(14.44115686416626, 1.5199918435073472, 19.355849266052246, 1.5233685706910973)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.eval_split_output = False\n",
    "trainer.test(test_dataloader)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Furthermore, we can disable measurement splitting at evaluation altogether by setting `eval_split_input` to False (this is done in [SSDU](https://pubmed.ncbi.nlm.nih.gov/32614100/)):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAn8AAAC6CAYAAADf9CvVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAASIElEQVR4nO3dXYicV/048LN5EQsiswktJL5mFqXeiJ1prlSo7Kz4srGCu6mIl2YGKXpld11BkvQm7IKCFyqzey002b2pLSW6G6FeKGgn4BtqcaZCbVMQsuMLUmyz87uQnX/2/5zZPNPZ2bfz+UCh+805z3OyOXnyfc6c756RTqfTCQAAJOHIXg8AAIDdI/kDAEiI5A8AICGSPwCAhEj+AAASIvkDAEiI5A8AICGSPwCAhEj+tjE2NhZWVlb2ehgcEOVyOYyMjIR2u92N3bx5M5TL5b6vNTIysoMji2u325nxAnD4Sf72UKvV2ushsMMqlUqYnZ3d62HkUigUQqPRCIVCYa+HwpCVy+UwNjYWJiYmwsTERCiXy15s2Zfa7XaYnp4OY2NjYXR0NJTL5bC2thZCCGFhYSGMjY11/xsdHd3y9c2bN/d49AeH5G8Pzc7OmqyHzOzsbLh27dqBWU0rlUp7PQR2yfz8fFhdXQ2rq6vhxo0bnj/sO61WK5w5cyZMTEyEZrMZ1tfXw9LSUvfXZ2ZmQrPZDM1mMywvL4d2u939utlsep71QfLXh9HR0bCwsNB9i7575W5sbCzMzs5230bufqje/RHewsJCqNVqoVarhZWVlTA9Pf2WPhZkfzpx4kSoVqs9V/9WVlZCuVwO5XI51Gq1XNdcWVnpvtnevVqz+XY8NjbWfTMOIXS/LpfLodVqbTtv756b27VbXFzsjntzzh6UBJesQqEQKpXKlnkDe212djZUq9VQrVa7sVKpFCqVyh6O6nCS/PWh3W6HYrEYGo1GKJVKW/4hbrVa3beVubm5MD09ve216vV6qFQqYXl5OTQajWEPnV00NzcXXf27efNmuHLlSrhx40b349Z7fUR88+bNMDs7GxqNRmg2m6Fer3d/rVardd+A755vrVYrzM/Phxs3boRisbjtvL1br3btdrs7hqWlpdBut31cfMCtra2Fa9euhampqRBC7xeMtbW1bvzuhL/XS0yvF4hWq9WN+XiOXlZWVnK/FDMYyV+fNt9Azp49G5rNZvTXZmZmQqvVsjKSqEKhEKrVarhy5cqW+NWrV8Njjz3WTZrm5ubC4uLitte6evVqCOF/q3wTExOh1Wp1V2s251upVArtdnvLfKvX61uSs+3m7d16tTtx4kT3/2/fvr3tmNmfLly4EMrlchgdHQ3z8/PhpZdeCsVisecLRqvVCtPT09348vJyKBQK277E9HqBWFlZCZVKxcdz9LT5olAsFvd4JGk4ttcDOGj6We2wMpKuubm57t6VQZw8eTJMTU2F+fn5zK8tLCyEX//619F+//8DNO9cjLUrFAqhUCh0tyfcvQeHg2NpaSlMTU2FtbW1UKvVun/Wd79ghBC6Lxg3b94M58+f77bbnFOxl5gzZ85052jsBaJUKoXp6enQarVCrVbzMR4Zm/Or1Wp1/3/zmVOpVKLPQN46K387aPOjjIWFhcyb7eaqzOrqajdWKBSsEB5ShUIhnD9/fssD67HHHgv1er375z07OxvOnz+/7XWmpqbCyspKt8/mfFlZWQmrq6theXl56MnY3as5mys6HFyVSiUUi8WwsLAQQvh/LxibxSDNZnOg5Cz2AlGpVEKj0Qhnz54N09PTKo2JqlQqW7a2NBqNUKvV/GSMIZD87aB6vR7K5XK4evVqWF5e7sar1WoYHx8P09PToVgsdh+OExMT4cKFC2F8fFwCeAjNz89v2VBfKpXC/Px8d+9TCGHLgy6mWCyG+fn5MD4+vmV/VaVS6e6junDhwlA/KikUCmFlZSWMjIx0f/SCPVsHW71eD1euXAntdrvnC8bU1NSWvaub8bfyErO5mjMzMxPm5uZ6rliTtnq9HhYXF++5HYYd0GFH+FZyWM3MzHTm5+e7X6+urnaKxeIejoh+lUqlzvLy8pZYtVrtVKvVTqfT6SwvL3dKpVKnVCp1KpVKZ319vRsvFoudYrHYqVQqnWazmYlvXqPT2focnJ+f33L92HVKpVJndXV1aL9vDp719fXO1NRUd77E5m6n0+k0Gg3/7g5gpNPpdPY4/zwURkZGgm8lh9Hs7Gw4efJkmJmZCSFsrWwH4OCR/O0QyR+HWa1WC2tra90tC0tLS/b+ARxQkj8AgIQo+AAASIjkDwAgIZI/AICESP4AABIi+QMASEjus31HRkaGOQ4StV2xuTnHMJhz7DZzjt12rx/kYuUPACAhkj8AgIRI/gAAEpJ7zx8AsL0jR6ypsP+ZpQAACZH8AQAkRPIHAJAQyR8AQEIkfwAACVHtCwA7ZGNjY6+HAPdk5Q8AICGSPwCAhEj+AAASIvkDAEiIgg8A6NPIyEg03ul0dnkk0D8rfwAACZH8AQAkRPIHAJAQyR8AQEIkfwAACZH8AQAkRPIHAJAQyR8AQEIkfwAACZH8AQAkxPFuAEPU6xiwQdvG7ObRYrF7HTkSX0/Y2NgY9nB2nWPcOMis/AEAJETyBwCQEMkfAEBCJH8AAAlR8AEwRL0KA2LFEUePHs3Ejh2LP6bfeOONTKxXYcWgxQl5C1GGVQQRu38sthP3V8hBCqz8AQAkRPIHAJAQyR8AQEIkfwAACTkwBR/lcjkTu3TpUrRto9HIfd16vZ7rXs8++2zua3I4mHMM09ve9rZM7NFHH83EnnrqqdzXbLfb0fh3vvOdTCxWMLK2thbt/+KLL+bq36sw5Pjx42851s+9/vGPf+Tu3+s0ktQKPqampjKxCxcuRNu++uqrmdjrr78ebfujH/0oE3vttdcysb/85S/3GiJDYOUPACAhkj8AgIRI/gAAEiL5AwBIiOQPACAhI52cpU15j/cZlmeeeSYTm5yczN2/VzVmrMoyplarRePnzp3Lda9eYz19+nQmdurUqUzs8uXL0f6x68buv7i4GO3fT5XqMGw3/cw5c24Yhjnn8h5DFkII73znOzOx9fX1ge6fuk9/+tPR+PXr1zOxXtW+Mb2OzctrPz/nWq1WJvb+979/KPf617/+lYn94Q9/GMq9huFvf/tbJrawsBBt+8ILLwx7ONu6V2pn5Q8AICGSPwCAhEj+AAASIvkDAEjIgSn4iOln8/2tW7ei8djm+2q1mqtdCPFN8bH+sWNxel23n2O9Yv1jm/d7FQ/02pS/W/bzRugYc86c204/BR9vf/vbM7EPfOADmdjXv/71aP8HHnggE3vppZeibb/4xS9mYvfff3+07UH24x//OBqPHZu3m/bzc258fDwT+/CHPxxt+8c//jET+9CHPhRtWyqVMrFHHnkkE3vXu94V7f/yyy9nYu95z3uibfN68803M7G///3v0baxZ1rMd7/73Wj8G9/4Rv6BDYGCDwAAuiR/AAAJkfwBACRE8gcAkBDJHwBAQg50te9uunTpUjQeOwIspteRVnmrNHtVY8b++GL3evjhh+81xD2xn6vg9po5Nxy7Xe3b635Hjx7N1b/X0WKx6/Ya/+joaCb24IMPZmJf+cpXov1/+9vfZmInTpzIxF555ZVo/89//vOZ2PPPP5+JffKTn4z2j1WJ5r1PCCE8/fTTufoPi+fc/8Tm4Uc+8pFo29gz5ezZswPd//XXX8/EXnzxxWjbWGVzbM4//vjj0f4//OEP+xzdzlLtCwBAl+QPACAhkj8AgIRI/gAAEqLg44Dotfn/4sWLufrv1z8/G6H3L3Nu9w2j4CSEEI4cyb7nx2K9iktix2LFClZ6ibW97777MrHYJvsQ8h+11avda6+9lqv/sOznOZe6L3zhC9H4tWvXMrHf//73mdgnPvGJaP/bt28PNrABKfgAAKBL8gcAkBDJHwBAQiR/AAAJObbXAyArtmk5dipDCCFcvnw5E3vhhRd2fEwcbubc8PTa0B/bkN3PqR15rxlCCHfu3MkVy1n/17N/ryKQWNuxsbFMLG9hRwgh/OQnP8nE9rqwg/3tgQceyMR+8IMfRNvGCqKefPLJTGyvCzveKit/AAAJkfwBACRE8gcAkBDJHwBAQiR/AAAJUe27D5XL5UysVxXc5ORkJlav13d8TBxu5tzw9KqgjVXx5q0A7qWfyuC89+9HrEIyhHi17/Ly8kD3+uY3v5m77U5UTHPwPf7445nY/fffH227vr6eif35z3/e8THtFSt/AAAJkfwBACRE8gcAkBDJHwBAQkY6OXe39rNhlvxiG+37OSrr9OnTmditW7cGGtNu2m76mXPDYc7t3zm3m4UJseKMjY2Nga557Fi8hnBiYiITe+6553JfN3ak4KVLl3L3H0ZxSz/285w7rD760Y9mYj/72c8ysePHj0f7P/LII5nYz3/+84HHtVvuNb+t/AEAJETyBwCQEMkfAEBCJH8AAAlxwscei52i0Gg0csVCOFgb7dkfzLn9azeLEAYt7uinUOH73//+QPf63ve+N1B/0vOZz3wmE4sVd9y4cSPa/5e//OWOj2k/sfIHAJAQyR8AQEIkfwAACZH8AQAkRPIHAJAQ1b57rFarZWKxasp+jjKC7Zhz7IRYte+DDz4YbXvmzJlc1/zqV78aja+vr+cfGEm57777ovFPfepTmdh///vfTOzixYvR/m+88cZgA9vnrPwBACRE8gcAkBDJHwBAQiR/AAAJUfCxS6rVajQ+OTmZiZ07dy4Tc6QW/TLn2ClHjmTXCY4ePZqJ/e53vxvoPktLSwP17+fIuV5td/OIPQb3xBNPROMPPfRQJnb9+vVM7Be/+MWOj+kgsPIHAJAQyR8AQEIkfwAACZH8AQAkRMHHEJw6dSoTq9fr0bbPPvtsJvbwww/nagebzDl2Qqywo1f8y1/+8kD3evrppzOxO3fuDHRNDrfPfvazmdi3v/3taNt//vOfmdiTTz6542M6qKz8AQAkRPIHAJAQyR8AQEIkfwAACZH8AQAkZKST8yybfo7NSV0/xwPFKipjR20dVtt9r8y5/My5/My5/4lV8Pb6/b/73e/OxP76178OdP9Bj2LrZ87Hfq8bGxu5+w/KnOvfyZMnM7Ff/epXmdiZM2ei/Z966qlM7Etf+tLgAzsg7vX3w8ofAEBCJH8AAAmR/AEAJETyBwCQEMe7DeiZZ57J1a5Wqw3UHzaZc/QrVvAQix07Fv8nYdDijrNnzw7Uv5/iDg6Wo0ePRuPXr1/PxGLFHc1mM9q/17Fv/I+VPwCAhEj+AAASIvkDAEiI5A8AICFO+MhpcnIyGs+7eT52qkIIaZ2sEOMn3/dmzg3HYZ5zvcYfK+SItX3uueei/cfHx4cyrrzt8hZ8DNp/WA7znBvUBz/4wWj8T3/6U67+jz76aDSeemGbEz4AAOiS/AEAJETyBwCQEMkfAEBCJH8AAAlxvFtEtVrNxOr1eu7+ly9fzsQuXbo0yJA45Mw5dkI/laMXL17MxAat6j1+/HjutrGx7nVVLsP1vve9LxP76U9/mrv/E088kYn1+qkGbM/KHwBAQiR/AAAJkfwBACRE8gcAkBAFHwOKbTbtZ6M+9MucO7yOHMm+j29sbOTu38/xbt/61rfyDyzic5/7XCb25ptv5u4/aHFH6seiHUSxwrb3vve9ufs///zzmZgiobfGyh8AQEIkfwAACZH8AQAkRPIHAJAQyR8AQEKSr/adnJzMxGKVk4uLi9H+seqlW7duDT4wDi1zjl76qeyNVQb3qoCdmpp6y2Pq5eWXX97xa/aSt7JX5ef+8bGPfSwT+9rXvrYHIyHGyh8AQEIkfwAACZH8AQAkRPIHAJCQ5As+Go1Grnbnzp3Lfc1Tp05lYjbks8mco5dYYUOvIoZYvFdhRD9HaOX173//e8ev2Q/FHfvbxz/+8UzsHe94R+7+zWYzE9vrOXeYWPkDAEiI5A8AICGSPwCAhEj+AAASknzBx+nTpzOxvBvye/WH7Zhzh8Ogp070U9wxyP1DCOGVV17J3TbmoYceysReffXV3P1jY43Fep1wMozijl7fv7z36uf7T2+/+c1vovHx8fFM7Pbt28MeTjKs/AEAJETyBwCQEMkfAEBCJH8AAAmR/AEAJGSkk7O0SWUTw7Dd9DPnGIZ+51w/83DQat1Y/yNH8r+jHz9+PHf8P//5T+7+d+7cycRilbn9HEV30PVTLew5x2671985K38AAAmR/AEAJETyBwCQEMkfAEBCchd8AABw8Fn5AwBIiOQPACAhkj8AgIRI/gAAEiL5AwBIiOQPACAhkj8AgIRI/gAAEiL5AwBIyP8BazPu/dSWC3YAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 800x200 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test PSNR: No learning rec.: 19.356+-1.523 | Model: 9.650+-1.670. \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(9.650477170944214, 1.6700495788637173, 19.355849266052246, 1.5233685706910973)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.eval_split_input = False\n",
    "trainer.test(test_dataloader)"
   ]
  }
 ],
 "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}