examples/paper-examples/2-single_particle_tracking.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:53.875636Z",
          "iopub.status.busy": "2022-06-30T10:52:53.875636Z",
          "iopub.status.idle": "2022-06-30T10:52:55.884639Z",
          "shell.execute_reply": "2022-06-30T10:52:55.884140Z"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: deeptrack in c:\\users\\gu\\deeptrack\\deeptrack-2.0 (1.2.1)\n",
            "Requirement already satisfied: tensorflow in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (2.9.1)\n",
            "Requirement already satisfied: tensorflow-probability in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (0.17.0)\n",
            "Requirement already satisfied: numpy in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (1.23.0)\n",
            "Requirement already satisfied: scipy in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (1.8.1)\n",
            "Requirement already satisfied: pint in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (0.19.2)\n",
            "Requirement already satisfied: pandas in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (1.4.3)\n",
            "Requirement already satisfied: tqdm in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (4.64.0)\n",
            "Requirement already satisfied: scikit-image>=0.18.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (0.19.3)\n",
            "Requirement already satisfied: pydeepimagej in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (1.1.0)\n",
            "Requirement already satisfied: more_itertools in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (8.13.0)\n",
            "Requirement already satisfied: tensorflow_addons in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from deeptrack) (0.17.1)\n",
            "Requirement already satisfied: imageio>=2.4.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-image>=0.18.0->deeptrack) (2.19.3)\n",
            "Requirement already satisfied: PyWavelets>=1.1.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-image>=0.18.0->deeptrack) (1.3.0)\n",
            "Requirement already satisfied: packaging>=20.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-image>=0.18.0->deeptrack) (21.3)\n",
            "Requirement already satisfied: pillow!=7.1.0,!=7.1.1,!=8.3.0,>=6.1.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-image>=0.18.0->deeptrack) (9.1.1)\n",
            "Requirement already satisfied: tifffile>=2019.7.26 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-image>=0.18.0->deeptrack) (2022.5.4)\n",
            "Requirement already satisfied: networkx>=2.2 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-image>=0.18.0->deeptrack) (2.8.4)\n",
            "Requirement already satisfied: pytz>=2020.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from pandas->deeptrack) (2022.1)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from pandas->deeptrack) (2.8.2)\n",
            "Requirement already satisfied: keras<2.10.0,>=2.9.0rc0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (2.9.0)\n",
            "Requirement already satisfied: protobuf<3.20,>=3.9.2 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (3.19.4)\n",
            "Requirement already satisfied: google-pasta>=0.1.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (0.2.0)\n",
            "Requirement already satisfied: six>=1.12.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (1.16.0)\n",
            "Requirement already satisfied: keras-preprocessing>=1.1.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (1.1.2)\n",
            "Requirement already satisfied: tensorflow-estimator<2.10.0,>=2.9.0rc0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (2.9.0)\n",
            "Requirement already satisfied: absl-py>=1.0.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (1.1.0)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (3.3.0)\n",
            "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (0.26.0)\n",
            "Requirement already satisfied: libclang>=13.0.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (14.0.1)\n",
            "Requirement already satisfied: grpcio<2.0,>=1.24.3 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (1.47.0)\n",
            "Requirement already satisfied: flatbuffers<2,>=1.12 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (1.12)\n",
            "Requirement already satisfied: gast<=0.4.0,>=0.2.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (0.4.0)\n",
            "Requirement already satisfied: wrapt>=1.11.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (1.14.1)\n",
            "Requirement already satisfied: termcolor>=1.1.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (1.1.0)\n",
            "Requirement already satisfied: astunparse>=1.6.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (1.6.3)\n",
            "Requirement already satisfied: setuptools in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (58.1.0)\n",
            "Requirement already satisfied: h5py>=2.9.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (3.7.0)\n",
            "Requirement already satisfied: typing-extensions>=3.6.6 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (4.2.0)\n",
            "Requirement already satisfied: tensorboard<2.10,>=2.9 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow->deeptrack) (2.9.1)\n",
            "Requirement already satisfied: typeguard>=2.7 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow_addons->deeptrack) (2.13.3)\n",
            "Requirement already satisfied: decorator in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow-probability->deeptrack) (5.1.1)\n",
            "Requirement already satisfied: cloudpickle>=1.3 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow-probability->deeptrack) (2.1.0)\n",
            "Requirement already satisfied: dm-tree in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorflow-probability->deeptrack) (0.1.7)\n",
            "Requirement already satisfied: colorama in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tqdm->deeptrack) (0.4.5)\n",
            "Requirement already satisfied: wheel<1.0,>=0.23.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from astunparse>=1.6.0->tensorflow->deeptrack) (0.37.1)\n",
            "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from packaging>=20.0->scikit-image>=0.18.0->deeptrack) (3.0.9)\n",
            "Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorboard<2.10,>=2.9->tensorflow->deeptrack) (0.4.6)\n",
            "Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorboard<2.10,>=2.9->tensorflow->deeptrack) (0.6.1)\n",
            "Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorboard<2.10,>=2.9->tensorflow->deeptrack) (1.8.1)\n",
            "Requirement already satisfied: requests<3,>=2.21.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorboard<2.10,>=2.9->tensorflow->deeptrack) (2.28.0)\n",
            "Requirement already satisfied: werkzeug>=1.0.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorboard<2.10,>=2.9->tensorflow->deeptrack) (2.1.2)\n",
            "Requirement already satisfied: markdown>=2.6.8 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorboard<2.10,>=2.9->tensorflow->deeptrack) (3.3.7)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from tensorboard<2.10,>=2.9->tensorflow->deeptrack) (2.9.0)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (0.2.8)\n",
            "Requirement already satisfied: cachetools<6.0,>=2.0.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (5.2.0)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (4.8)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (1.3.1)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (1.26.9)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (2.0.12)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (2022.6.15)\n",
            "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (3.3)\n",
            "Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (0.4.8)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in c:\\users\\gu\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.10,>=2.9->tensorflow->deeptrack) (3.2.0)\n"
          ]
        }
      ],
      "source": [
        "%matplotlib inline\n",
        "!pip install deeptrack"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "<!--<badge>--><a href=\"https://colab.research.google.com/github/softmatterlab/DeepTrack-2.0/blob/develop/examples/paper-examples/2-single_particle_tracking.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a><!--</badge>-->"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Example 2. Single particle tracking\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## 1. Setup\n",
        "\n",
        "Imports the objects needed for this example."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:55.901639Z",
          "iopub.status.busy": "2022-06-30T10:52:55.901639Z",
          "iopub.status.idle": "2022-06-30T10:52:59.013433Z",
          "shell.execute_reply": "2022-06-30T10:52:59.012934Z"
        }
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "c:\\users\\gu\\deeptrack\\deeptrack-2.0\\deeptrack\\backend\\_config.py:11: UserWarning: cupy not installed. GPU-accelerated simulations will not be possible\n",
            "  warnings.warn(\n",
            "c:\\users\\gu\\deeptrack\\deeptrack-2.0\\deeptrack\\backend\\_config.py:25: UserWarning: cupy not installed, CPU acceleration not enabled\n",
            "  warnings.warn(\"cupy not installed, CPU acceleration not enabled\")\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Dataset already downloaded.\n"
          ]
        }
      ],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import deeptrack as dt\n",
        "import deeptrack.extras\n",
        "deeptrack.extras.datasets.load(\"ParticleTracking\")\n",
        "\n",
        "IMAGE_SIZE = 51"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## 2. Defining the dataset\n",
        "\n",
        "### 2.1 Defining the training set\n",
        "\n",
        "The training set consists of simulated 51 by 51 pixel images, containing a single particle each. The particles are simulated as spheres with a radius between 1 micron and 2 microns, and a refractive index between 1.5 and 1.6. Its position in the camera plane is constrained to be within the image, and is sampled with a normal distribution with standard deviation of 5 pixel units in along the axis normal to the camera plane. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:59.015933Z",
          "iopub.status.busy": "2022-06-30T10:52:59.015933Z",
          "iopub.status.idle": "2022-06-30T10:52:59.019933Z",
          "shell.execute_reply": "2022-06-30T10:52:59.019446Z"
        }
      },
      "outputs": [],
      "source": [
        "particle = dt.MieSphere(\n",
        "    position=lambda: np.random.uniform(IMAGE_SIZE / 2 - 3, IMAGE_SIZE / 2 + 3, 2) * dt.units.pixel,\n",
        "    z=lambda: -np.random.uniform(0, 5) * dt.units.pixel,\n",
        "    radius=lambda: np.random.uniform(300, 500) * 1e-9,\n",
        "    refractive_index=lambda: np.random.uniform(1.37, 1.42),\n",
        "    position_unit=\"pixel\",\n",
        "    L=10\n",
        ")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "The particle is imaged using a brightfield microscope with NA between 0.15 and 0.25 and a illuminating laser wavelength between 400 and 700 nm. To simulate the broad spectrum we define 10 individual optical devices, each imaging the particle at a single wavelength. The result is then averaged."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:59.021934Z",
          "iopub.status.busy": "2022-06-30T10:52:59.021934Z",
          "iopub.status.idle": "2022-06-30T10:52:59.027443Z",
          "shell.execute_reply": "2022-06-30T10:52:59.027443Z"
        }
      },
      "outputs": [],
      "source": [
        "\n",
        "spectrum = np.linspace(500e-9, 700e-9, 5)\n",
        "\n",
        "imaged_particle_list = []\n",
        "\n",
        "\n",
        "for wavelength in spectrum:\n",
        "    single_wavelength_optics = dt.Brightfield(\n",
        "        NA=0.8,\n",
        "        resolution=1e-6,\n",
        "        magnification=15,\n",
        "        wavelength=wavelength,\n",
        "        padding=(32, 32, 32, 32),\n",
        "        output_region=(0, 0, IMAGE_SIZE, IMAGE_SIZE),\n",
        "    )\n",
        "    \n",
        "    imaged_particle_list.append(\n",
        "        single_wavelength_optics(particle)\n",
        "    )\n",
        "\n",
        "dataset = sum(imaged_particle_list) / len(imaged_particle_list)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### 2.2 Defining the training label\n",
        "\n",
        "The training label is extracted directly from the image as the `position` property divided by the image size, such that the posible values are contained within -0.5 and 0.5."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:59.029934Z",
          "iopub.status.busy": "2022-06-30T10:52:59.029934Z",
          "iopub.status.idle": "2022-06-30T10:52:59.032944Z",
          "shell.execute_reply": "2022-06-30T10:52:59.032944Z"
        }
      },
      "outputs": [],
      "source": [
        "def get_label(image):\n",
        "    px = np.array(image.get_property(\"position\")) / IMAGE_SIZE - 0.5\n",
        "    return px"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### 2.3 Visualizing the dataset\n",
        "\n",
        "We resolve and show 16 images, with a green circle indicating the particle position."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:59.035434Z",
          "iopub.status.busy": "2022-06-30T10:52:59.035434Z",
          "iopub.status.idle": "2022-06-30T10:52:59.725934Z",
          "shell.execute_reply": "2022-06-30T10:52:59.725463Z"
        },
        "scrolled": true
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "c:\\users\\gu\\deeptrack\\deeptrack-2.0\\deeptrack\\optics.py:207: UserWarning: Likely bad optical parameters. NA / wavelength * resolution * magnification = 24.0 should be at most 0.5\n",
            "To fix, set magnification to 48.0, and downsample the resulting image with dt.AveragePooling((48.0, 48.0, 1))\n",
            "\n",
            "  warnings.warn(\n",
            "c:\\users\\gu\\deeptrack\\deeptrack-2.0\\deeptrack\\optics.py:276: RuntimeWarning: invalid value encountered in sqrt\n",
            "  * np.sqrt(1 - (NA / refractive_index_medium) ** 2 * RHO),\n",
            "c:\\users\\gu\\deeptrack\\deeptrack-2.0\\deeptrack\\optics.py:207: UserWarning: Likely bad optical parameters. NA / wavelength * resolution * magnification = 21.81818181818182 should be at most 0.5\n",
            "To fix, set magnification to 44.0, and downsample the resulting image with dt.AveragePooling((44.0, 44.0, 1))\n",
            "\n",
            "  warnings.warn(\n",
            "c:\\users\\gu\\deeptrack\\deeptrack-2.0\\deeptrack\\optics.py:207: UserWarning: Likely bad optical parameters. NA / wavelength * resolution * magnification = 20.000000000000004 should be at most 0.5\n",
            "To fix, set magnification to 41.0, and downsample the resulting image with dt.AveragePooling((41.0, 41.0, 1))\n",
            "\n",
            "  warnings.warn(\n",
            "c:\\users\\gu\\deeptrack\\deeptrack-2.0\\deeptrack\\optics.py:207: UserWarning: Likely bad optical parameters. NA / wavelength * resolution * magnification = 18.461538461538463 should be at most 0.5\n",
            "To fix, set magnification to 37.0, and downsample the resulting image with dt.AveragePooling((37.0, 37.0, 1))\n",
            "\n",
            "  warnings.warn(\n",
            "c:\\users\\gu\\deeptrack\\deeptrack-2.0\\deeptrack\\optics.py:207: UserWarning: Likely bad optical parameters. NA / wavelength * resolution * magnification = 17.142857142857146 should be at most 0.5\n",
            "To fix, set magnification to 35.0, and downsample the resulting image with dt.AveragePooling((35.0, 35.0, 1))\n",
            "\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "NUMBER_OF_IMAGES = 4\n",
        "\n",
        "for _ in range(NUMBER_OF_IMAGES):\n",
        "\n",
        "    dataset.update()\n",
        "    image_of_particle = dataset.resolve()\n",
        "    position_of_particle = get_label(image_of_particle) * IMAGE_SIZE + IMAGE_SIZE / 2\n",
        "    plt.imshow(image_of_particle[..., 0], cmap=\"gray\")\n",
        "    plt.colorbar()\n",
        "    plt.scatter(position_of_particle[1], position_of_particle[0], s=120, facecolors='none', edgecolors=\"g\", linewidth=2)\n",
        "    plt.show()\n",
        "    "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### 2.4 Augmenting dataset\n",
        "\n",
        "Simulating mie particles is slow. To speed up training we implement augmentation techniques. Here we flip and mirror the image. Note that DeepTrack ensures that the position is still correct after the augmentation. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:59.728434Z",
          "iopub.status.busy": "2022-06-30T10:52:59.727933Z",
          "iopub.status.idle": "2022-06-30T10:52:59.731440Z",
          "shell.execute_reply": "2022-06-30T10:52:59.731440Z"
        }
      },
      "outputs": [],
      "source": [
        "augmented_dataset = dt.Reuse(dataset, 8) >> dt.FlipLR() >> dt.FlipUD() >> dt.FlipDiagonal()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "We add noises after augmentation. This allows the augmented images to be more distinct."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:59.733933Z",
          "iopub.status.busy": "2022-06-30T10:52:59.733933Z",
          "iopub.status.idle": "2022-06-30T10:52:59.737932Z",
          "shell.execute_reply": "2022-06-30T10:52:59.737932Z"
        }
      },
      "outputs": [],
      "source": [
        "gradient = dt.IlluminationGradient(\n",
        "    gradient=lambda: np.random.randn(2) * 1e-4,\n",
        ")\n",
        "\n",
        "noise = dt.Poisson(\n",
        "    min_snr=5,\n",
        "    max_snr=100,\n",
        "    snr=lambda min_snr, max_snr: min_snr + np.random.rand() * (max_snr - min_snr),\n",
        "    background=1\n",
        ")\n",
        "\n",
        "normalization = dt.NormalizeMinMax(lambda: np.random.rand() * 0.2, lambda: 0.8 + np.random.rand() * 0.2)\n",
        "\n",
        "data_pipeline = augmented_dataset >> gradient >> noise >> normalization\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:52:59.740434Z",
          "iopub.status.busy": "2022-06-30T10:52:59.740434Z",
          "iopub.status.idle": "2022-06-30T10:53:00.285933Z",
          "shell.execute_reply": "2022-06-30T10:53:00.285434Z"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[24.99846677 25.88724352]\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[24.11275648 25.00153323]\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[25.00153323 24.11275648]\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[25.88724352 24.99846677]\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "NUMBER_OF_IMAGES = 4\n",
        "\n",
        "for _ in range(NUMBER_OF_IMAGES):\n",
        "\n",
        "    data_pipeline.update()\n",
        "    image_of_particle = data_pipeline()\n",
        "    position_of_particle = get_label(image_of_particle) * IMAGE_SIZE + IMAGE_SIZE / 2\n",
        "    print(position_of_particle)\n",
        "    plt.imshow(image_of_particle[..., 0], cmap=\"gray\")\n",
        "    plt.colorbar()\n",
        "    plt.scatter(position_of_particle[1], position_of_particle[0], s=120, facecolors='none', edgecolors=\"g\", linewidth=2)\n",
        "    plt.show()\n",
        "    "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## 3. Defining the network\n",
        "\n",
        "The network used is a Convolutional network, with mse as loss."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:53:00.288433Z",
          "iopub.status.busy": "2022-06-30T10:53:00.288433Z",
          "iopub.status.idle": "2022-06-30T10:53:00.793932Z",
          "shell.execute_reply": "2022-06-30T10:53:00.793932Z"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 51, 51, 1)]       0         \n",
            "                                                                 \n",
            " conv2d (Conv2D)             (None, 51, 51, 16)        160       \n",
            "                                                                 \n",
            " activation (Activation)     (None, 51, 51, 16)        0         \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 25, 25, 16)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 25, 25, 32)        4640      \n",
            "                                                                 \n",
            " activation_1 (Activation)   (None, 25, 25, 32)        0         \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 12, 12, 32)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_2 (Conv2D)           (None, 12, 12, 64)        18496     \n",
            "                                                                 \n",
            " activation_2 (Activation)   (None, 12, 12, 64)        0         \n",
            "                                                                 \n",
            " max_pooling2d_2 (MaxPooling  (None, 6, 6, 64)         0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 2304)              0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 32)                73760     \n",
            "                                                                 \n",
            " activation_3 (Activation)   (None, 32)                0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 32)                1056      \n",
            "                                                                 \n",
            " activation_4 (Activation)   (None, 32)                0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 2)                 66        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 98,178\n",
            "Trainable params: 98,178\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "import tensorflow.keras.backend as K\n",
        "import tensorflow.keras.optimizers as optimizers\n",
        "def pixel_error(T, P):\n",
        "    return K.mean(K.sqrt(K.sum(K.square(T - P), axis=-1))) * IMAGE_SIZE\n",
        "\n",
        "model = dt.models.Convolutional(\n",
        "    input_shape=(IMAGE_SIZE, IMAGE_SIZE, 1),\n",
        "    conv_layers_dimensions=(16, 32, 64),\n",
        "    dense_layers_dimensions=(32, 32),\n",
        "    steps_per_pooling=1,\n",
        "    number_of_outputs=2,\n",
        "    loss=\"mse\",\n",
        "    metrics=[pixel_error],\n",
        "    optimizer=\"adam\",\n",
        "    dense_block=dt.layers.DenseBlock(activation=\"relu\"),\n",
        "    pooling_block=dt.layers.PoolingBlock(padding=\"valid\")\n",
        ")\n",
        "\n",
        "model.summary()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## 4. Training the network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "We use the `ContinuousGenerator` to generate the images. It creates a new thread and generates images while the model is training. \n",
        "\n",
        "Set TRAIN_MODEL to True to train the model, otherwise a pretrained model is downloaded."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:53:00.803934Z",
          "iopub.status.busy": "2022-06-30T10:53:00.803934Z",
          "iopub.status.idle": "2022-06-30T10:57:48.493534Z",
          "shell.execute_reply": "2022-06-30T10:57:48.493534Z"
        },
        "scrolled": true
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Generating 1004 / 1000 samples before starting training\n",
            "Epoch 1/250\n",
            "15/15 [==============================] - 4s 52ms/step - loss: 9.9321e-04 - pixel_error: 2.0605 - val_loss: 5.2224e-04 - val_pixel_error: 1.4501\n",
            "Epoch 2/250\n",
            "15/15 [==============================] - 0s 19ms/step - loss: 2.1147e-04 - pixel_error: 0.8914 - val_loss: 1.4976e-04 - val_pixel_error: 0.6850\n",
            "Epoch 3/250\n",
            "15/15 [==============================] - 0s 27ms/step - loss: 1.0225e-04 - pixel_error: 0.5713 - val_loss: 1.0218e-04 - val_pixel_error: 0.5500\n",
            "Epoch 4/250\n",
            "15/15 [==============================] - 0s 19ms/step - loss: 8.0869e-05 - pixel_error: 0.4968 - val_loss: 9.7096e-05 - val_pixel_error: 0.5578\n",
            "Epoch 5/250\n",
            "15/15 [==============================] - 0s 28ms/step - loss: 6.7782e-05 - pixel_error: 0.4686 - val_loss: 8.4642e-05 - val_pixel_error: 0.4954\n",
            "Epoch 6/250\n",
            "15/15 [==============================] - 0s 19ms/step - loss: 5.8364e-05 - pixel_error: 0.4254 - val_loss: 1.0386e-04 - val_pixel_error: 0.5468\n",
            "Epoch 7/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 4.6099e-05 - pixel_error: 0.3836 - val_loss: 6.7298e-05 - val_pixel_error: 0.4317\n",
            "Epoch 8/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 3.5949e-05 - pixel_error: 0.3327 - val_loss: 5.1120e-05 - val_pixel_error: 0.3599\n",
            "Epoch 9/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 2.9882e-05 - pixel_error: 0.3113 - val_loss: 4.5586e-05 - val_pixel_error: 0.3360\n",
            "Epoch 10/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 2.4438e-05 - pixel_error: 0.2772 - val_loss: 4.0967e-05 - val_pixel_error: 0.3275\n",
            "Epoch 11/250\n",
            "15/15 [==============================] - 0s 28ms/step - loss: 2.1663e-05 - pixel_error: 0.2607 - val_loss: 3.3650e-05 - val_pixel_error: 0.2898\n",
            "Epoch 12/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 1.8973e-05 - pixel_error: 0.2500 - val_loss: 2.9210e-05 - val_pixel_error: 0.2713\n",
            "Epoch 13/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 1.6594e-05 - pixel_error: 0.2371 - val_loss: 2.7724e-05 - val_pixel_error: 0.2801\n",
            "Epoch 14/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 1.7209e-05 - pixel_error: 0.2438 - val_loss: 2.7063e-05 - val_pixel_error: 0.2672\n",
            "Epoch 15/250\n",
            "15/15 [==============================] - 1s 33ms/step - loss: 1.3916e-05 - pixel_error: 0.2223 - val_loss: 2.4086e-05 - val_pixel_error: 0.2721\n",
            "Epoch 16/250\n",
            "15/15 [==============================] - 0s 27ms/step - loss: 1.3216e-05 - pixel_error: 0.2135 - val_loss: 2.4207e-05 - val_pixel_error: 0.2569\n",
            "Epoch 17/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 1.4663e-05 - pixel_error: 0.2284 - val_loss: 2.2422e-05 - val_pixel_error: 0.2585\n",
            "Epoch 18/250\n",
            "15/15 [==============================] - 0s 24ms/step - loss: 1.3717e-05 - pixel_error: 0.2220 - val_loss: 1.8356e-05 - val_pixel_error: 0.2219\n",
            "Epoch 19/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 1.4686e-05 - pixel_error: 0.2347 - val_loss: 1.9984e-05 - val_pixel_error: 0.2454\n",
            "Epoch 20/250\n",
            "15/15 [==============================] - 0s 32ms/step - loss: 1.2575e-05 - pixel_error: 0.2080 - val_loss: 2.0585e-05 - val_pixel_error: 0.2440\n",
            "Epoch 21/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 9.8453e-06 - pixel_error: 0.1909 - val_loss: 1.6860e-05 - val_pixel_error: 0.2153\n",
            "Epoch 22/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 9.6152e-06 - pixel_error: 0.1895 - val_loss: 1.7617e-05 - val_pixel_error: 0.2258\n",
            "Epoch 23/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 1.1149e-05 - pixel_error: 0.2027 - val_loss: 1.4444e-05 - val_pixel_error: 0.2077\n",
            "Epoch 24/250\n",
            "15/15 [==============================] - 0s 13ms/step - loss: 1.3460e-05 - pixel_error: 0.2241 - val_loss: 3.1241e-05 - val_pixel_error: 0.3220\n",
            "Epoch 25/250\n",
            "15/15 [==============================] - 0s 14ms/step - loss: 1.1958e-05 - pixel_error: 0.2085 - val_loss: 2.0172e-05 - val_pixel_error: 0.2599\n",
            "Epoch 26/250\n",
            "15/15 [==============================] - 0s 11ms/step - loss: 1.0358e-05 - pixel_error: 0.2030 - val_loss: 1.6911e-05 - val_pixel_error: 0.2151\n",
            "Epoch 27/250\n",
            "15/15 [==============================] - 0s 12ms/step - loss: 7.6586e-06 - pixel_error: 0.1700 - val_loss: 1.1703e-05 - val_pixel_error: 0.1787\n",
            "Epoch 28/250\n",
            "15/15 [==============================] - 0s 6ms/step - loss: 6.4272e-06 - pixel_error: 0.1540 - val_loss: 1.1690e-05 - val_pixel_error: 0.1773\n",
            "Epoch 29/250\n",
            "15/15 [==============================] - 0s 10ms/step - loss: 6.9168e-06 - pixel_error: 0.1584 - val_loss: 1.4760e-05 - val_pixel_error: 0.2275\n",
            "Epoch 30/250\n",
            "15/15 [==============================] - 0s 10ms/step - loss: 7.9072e-06 - pixel_error: 0.1722 - val_loss: 1.2663e-05 - val_pixel_error: 0.2021\n",
            "Epoch 31/250\n",
            "15/15 [==============================] - 0s 13ms/step - loss: 7.0377e-06 - pixel_error: 0.1642 - val_loss: 1.6243e-05 - val_pixel_error: 0.2161\n",
            "Epoch 32/250\n",
            "15/15 [==============================] - 0s 12ms/step - loss: 8.1500e-06 - pixel_error: 0.1744 - val_loss: 9.8313e-06 - val_pixel_error: 0.1720\n",
            "Epoch 33/250\n",
            "15/15 [==============================] - 0s 12ms/step - loss: 6.3107e-06 - pixel_error: 0.1516 - val_loss: 1.1049e-05 - val_pixel_error: 0.1820\n",
            "Epoch 34/250\n",
            "15/15 [==============================] - 0s 10ms/step - loss: 6.2303e-06 - pixel_error: 0.1510 - val_loss: 9.4747e-06 - val_pixel_error: 0.1654\n",
            "Epoch 35/250\n",
            "15/15 [==============================] - 0s 10ms/step - loss: 5.6642e-06 - pixel_error: 0.1435 - val_loss: 1.1590e-05 - val_pixel_error: 0.1916\n",
            "Epoch 36/250\n",
            "15/15 [==============================] - 0s 12ms/step - loss: 6.1854e-06 - pixel_error: 0.1539 - val_loss: 1.0942e-05 - val_pixel_error: 0.1821\n",
            "Epoch 37/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 6.9994e-06 - pixel_error: 0.1488 - val_loss: 1.0294e-05 - val_pixel_error: 0.1627\n",
            "Epoch 38/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 7.2665e-06 - pixel_error: 0.1524 - val_loss: 1.1686e-05 - val_pixel_error: 0.1827\n",
            "Epoch 39/250\n",
            "15/15 [==============================] - 1s 36ms/step - loss: 8.0428e-06 - pixel_error: 0.1588 - val_loss: 9.8822e-06 - val_pixel_error: 0.1656\n",
            "Epoch 40/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 8.1281e-06 - pixel_error: 0.1623 - val_loss: 9.5136e-06 - val_pixel_error: 0.1604\n",
            "Epoch 41/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 7.8137e-06 - pixel_error: 0.1585 - val_loss: 1.1221e-05 - val_pixel_error: 0.1789\n",
            "Epoch 42/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 6.6661e-06 - pixel_error: 0.1460 - val_loss: 1.0974e-05 - val_pixel_error: 0.1828\n",
            "Epoch 43/250\n",
            "15/15 [==============================] - 0s 32ms/step - loss: 6.3068e-06 - pixel_error: 0.1495 - val_loss: 1.0646e-05 - val_pixel_error: 0.1619\n",
            "Epoch 44/250\n",
            "15/15 [==============================] - 1s 36ms/step - loss: 5.5909e-06 - pixel_error: 0.1390 - val_loss: 1.1994e-05 - val_pixel_error: 0.1940\n",
            "Epoch 45/250\n",
            "15/15 [==============================] - 0s 32ms/step - loss: 6.3072e-06 - pixel_error: 0.1465 - val_loss: 8.9497e-06 - val_pixel_error: 0.1548\n",
            "Epoch 46/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 5.3678e-06 - pixel_error: 0.1318 - val_loss: 1.0245e-05 - val_pixel_error: 0.1625\n",
            "Epoch 47/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 5.1252e-06 - pixel_error: 0.1325 - val_loss: 1.0634e-05 - val_pixel_error: 0.1833\n",
            "Epoch 48/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 4.4402e-06 - pixel_error: 0.1222 - val_loss: 8.8573e-06 - val_pixel_error: 0.1535\n",
            "Epoch 49/250\n",
            "15/15 [==============================] - 1s 36ms/step - loss: 4.5398e-06 - pixel_error: 0.1242 - val_loss: 1.1685e-05 - val_pixel_error: 0.1773\n",
            "Epoch 50/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 5.3407e-06 - pixel_error: 0.1397 - val_loss: 9.8983e-06 - val_pixel_error: 0.1769\n",
            "Epoch 51/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 5.3055e-06 - pixel_error: 0.1383 - val_loss: 1.2179e-05 - val_pixel_error: 0.1962\n",
            "Epoch 52/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 5.8199e-06 - pixel_error: 0.1449 - val_loss: 1.1396e-05 - val_pixel_error: 0.1674\n",
            "Epoch 53/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 4.4586e-06 - pixel_error: 0.1261 - val_loss: 8.5692e-06 - val_pixel_error: 0.1542\n",
            "Epoch 54/250\n",
            "15/15 [==============================] - 0s 35ms/step - loss: 3.8699e-06 - pixel_error: 0.1154 - val_loss: 9.6211e-06 - val_pixel_error: 0.1635\n",
            "Epoch 55/250\n",
            "15/15 [==============================] - 0s 18ms/step - loss: 3.6359e-06 - pixel_error: 0.1122 - val_loss: 8.9243e-06 - val_pixel_error: 0.1515\n",
            "Epoch 56/250\n",
            "15/15 [==============================] - 0s 16ms/step - loss: 3.3199e-06 - pixel_error: 0.1099 - val_loss: 1.1330e-05 - val_pixel_error: 0.1789\n",
            "Epoch 57/250\n",
            "15/15 [==============================] - 0s 9ms/step - loss: 3.7543e-06 - pixel_error: 0.1178 - val_loss: 8.8810e-06 - val_pixel_error: 0.1565\n",
            "Epoch 58/250\n",
            "15/15 [==============================] - 0s 11ms/step - loss: 4.0713e-06 - pixel_error: 0.1203 - val_loss: 1.0081e-05 - val_pixel_error: 0.1755\n",
            "Epoch 59/250\n",
            "15/15 [==============================] - 0s 8ms/step - loss: 4.7006e-06 - pixel_error: 0.1322 - val_loss: 7.4592e-06 - val_pixel_error: 0.1420\n",
            "Epoch 60/250\n",
            "15/15 [==============================] - 0s 18ms/step - loss: 3.5971e-06 - pixel_error: 0.1140 - val_loss: 1.1743e-05 - val_pixel_error: 0.1871\n",
            "Epoch 61/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 4.5321e-06 - pixel_error: 0.1309 - val_loss: 1.0220e-05 - val_pixel_error: 0.1826\n",
            "Epoch 62/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 4.4545e-06 - pixel_error: 0.1268 - val_loss: 7.9111e-06 - val_pixel_error: 0.1388\n",
            "Epoch 63/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 3.3516e-06 - pixel_error: 0.1086 - val_loss: 1.0877e-05 - val_pixel_error: 0.1557\n",
            "Epoch 64/250\n",
            "15/15 [==============================] - 1s 39ms/step - loss: 3.7513e-06 - pixel_error: 0.1136 - val_loss: 8.7817e-06 - val_pixel_error: 0.1476\n",
            "Epoch 65/250\n",
            "15/15 [==============================] - 1s 46ms/step - loss: 4.4302e-06 - pixel_error: 0.1266 - val_loss: 1.1125e-05 - val_pixel_error: 0.1921\n",
            "Epoch 66/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 7.2704e-06 - pixel_error: 0.1511 - val_loss: 1.4376e-05 - val_pixel_error: 0.2007\n",
            "Epoch 67/250\n",
            "15/15 [==============================] - 1s 39ms/step - loss: 7.1026e-06 - pixel_error: 0.1538 - val_loss: 1.1193e-05 - val_pixel_error: 0.1639\n",
            "Epoch 68/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 6.8167e-06 - pixel_error: 0.1494 - val_loss: 1.5848e-05 - val_pixel_error: 0.1902\n",
            "Epoch 69/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 6.5882e-06 - pixel_error: 0.1520 - val_loss: 1.8902e-05 - val_pixel_error: 0.1815\n",
            "Epoch 70/250\n",
            "15/15 [==============================] - 0s 33ms/step - loss: 6.3544e-06 - pixel_error: 0.1428 - val_loss: 1.0272e-05 - val_pixel_error: 0.1480\n",
            "Epoch 71/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 5.6043e-06 - pixel_error: 0.1428 - val_loss: 1.2553e-05 - val_pixel_error: 0.1750\n",
            "Epoch 72/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 5.4434e-06 - pixel_error: 0.1408 - val_loss: 1.0125e-05 - val_pixel_error: 0.1582\n",
            "Epoch 73/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 5.0608e-06 - pixel_error: 0.1339 - val_loss: 1.2165e-05 - val_pixel_error: 0.1857\n",
            "Epoch 74/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 6.7990e-06 - pixel_error: 0.1590 - val_loss: 1.0973e-05 - val_pixel_error: 0.1511\n",
            "Epoch 75/250\n",
            "15/15 [==============================] - 0s 32ms/step - loss: 3.8669e-06 - pixel_error: 0.1175 - val_loss: 1.0188e-05 - val_pixel_error: 0.1619\n",
            "Epoch 76/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 4.1854e-06 - pixel_error: 0.1248 - val_loss: 9.8436e-06 - val_pixel_error: 0.1572\n",
            "Epoch 77/250\n",
            "15/15 [==============================] - 0s 28ms/step - loss: 3.5623e-06 - pixel_error: 0.1135 - val_loss: 1.0369e-05 - val_pixel_error: 0.1746\n",
            "Epoch 78/250\n",
            "15/15 [==============================] - 0s 24ms/step - loss: 3.6182e-06 - pixel_error: 0.1158 - val_loss: 8.6271e-06 - val_pixel_error: 0.1324\n",
            "Epoch 79/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 3.1375e-06 - pixel_error: 0.1064 - val_loss: 1.0520e-05 - val_pixel_error: 0.1442\n",
            "Epoch 80/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 3.5666e-06 - pixel_error: 0.1152 - val_loss: 1.0114e-05 - val_pixel_error: 0.1533\n",
            "Epoch 81/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 3.6339e-06 - pixel_error: 0.1153 - val_loss: 1.0926e-05 - val_pixel_error: 0.1490\n",
            "Epoch 82/250\n",
            "15/15 [==============================] - 1s 28ms/step - loss: 3.6545e-06 - pixel_error: 0.1178 - val_loss: 9.2072e-06 - val_pixel_error: 0.1379\n",
            "Epoch 83/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 3.6701e-06 - pixel_error: 0.1187 - val_loss: 1.1505e-05 - val_pixel_error: 0.1672\n",
            "Epoch 84/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 3.0225e-06 - pixel_error: 0.1054 - val_loss: 9.7525e-06 - val_pixel_error: 0.1309\n",
            "Epoch 85/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 2.3899e-06 - pixel_error: 0.0941 - val_loss: 1.0036e-05 - val_pixel_error: 0.1523\n",
            "Epoch 86/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.2676e-06 - pixel_error: 0.0910 - val_loss: 1.0086e-05 - val_pixel_error: 0.1346\n",
            "Epoch 87/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 2.0701e-06 - pixel_error: 0.0872 - val_loss: 9.3856e-06 - val_pixel_error: 0.1314\n",
            "Epoch 88/250\n",
            "15/15 [==============================] - 0s 24ms/step - loss: 2.1661e-06 - pixel_error: 0.0896 - val_loss: 1.0162e-05 - val_pixel_error: 0.1396\n",
            "Epoch 89/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 1.9752e-06 - pixel_error: 0.0858 - val_loss: 9.0896e-06 - val_pixel_error: 0.1337\n",
            "Epoch 90/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 2.2199e-06 - pixel_error: 0.0922 - val_loss: 9.6267e-06 - val_pixel_error: 0.1400\n",
            "Epoch 91/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 2.4130e-06 - pixel_error: 0.0934 - val_loss: 1.0864e-05 - val_pixel_error: 0.1414\n",
            "Epoch 92/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 2.7422e-06 - pixel_error: 0.0970 - val_loss: 1.0303e-05 - val_pixel_error: 0.1358\n",
            "Epoch 93/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 3.4587e-06 - pixel_error: 0.1084 - val_loss: 1.0109e-05 - val_pixel_error: 0.1436\n",
            "Epoch 94/250\n",
            "15/15 [==============================] - 1s 39ms/step - loss: 4.5100e-06 - pixel_error: 0.1262 - val_loss: 1.3133e-05 - val_pixel_error: 0.1722\n",
            "Epoch 95/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 4.5408e-06 - pixel_error: 0.1285 - val_loss: 1.3575e-05 - val_pixel_error: 0.1831\n",
            "Epoch 96/250\n",
            "15/15 [==============================] - 1s 39ms/step - loss: 4.2682e-06 - pixel_error: 0.1210 - val_loss: 1.2351e-05 - val_pixel_error: 0.1471\n",
            "Epoch 97/250\n",
            "15/15 [==============================] - 1s 36ms/step - loss: 4.2937e-06 - pixel_error: 0.1164 - val_loss: 1.0659e-05 - val_pixel_error: 0.1379\n",
            "Epoch 98/250\n",
            "15/15 [==============================] - 1s 34ms/step - loss: 4.0562e-06 - pixel_error: 0.1139 - val_loss: 1.1458e-05 - val_pixel_error: 0.1562\n",
            "Epoch 99/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 3.8462e-06 - pixel_error: 0.1172 - val_loss: 9.8971e-06 - val_pixel_error: 0.1426\n",
            "Epoch 100/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 3.2203e-06 - pixel_error: 0.1070 - val_loss: 1.1237e-05 - val_pixel_error: 0.1729\n",
            "Epoch 101/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 3.3677e-06 - pixel_error: 0.1086 - val_loss: 1.1017e-05 - val_pixel_error: 0.1682\n",
            "Epoch 102/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 3.2241e-06 - pixel_error: 0.1077 - val_loss: 1.2114e-05 - val_pixel_error: 0.1505\n",
            "Epoch 103/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 3.0551e-06 - pixel_error: 0.1025 - val_loss: 9.8671e-06 - val_pixel_error: 0.1556\n",
            "Epoch 104/250\n",
            "15/15 [==============================] - 0s 27ms/step - loss: 3.1907e-06 - pixel_error: 0.1069 - val_loss: 1.1201e-05 - val_pixel_error: 0.1577\n",
            "Epoch 105/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 4.1720e-06 - pixel_error: 0.1241 - val_loss: 9.3723e-06 - val_pixel_error: 0.1650\n",
            "Epoch 106/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 4.4014e-06 - pixel_error: 0.1277 - val_loss: 1.3894e-05 - val_pixel_error: 0.1933\n",
            "Epoch 107/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 4.7520e-06 - pixel_error: 0.1331 - val_loss: 1.0801e-05 - val_pixel_error: 0.1810\n",
            "Epoch 108/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 3.7241e-06 - pixel_error: 0.1152 - val_loss: 1.0713e-05 - val_pixel_error: 0.1558\n",
            "Epoch 109/250\n",
            "15/15 [==============================] - 0s 24ms/step - loss: 3.3055e-06 - pixel_error: 0.1099 - val_loss: 8.5083e-06 - val_pixel_error: 0.1397\n",
            "Epoch 110/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.4471e-06 - pixel_error: 0.0933 - val_loss: 7.8698e-06 - val_pixel_error: 0.1331\n",
            "Epoch 111/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 2.4328e-06 - pixel_error: 0.0944 - val_loss: 7.7248e-06 - val_pixel_error: 0.1317\n",
            "Epoch 112/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 2.8907e-06 - pixel_error: 0.1046 - val_loss: 9.4337e-06 - val_pixel_error: 0.1299\n",
            "Epoch 113/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 3.0501e-06 - pixel_error: 0.1074 - val_loss: 8.4650e-06 - val_pixel_error: 0.1324\n",
            "Epoch 114/250\n",
            "15/15 [==============================] - 1s 33ms/step - loss: 2.2703e-06 - pixel_error: 0.0918 - val_loss: 9.1721e-06 - val_pixel_error: 0.1298\n",
            "Epoch 115/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 2.2686e-06 - pixel_error: 0.0920 - val_loss: 9.3033e-06 - val_pixel_error: 0.1427\n",
            "Epoch 116/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 2.1954e-06 - pixel_error: 0.0895 - val_loss: 7.6707e-06 - val_pixel_error: 0.1270\n",
            "Epoch 117/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 2.3452e-06 - pixel_error: 0.0915 - val_loss: 7.4559e-06 - val_pixel_error: 0.1266\n",
            "Epoch 118/250\n",
            "15/15 [==============================] - 1s 42ms/step - loss: 2.5111e-06 - pixel_error: 0.0874 - val_loss: 6.3885e-06 - val_pixel_error: 0.1207\n",
            "Epoch 119/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 2.4896e-06 - pixel_error: 0.0906 - val_loss: 6.4136e-06 - val_pixel_error: 0.1196\n",
            "Epoch 120/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 2.8837e-06 - pixel_error: 0.0989 - val_loss: 5.7658e-06 - val_pixel_error: 0.1216\n",
            "Epoch 121/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 2.5629e-06 - pixel_error: 0.0919 - val_loss: 5.2348e-06 - val_pixel_error: 0.1306\n",
            "Epoch 122/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 3.1230e-06 - pixel_error: 0.1059 - val_loss: 6.3768e-06 - val_pixel_error: 0.1487\n",
            "Epoch 123/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 3.9540e-06 - pixel_error: 0.1228 - val_loss: 4.4599e-06 - val_pixel_error: 0.1108\n",
            "Epoch 124/250\n",
            "15/15 [==============================] - 0s 28ms/step - loss: 2.5788e-06 - pixel_error: 0.0957 - val_loss: 5.0195e-06 - val_pixel_error: 0.1271\n",
            "Epoch 125/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 8.1490e-06 - pixel_error: 0.1475 - val_loss: 9.8304e-06 - val_pixel_error: 0.1822\n",
            "Epoch 126/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 8.9633e-06 - pixel_error: 0.1772 - val_loss: 1.0622e-05 - val_pixel_error: 0.1603\n",
            "Epoch 127/250\n",
            "15/15 [==============================] - 0s 28ms/step - loss: 6.1088e-06 - pixel_error: 0.1504 - val_loss: 7.5003e-06 - val_pixel_error: 0.1368\n",
            "Epoch 128/250\n",
            "15/15 [==============================] - 0s 28ms/step - loss: 4.7418e-06 - pixel_error: 0.1310 - val_loss: 7.1705e-06 - val_pixel_error: 0.1552\n",
            "Epoch 129/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 3.9840e-06 - pixel_error: 0.1175 - val_loss: 1.1678e-05 - val_pixel_error: 0.1460\n",
            "Epoch 130/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 5.3478e-06 - pixel_error: 0.1246 - val_loss: 8.8417e-06 - val_pixel_error: 0.1565\n",
            "Epoch 131/250\n",
            "15/15 [==============================] - 1s 34ms/step - loss: 4.9904e-06 - pixel_error: 0.1285 - val_loss: 8.4111e-06 - val_pixel_error: 0.1685\n",
            "Epoch 132/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 4.7955e-06 - pixel_error: 0.1343 - val_loss: 6.7845e-06 - val_pixel_error: 0.1487\n",
            "Epoch 133/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 3.3907e-06 - pixel_error: 0.1089 - val_loss: 5.9403e-06 - val_pixel_error: 0.1270\n",
            "Epoch 134/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 3.1255e-06 - pixel_error: 0.1043 - val_loss: 7.0542e-06 - val_pixel_error: 0.1549\n",
            "Epoch 135/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 3.4219e-06 - pixel_error: 0.1118 - val_loss: 6.9392e-06 - val_pixel_error: 0.1451\n",
            "Epoch 136/250\n",
            "15/15 [==============================] - 0s 26ms/step - loss: 2.9780e-06 - pixel_error: 0.1046 - val_loss: 5.7588e-06 - val_pixel_error: 0.1285\n",
            "Epoch 137/250\n",
            "15/15 [==============================] - 0s 27ms/step - loss: 2.6487e-06 - pixel_error: 0.0963 - val_loss: 5.6203e-06 - val_pixel_error: 0.1160\n",
            "Epoch 138/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.1706e-06 - pixel_error: 0.0862 - val_loss: 5.0523e-06 - val_pixel_error: 0.1242\n",
            "Epoch 139/250\n",
            "15/15 [==============================] - 0s 17ms/step - loss: 2.2500e-06 - pixel_error: 0.0911 - val_loss: 5.0690e-06 - val_pixel_error: 0.1149\n",
            "Epoch 140/250\n",
            "15/15 [==============================] - 0s 6ms/step - loss: 1.7891e-06 - pixel_error: 0.0812 - val_loss: 4.4571e-06 - val_pixel_error: 0.1064\n",
            "Epoch 141/250\n",
            "15/15 [==============================] - 0s 11ms/step - loss: 1.8296e-06 - pixel_error: 0.0826 - val_loss: 4.0510e-06 - val_pixel_error: 0.1009\n",
            "Epoch 142/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.5766e-06 - pixel_error: 0.0757 - val_loss: 4.9803e-06 - val_pixel_error: 0.1110\n",
            "Epoch 143/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 2.9056e-06 - pixel_error: 0.0992 - val_loss: 5.0983e-06 - val_pixel_error: 0.1259\n",
            "Epoch 144/250\n",
            "15/15 [==============================] - 0s 19ms/step - loss: 3.3747e-06 - pixel_error: 0.1067 - val_loss: 4.1659e-06 - val_pixel_error: 0.1097\n",
            "Epoch 145/250\n",
            "15/15 [==============================] - 0s 6ms/step - loss: 2.8323e-06 - pixel_error: 0.0943 - val_loss: 6.1719e-06 - val_pixel_error: 0.1312\n",
            "Epoch 146/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 2.7529e-06 - pixel_error: 0.0967 - val_loss: 5.8310e-06 - val_pixel_error: 0.1374\n",
            "Epoch 147/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 3.4087e-06 - pixel_error: 0.1103 - val_loss: 5.2910e-06 - val_pixel_error: 0.1214\n",
            "Epoch 148/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 2.9004e-06 - pixel_error: 0.0980 - val_loss: 4.5106e-06 - val_pixel_error: 0.1131\n",
            "Epoch 149/250\n",
            "15/15 [==============================] - 1s 40ms/step - loss: 3.2610e-06 - pixel_error: 0.1077 - val_loss: 4.8397e-06 - val_pixel_error: 0.1268\n",
            "Epoch 150/250\n",
            "15/15 [==============================] - 0s 32ms/step - loss: 3.4867e-06 - pixel_error: 0.1160 - val_loss: 5.1927e-06 - val_pixel_error: 0.1344\n",
            "Epoch 151/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 3.0300e-06 - pixel_error: 0.1061 - val_loss: 3.5412e-06 - val_pixel_error: 0.0991\n",
            "Epoch 152/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 2.6018e-06 - pixel_error: 0.0988 - val_loss: 4.9222e-06 - val_pixel_error: 0.1260\n",
            "Epoch 153/250\n",
            "15/15 [==============================] - 0s 33ms/step - loss: 2.8628e-06 - pixel_error: 0.1019 - val_loss: 3.2590e-06 - val_pixel_error: 0.0953\n",
            "Epoch 154/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.4041e-06 - pixel_error: 0.0933 - val_loss: 3.9361e-06 - val_pixel_error: 0.1039\n",
            "Epoch 155/250\n",
            "15/15 [==============================] - 0s 26ms/step - loss: 1.8998e-06 - pixel_error: 0.0826 - val_loss: 3.5383e-06 - val_pixel_error: 0.0999\n",
            "Epoch 156/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 2.1767e-06 - pixel_error: 0.0898 - val_loss: 3.9542e-06 - val_pixel_error: 0.1111\n",
            "Epoch 157/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 2.3564e-06 - pixel_error: 0.0958 - val_loss: 3.7837e-06 - val_pixel_error: 0.0986\n",
            "Epoch 158/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.6538e-06 - pixel_error: 0.0786 - val_loss: 3.5803e-06 - val_pixel_error: 0.0987\n",
            "Epoch 159/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 1.7865e-06 - pixel_error: 0.0820 - val_loss: 3.5408e-06 - val_pixel_error: 0.1002\n",
            "Epoch 160/250\n",
            "15/15 [==============================] - 1s 31ms/step - loss: 1.6430e-06 - pixel_error: 0.0776 - val_loss: 3.6171e-06 - val_pixel_error: 0.0989\n",
            "Epoch 161/250\n",
            "15/15 [==============================] - 0s 26ms/step - loss: 1.9331e-06 - pixel_error: 0.0863 - val_loss: 3.5091e-06 - val_pixel_error: 0.1052\n",
            "Epoch 162/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.9944e-06 - pixel_error: 0.0873 - val_loss: 3.2378e-06 - val_pixel_error: 0.0942\n",
            "Epoch 163/250\n",
            "15/15 [==============================] - 1s 39ms/step - loss: 1.8029e-06 - pixel_error: 0.0836 - val_loss: 3.6552e-06 - val_pixel_error: 0.1036\n",
            "Epoch 164/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 1.5712e-06 - pixel_error: 0.0774 - val_loss: 3.5098e-06 - val_pixel_error: 0.0971\n",
            "Epoch 165/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.6388e-06 - pixel_error: 0.0792 - val_loss: 3.5814e-06 - val_pixel_error: 0.0967\n",
            "Epoch 166/250\n",
            "15/15 [==============================] - 0s 19ms/step - loss: 1.6081e-06 - pixel_error: 0.0784 - val_loss: 4.2741e-06 - val_pixel_error: 0.1163\n",
            "Epoch 167/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 1.9462e-06 - pixel_error: 0.0851 - val_loss: 4.0566e-06 - val_pixel_error: 0.1124\n",
            "Epoch 168/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.8384e-06 - pixel_error: 0.0835 - val_loss: 3.6357e-06 - val_pixel_error: 0.1086\n",
            "Epoch 169/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.9604e-06 - pixel_error: 0.0861 - val_loss: 4.3981e-06 - val_pixel_error: 0.1199\n",
            "Epoch 170/250\n",
            "15/15 [==============================] - 0s 24ms/step - loss: 2.1545e-06 - pixel_error: 0.0915 - val_loss: 3.9779e-06 - val_pixel_error: 0.1064\n",
            "Epoch 171/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.4991e-06 - pixel_error: 0.0973 - val_loss: 4.0265e-06 - val_pixel_error: 0.1136\n",
            "Epoch 172/250\n",
            "15/15 [==============================] - 1s 39ms/step - loss: 2.9934e-06 - pixel_error: 0.0982 - val_loss: 5.8044e-06 - val_pixel_error: 0.1472\n",
            "Epoch 173/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 3.2872e-06 - pixel_error: 0.1097 - val_loss: 4.9825e-06 - val_pixel_error: 0.1191\n",
            "Epoch 174/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 3.0545e-06 - pixel_error: 0.1066 - val_loss: 4.4304e-06 - val_pixel_error: 0.1237\n",
            "Epoch 175/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 3.3778e-06 - pixel_error: 0.1067 - val_loss: 2.9174e-06 - val_pixel_error: 0.0899\n",
            "Epoch 176/250\n",
            "15/15 [==============================] - 0s 24ms/step - loss: 3.3128e-06 - pixel_error: 0.1095 - val_loss: 3.5187e-06 - val_pixel_error: 0.1051\n",
            "Epoch 177/250\n",
            "15/15 [==============================] - 0s 28ms/step - loss: 3.3645e-06 - pixel_error: 0.1101 - val_loss: 4.0165e-06 - val_pixel_error: 0.1126\n",
            "Epoch 178/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 2.8522e-06 - pixel_error: 0.1016 - val_loss: 3.9981e-06 - val_pixel_error: 0.1038\n",
            "Epoch 179/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 2.8667e-06 - pixel_error: 0.1036 - val_loss: 5.4430e-06 - val_pixel_error: 0.1403\n",
            "Epoch 180/250\n",
            "15/15 [==============================] - 0s 26ms/step - loss: 3.7177e-06 - pixel_error: 0.1170 - val_loss: 4.6635e-06 - val_pixel_error: 0.1289\n",
            "Epoch 181/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 3.3595e-06 - pixel_error: 0.1096 - val_loss: 4.9066e-06 - val_pixel_error: 0.1337\n",
            "Epoch 182/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.7074e-06 - pixel_error: 0.0974 - val_loss: 4.3886e-06 - val_pixel_error: 0.1086\n",
            "Epoch 183/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 2.6258e-06 - pixel_error: 0.0983 - val_loss: 4.2692e-06 - val_pixel_error: 0.1214\n",
            "Epoch 184/250\n",
            "15/15 [==============================] - 0s 24ms/step - loss: 2.0227e-06 - pixel_error: 0.0867 - val_loss: 3.2884e-06 - val_pixel_error: 0.0991\n",
            "Epoch 185/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.0090e-06 - pixel_error: 0.0860 - val_loss: 3.5283e-06 - val_pixel_error: 0.1061\n",
            "Epoch 186/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 2.0063e-06 - pixel_error: 0.0863 - val_loss: 3.4584e-06 - val_pixel_error: 0.1050\n",
            "Epoch 187/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.4740e-06 - pixel_error: 0.0956 - val_loss: 4.0184e-06 - val_pixel_error: 0.1192\n",
            "Epoch 188/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.1507e-06 - pixel_error: 0.0891 - val_loss: 3.1599e-06 - val_pixel_error: 0.0970\n",
            "Epoch 189/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 2.1487e-06 - pixel_error: 0.0870 - val_loss: 3.9209e-06 - val_pixel_error: 0.1052\n",
            "Epoch 190/250\n",
            "15/15 [==============================] - 0s 26ms/step - loss: 2.4952e-06 - pixel_error: 0.0956 - val_loss: 4.6875e-06 - val_pixel_error: 0.1331\n",
            "Epoch 191/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 3.6430e-06 - pixel_error: 0.1153 - val_loss: 4.0159e-06 - val_pixel_error: 0.1078\n",
            "Epoch 192/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 3.0871e-06 - pixel_error: 0.1023 - val_loss: 4.1708e-06 - val_pixel_error: 0.1148\n",
            "Epoch 193/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 4.9780e-06 - pixel_error: 0.1378 - val_loss: 4.8898e-06 - val_pixel_error: 0.1316\n",
            "Epoch 194/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 2.2225e-06 - pixel_error: 0.0906 - val_loss: 3.9304e-06 - val_pixel_error: 0.1057\n",
            "Epoch 195/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.8668e-06 - pixel_error: 0.0961 - val_loss: 5.7526e-06 - val_pixel_error: 0.1542\n",
            "Epoch 196/250\n",
            "15/15 [==============================] - 1s 36ms/step - loss: 4.2370e-06 - pixel_error: 0.1203 - val_loss: 6.1577e-06 - val_pixel_error: 0.1522\n",
            "Epoch 197/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 5.1794e-06 - pixel_error: 0.1409 - val_loss: 6.5238e-06 - val_pixel_error: 0.1474\n",
            "Epoch 198/250\n",
            "15/15 [==============================] - 1s 42ms/step - loss: 5.3248e-06 - pixel_error: 0.1437 - val_loss: 6.0847e-06 - val_pixel_error: 0.1434\n",
            "Epoch 199/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 4.3674e-06 - pixel_error: 0.1312 - val_loss: 7.3619e-06 - val_pixel_error: 0.1620\n",
            "Epoch 200/250\n",
            "15/15 [==============================] - 1s 40ms/step - loss: 3.9579e-06 - pixel_error: 0.1234 - val_loss: 4.6213e-06 - val_pixel_error: 0.1218\n",
            "Epoch 201/250\n",
            "15/15 [==============================] - 1s 39ms/step - loss: 3.0311e-06 - pixel_error: 0.1060 - val_loss: 3.8859e-06 - val_pixel_error: 0.1044\n",
            "Epoch 202/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 2.8213e-06 - pixel_error: 0.1013 - val_loss: 4.6032e-06 - val_pixel_error: 0.1169\n",
            "Epoch 203/250\n",
            "15/15 [==============================] - 0s 28ms/step - loss: 2.9346e-06 - pixel_error: 0.1008 - val_loss: 3.9858e-06 - val_pixel_error: 0.1085\n",
            "Epoch 204/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 3.4698e-06 - pixel_error: 0.1105 - val_loss: 6.1462e-06 - val_pixel_error: 0.1355\n",
            "Epoch 205/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 3.3648e-06 - pixel_error: 0.1073 - val_loss: 4.0396e-06 - val_pixel_error: 0.1143\n",
            "Epoch 206/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 3.0023e-06 - pixel_error: 0.1049 - val_loss: 3.5946e-06 - val_pixel_error: 0.1091\n",
            "Epoch 207/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 3.3773e-06 - pixel_error: 0.1090 - val_loss: 4.8135e-06 - val_pixel_error: 0.1135\n",
            "Epoch 208/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 3.8650e-06 - pixel_error: 0.1201 - val_loss: 4.8167e-06 - val_pixel_error: 0.1247\n",
            "Epoch 209/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 3.6807e-06 - pixel_error: 0.1188 - val_loss: 4.6940e-06 - val_pixel_error: 0.1251\n",
            "Epoch 210/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 3.4131e-06 - pixel_error: 0.1147 - val_loss: 4.8330e-06 - val_pixel_error: 0.1252\n",
            "Epoch 211/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 2.9450e-06 - pixel_error: 0.1080 - val_loss: 3.2747e-06 - val_pixel_error: 0.0924\n",
            "Epoch 212/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 1.9086e-06 - pixel_error: 0.0842 - val_loss: 4.3691e-06 - val_pixel_error: 0.1259\n",
            "Epoch 213/250\n",
            "15/15 [==============================] - 0s 23ms/step - loss: 2.1262e-06 - pixel_error: 0.0906 - val_loss: 3.1346e-06 - val_pixel_error: 0.0947\n",
            "Epoch 214/250\n",
            "15/15 [==============================] - 1s 38ms/step - loss: 2.0843e-06 - pixel_error: 0.0886 - val_loss: 3.3668e-06 - val_pixel_error: 0.0996\n",
            "Epoch 215/250\n",
            "15/15 [==============================] - 0s 30ms/step - loss: 1.8478e-06 - pixel_error: 0.0825 - val_loss: 2.9667e-06 - val_pixel_error: 0.0871\n",
            "Epoch 216/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 1.6879e-06 - pixel_error: 0.0790 - val_loss: 3.3386e-06 - val_pixel_error: 0.0992\n",
            "Epoch 217/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.7894e-06 - pixel_error: 0.0833 - val_loss: 3.0499e-06 - val_pixel_error: 0.0909\n",
            "Epoch 218/250\n",
            "15/15 [==============================] - 0s 31ms/step - loss: 2.0485e-06 - pixel_error: 0.0876 - val_loss: 2.9759e-06 - val_pixel_error: 0.0920\n",
            "Epoch 219/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 1.9134e-06 - pixel_error: 0.0864 - val_loss: 3.6668e-06 - val_pixel_error: 0.1085\n",
            "Epoch 220/250\n",
            "15/15 [==============================] - 0s 24ms/step - loss: 1.8068e-06 - pixel_error: 0.0833 - val_loss: 3.2944e-06 - val_pixel_error: 0.0972\n",
            "Epoch 221/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 2.0132e-06 - pixel_error: 0.0892 - val_loss: 3.2424e-06 - val_pixel_error: 0.1032\n",
            "Epoch 222/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.3676e-06 - pixel_error: 0.0719 - val_loss: 2.7122e-06 - val_pixel_error: 0.0861\n",
            "Epoch 223/250\n",
            "15/15 [==============================] - 1s 35ms/step - loss: 1.4266e-06 - pixel_error: 0.0726 - val_loss: 2.9089e-06 - val_pixel_error: 0.0944\n",
            "Epoch 224/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.5277e-06 - pixel_error: 0.0749 - val_loss: 3.3312e-06 - val_pixel_error: 0.1038\n",
            "Epoch 225/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.7447e-06 - pixel_error: 0.0819 - val_loss: 3.2504e-06 - val_pixel_error: 0.1009\n",
            "Epoch 226/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 1.7818e-06 - pixel_error: 0.0811 - val_loss: 3.0442e-06 - val_pixel_error: 0.0918\n",
            "Epoch 227/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 1.6679e-06 - pixel_error: 0.0774 - val_loss: 2.9479e-06 - val_pixel_error: 0.0968\n",
            "Epoch 228/250\n",
            "15/15 [==============================] - 0s 29ms/step - loss: 1.8140e-06 - pixel_error: 0.0784 - val_loss: 3.5870e-06 - val_pixel_error: 0.1063\n",
            "Epoch 229/250\n",
            "15/15 [==============================] - 1s 37ms/step - loss: 2.9484e-06 - pixel_error: 0.0988 - val_loss: 4.6449e-06 - val_pixel_error: 0.1086\n",
            "Epoch 230/250\n",
            "15/15 [==============================] - 1s 36ms/step - loss: 2.8687e-06 - pixel_error: 0.1028 - val_loss: 3.6940e-06 - val_pixel_error: 0.1067\n",
            "Epoch 231/250\n",
            "15/15 [==============================] - 1s 34ms/step - loss: 2.2262e-06 - pixel_error: 0.0889 - val_loss: 3.1824e-06 - val_pixel_error: 0.0983\n",
            "Epoch 232/250\n",
            "15/15 [==============================] - 1s 36ms/step - loss: 2.2042e-06 - pixel_error: 0.0886 - val_loss: 3.2113e-06 - val_pixel_error: 0.1000\n",
            "Epoch 233/250\n",
            "15/15 [==============================] - 1s 41ms/step - loss: 1.7296e-06 - pixel_error: 0.0784 - val_loss: 2.8556e-06 - val_pixel_error: 0.0927\n",
            "Epoch 234/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 1.6004e-06 - pixel_error: 0.0751 - val_loss: 3.1331e-06 - val_pixel_error: 0.0986\n",
            "Epoch 235/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 1.6648e-06 - pixel_error: 0.0785 - val_loss: 2.6895e-06 - val_pixel_error: 0.0905\n",
            "Epoch 236/250\n",
            "15/15 [==============================] - 0s 19ms/step - loss: 1.9118e-06 - pixel_error: 0.0793 - val_loss: 4.4535e-06 - val_pixel_error: 0.1156\n",
            "Epoch 237/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 2.4843e-06 - pixel_error: 0.0962 - val_loss: 3.8513e-06 - val_pixel_error: 0.1178\n",
            "Epoch 238/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 2.6221e-06 - pixel_error: 0.1002 - val_loss: 4.0366e-06 - val_pixel_error: 0.1135\n",
            "Epoch 239/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 2.7419e-06 - pixel_error: 0.1040 - val_loss: 3.3747e-06 - val_pixel_error: 0.1074\n",
            "Epoch 240/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.7873e-06 - pixel_error: 0.0818 - val_loss: 3.2266e-06 - val_pixel_error: 0.0989\n",
            "Epoch 241/250\n",
            "15/15 [==============================] - 1s 36ms/step - loss: 1.3829e-06 - pixel_error: 0.0716 - val_loss: 2.7855e-06 - val_pixel_error: 0.0889\n",
            "Epoch 242/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.2346e-06 - pixel_error: 0.0677 - val_loss: 3.1284e-06 - val_pixel_error: 0.0901\n",
            "Epoch 243/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.3413e-06 - pixel_error: 0.0689 - val_loss: 3.5255e-06 - val_pixel_error: 0.0967\n",
            "Epoch 244/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.7344e-06 - pixel_error: 0.0782 - val_loss: 3.3234e-06 - val_pixel_error: 0.1018\n",
            "Epoch 245/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.5913e-06 - pixel_error: 0.0789 - val_loss: 3.7795e-06 - val_pixel_error: 0.1102\n",
            "Epoch 246/250\n",
            "15/15 [==============================] - 0s 20ms/step - loss: 1.7800e-06 - pixel_error: 0.0820 - val_loss: 3.8597e-06 - val_pixel_error: 0.1137\n",
            "Epoch 247/250\n",
            "15/15 [==============================] - 0s 21ms/step - loss: 1.5677e-06 - pixel_error: 0.0777 - val_loss: 3.3133e-06 - val_pixel_error: 0.0982\n",
            "Epoch 248/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 1.4396e-06 - pixel_error: 0.0733 - val_loss: 3.1527e-06 - val_pixel_error: 0.0955\n",
            "Epoch 249/250\n",
            "15/15 [==============================] - 0s 25ms/step - loss: 1.5241e-06 - pixel_error: 0.0770 - val_loss: 3.4084e-06 - val_pixel_error: 0.0992\n",
            "Epoch 250/250\n",
            "15/15 [==============================] - 0s 22ms/step - loss: 1.2053e-06 - pixel_error: 0.0664 - val_loss: 3.1460e-06 - val_pixel_error: 0.0928\n",
            "                                                                        \r"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "TRAIN_MODEL = True\n",
        "\n",
        "from tensorflow.keras.callbacks import EarlyStopping\n",
        "\n",
        "validation_set_size = 200\n",
        "validation_set = [data_pipeline.update().resolve() for _ in range(validation_set_size)]\n",
        "validation_labels = [get_label(image) for image in validation_set]\n",
        "\n",
        "if TRAIN_MODEL:\n",
        "    generator = dt.generators.ContinuousGenerator(\n",
        "        data_pipeline & (data_pipeline >> get_label),\n",
        "        min_data_size=int(1e3),\n",
        "        max_data_size=int(2e3),\n",
        "        batch_size=64,\n",
        "        max_epochs_per_sample=25\n",
        "    )\n",
        "\n",
        "    histories = []\n",
        "\n",
        "    with generator:\n",
        "        h = model.fit(\n",
        "            generator,\n",
        "            validation_data=(\n",
        "                np.array(validation_set), \n",
        "                np.array(validation_labels)\n",
        "            ),\n",
        "            epochs=250\n",
        "        )\n",
        "\n",
        "    plt.plot(h.history[\"loss\"], 'g')\n",
        "    plt.plot(h.history[\"val_loss\"], 'r')\n",
        "    plt.legend([\"loss\", \"val_loss\"])\n",
        "    plt.yscale('log')\n",
        "    plt.show()\n",
        "    \n",
        "else:\n",
        "    model_path = datasets.load_model(\"ParticleTracking\")\n",
        "    model.load_weights(model_path)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## 5. Evaluating the network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### 5.1 Prediction vs actual\n",
        "\n",
        "We show the prediction of each output versus the ground truth"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:57:48.497035Z",
          "iopub.status.busy": "2022-06-30T10:57:48.496534Z",
          "iopub.status.idle": "2022-06-30T10:57:48.776035Z",
          "shell.execute_reply": "2022-06-30T10:57:48.776035Z"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "7/7 [==============================] - 0s 3ms/step\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "validation_prediction = model.predict(np.array(validation_set))\n",
        "labels = np.array(validation_labels)\n",
        "for col in range(validation_prediction.shape[-1]):\n",
        "    label_col = labels[:, col]\n",
        "    prediction_col = validation_prediction[:, col]\n",
        "    plt.scatter(label_col, prediction_col, alpha=0.1)\n",
        "    \n",
        "    plt.plot([np.min(label_col), np.max(label_col)], \n",
        "             [np.min(label_col), np.max(label_col)], c='k')\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### 5.3 Prediction vs property value\n",
        "\n",
        "We show the the pixel error as a function of some properties."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:57:48.779035Z",
          "iopub.status.busy": "2022-06-30T10:57:48.779035Z",
          "iopub.status.idle": "2022-06-30T10:57:50.211034Z",
          "shell.execute_reply": "2022-06-30T10:57:50.210534Z"
        },
        "scrolled": true
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "7/7 [==============================] - 0s 2ms/step\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "C:\\Users\\GU\\AppData\\Local\\Temp\\ipykernel_25976\\3554977865.py:20: UserWarning: Attempting to set identical left == right == 0.8 results in singular transformations; automatically expanding.\n",
            "  plt.xlim([np.min(values), np.max(values)])\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "properties = [\"position\", \"z\", \"snr\", \"gradient\", \"refractive_index\", \"NA\"]\n",
        "\n",
        "validation_prediction = model.predict(np.array(validation_set))\n",
        "\n",
        "snr = [image.get_property(\"snr\") for image in validation_set]\n",
        "\n",
        "validation_error = np.mean(np.abs(validation_prediction - validation_labels), axis=-1) * 51\n",
        "\n",
        "for property_name in properties:\n",
        "    property_values = np.array([image.get_property(property_name) for image in validation_set])\n",
        "    if property_values.ndim == 1:\n",
        "        property_values = np.expand_dims(property_values, axis=-1)\n",
        "    \n",
        "    for col in range(property_values.shape[1]):\n",
        "        values = property_values[:, col]\n",
        "\n",
        "        plt.subplot(1, property_values.shape[1], col + 1)\n",
        "\n",
        "        plt.scatter(values, validation_error, alpha=0.1)\n",
        "        plt.xlim([np.min(values), np.max(values)])\n",
        "        plt.ylim([np.min(validation_error), np.max(validation_error)])\n",
        "        plt.yscale(\"log\")\n",
        "        plt.ylabel(\"Pixel error\")\n",
        "        plt.xlabel(\"{0}[{1}]\".format(property_name, col))\n",
        "\n",
        "    \n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### 5.3 Experimental data\n",
        "We play some experimental videos tracked by the DeepTrack model, compared to radial center method."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:57:50.214035Z",
          "iopub.status.busy": "2022-06-30T10:57:50.213534Z",
          "iopub.status.idle": "2022-06-30T10:57:50.230034Z",
          "shell.execute_reply": "2022-06-30T10:57:50.229544Z"
        }
      },
      "outputs": [],
      "source": [
        "import cv2\n",
        "import IPython\n",
        "import numpy as np\n",
        "import os\n",
        "\n",
        "from radialcenter import radialcenter\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def track_video(video, frames_to_track):\n",
        "    video_width = int(video.get(cv2.CAP_PROP_FRAME_WIDTH))\n",
        "    video_height = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT))\n",
        "\n",
        "    # Initialize variables\n",
        "    predicted_positions = np.zeros((frames_to_track, 2))\n",
        "    predicted_positions_radial = np.zeros((frames_to_track, 2))\n",
        "\n",
        "    # Track the positions of the particles frame by frame\n",
        "\n",
        "    for i in range(frames_to_track):\n",
        "\n",
        "        # Read the current frame from the video\n",
        "        (ret, frame) = video.read()\n",
        "\n",
        "        frame = cv2.normalize(frame, None, 0, 255, cv2.NORM_MINMAX)\n",
        "\n",
        "        # Convert color image to grayscale.\n",
        "        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) / 255\n",
        "        \n",
        "        radial_x, radial_y = radialcenter(frame)\n",
        "        predicted_positions_radial[i, 0] = radial_x \n",
        "        predicted_positions_radial[i, 1] = radial_y\n",
        "\n",
        "        ### Resize the frame\n",
        "        frame_resize = cv2.resize(frame, (51, 51))\n",
        "\n",
        "        predicted_position = model.predict(np.reshape(frame_resize, (1, 51, 51, 1)))\n",
        "        \n",
        "        predicted_position_y = predicted_position[0,0] * video_width + video_width / 2 + 1\n",
        "        predicted_position_x = predicted_position[0,1] * video_height + video_height / 2 + 1\n",
        "        \n",
        "        predicted_positions[i, 0] = predicted_position_x\n",
        "        predicted_positions[i, 1] = predicted_position_y\n",
        "\n",
        "        IPython.display.clear_output(wait=True)\n",
        "\n",
        "        plt.imshow(frame, cmap=\"gray\")\n",
        "        # Make markerfacecolor transparent (marker edge color default).\n",
        "\n",
        "        plt.scatter(predicted_position_x, predicted_position_y, marker=\"o\", s=500, facecolors=\"none\", linewidths=3, edgecolors=\"r\")\n",
        "        plt.scatter(radial_x, radial_y, marker='x', s=400, c=\"g\", linewidths=3)\n",
        "        plt.axis(\"off\")\n",
        "\n",
        "        plt.show()\n",
        "        \n",
        "    return predicted_positions, predicted_positions_radial\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Here, the blue circle is deeptrack, and the cross is radial center"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:57:50.233035Z",
          "iopub.status.busy": "2022-06-30T10:57:50.232535Z",
          "iopub.status.idle": "2022-06-30T10:57:58.162534Z",
          "shell.execute_reply": "2022-06-30T10:57:58.162534Z"
        },
        "scrolled": false
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "video = cv2.VideoCapture(\"./datasets/ParticleTracking/ideal.avi\")\n",
        "p, pr = track_video(video, 100)\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "execution": {
          "iopub.execute_input": "2022-06-30T10:57:58.165534Z",
          "iopub.status.busy": "2022-06-30T10:57:58.165034Z",
          "iopub.status.idle": "2022-06-30T10:58:06.180534Z",
          "shell.execute_reply": "2022-06-30T10:58:06.180534Z"
        },
        "scrolled": true
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "video = cv2.VideoCapture(\"./datasets/ParticleTracking/bad.avi\")\n",
        "p, pr = track_video(video, 100)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.10.5"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}