{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Cellpose_cell_segmentation_2D_prediction_only.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "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": "3.7.6"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Nb90LCrotIx4"
      },
      "source": [
        "#**Cellpose Prediction for 2D v0.2**\n",
        "A generalist algorithm for cell and nucleus segmentation.\n",
        "\n",
        "***Cellpose code: Carsen Stringer & Marius Pachitariu***\n",
        "\n",
        "[Link to Paper](https://www.nature.com/articles/s41592-020-01018-x)\n",
        "\n",
        "[Link to Video talk](https://t.co/JChCsTD0SK?amp=1) \n",
        "\n",
        "Github Repo: https://github.com/MouseLand/cellpose\n",
        "\n",
        "The notebook is for processing **2D** images using the Cellpose package on Google Colab using the GPU. It processes images within the folder specified by the user. It will generate a folder called Masks which will contain the label images generated by Cellpose. \n",
        "\n",
        "The user can fine-tune and preview the parameters on one image before running it on all images within the folder. The user-specified channel will be used on all images for segmentation.\n",
        "\n",
        "This notebook was inspired by the [ZeroCostDL4Mic notebook series](https://github.com/HenriquesLab/ZeroCostDL4Mic/wiki). It was also adapted from colab notebook by Matteo Carandini hosted on the Cellpose github page. \n",
        "\n",
        "I have only tested it on jpg and tiff files. Feel free to email me if you have any questions. \n",
        "\n",
        "v0.2 update:\n",
        "\n",
        "* New models added (Cyto2, Omnipose...) in the notebook, which is quite similar to ZerCostDL4Mic notebook implementation\n",
        "* Parameters used in cellpose will be saved as a txt file (Thanks for suggestion Kota Miura)\n",
        "\n",
        "***Author of this notebook: Pradeep Rajasekhar***\n",
        "\n",
        "***Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia***\n",
        "\n",
        "rajasekhar.p@wehi.edu.au"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VbqFni8kuFar"
      },
      "source": [
        "#**Configuration**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z0s2fz5hUk75"
      },
      "source": [
        "### Make sure you have GPU access enabled by going to Runtime -> Change Runtime Type -> Hardware accelerator and selecting GPU\n",
        "\n",
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cJdG639F8uRL",
        "cellView": "form"
      },
      "source": [
        "#@markdown ##Step 1: Run this cell to connect your Google Drive to colab\n",
        "\n",
        "#@markdown * Click on the URL. \n",
        "\n",
        "#@markdown * Sign in your Google Account. \n",
        "\n",
        "#@markdown You will either have to:\n",
        "#@markdown * copy the authorisation code and enter it into box below OR\n",
        "\n",
        "#@markdown * in the new google colab, you can just click \"Allow\" and it should connect.\n",
        "\n",
        "#@markdown * Click on \"Folder\" icon on the Left, press the refresh button. Your Google Drive folder should now be available here as \"gdrive\". \n",
        "\n",
        "# mount user's Google Drive to Google Colab.\n",
        "from google.colab import drive\n",
        "drive.mount('/content/gdrive')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yH4zTaWwvH9x"
      },
      "source": [
        "###**How to get google drive path to your images**\n",
        "\n",
        "The above step allows this colab notebook to access your images from google drive. Now, you need the path of your folder containing the images. On the left of your notebook you will see a file icon:\n",
        "\n",
        "![image.png](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAkAAAAE1CAYAAAAPnV/ZAAAgAElEQVR4Aey9fZBVxbn/O3/fqlROnap779yqX1XuVMV7xphEPb8cA4lB5aBRfDkIvuLxZSRRCRHiwRERHQkOooKKOGKUFwUEgwQEFEQU0QiDSBiivCmjoIiOvA4zA/M+PLe+vfezdu/ea+299tuwZ+bbVXvW3mv16n760726v+vpXmuKhIEESIAESIAESIAE+hiBoj5WXhaXBEiABEiABEiABIQCiI2ABEiABEiABEigzxGgAOpzVc4CkwAJkAAJkAAJUACxDZAACZAACZAACfQ5AhRAfa7KWWASIAESIAESIAEKILYBEiABEiABEiCBPkeAAqjPVTkLTAIkQAIkkGsC7e3tgg9DzyFAAZSqrtoPibifVOfwOAmQAAmQQJ8isG/fPsGHoecQoABKUlfL9m6Stzf9XmTrhfGfY+uTnMVDJEACJNDLCexbJWMvK5GioiIpeaKmlxc2XPHefPNNWbVqVbjIjFUQBCiAAqrhlJySc5f8Sf7PudfL4S2XxQugHTeJyKmAM312d9RJ9aJKGXHRmVJcVCRFRcVy5kUjpHLZLqn3iZ5sV/WkIimaVG2i1C0aKkXXLpS6ZCf09mPfLJShRUNl4TcoaLVUFhVJ5abeXujTXL6WGpl2UYkMnHEaBr7D1TKtrL+U4DoqPlMGly+UXU0Wj02VZlDGwKyfkn5DZdSMVbI3zYutbtkIKfnxKFl12Eq/r331reu9MvuyIuk/HkzrZNe+eLCnj9vpu/4PHTokr7/+uvngO0PPIEABFFBPf/7Hq/LDOTeYz++WDYkXQPAIfTsr4Exn976lMuLHRVJ80ViZvbpG9tbXS339XqlZNk2GYv/NC2Vvh3NOkp8UQA4cCiAHSMBPIwwqJSKdA+L47Y7jG43QUiOV/Yq7XwA1VUtlvyLpf/dCqamrl/p966TyymIpvnlp7CbAlHOCrDLXGa61Otm1aalMu7ZEiooHy7RtLX6l9N1X99pwKe5LAihsXWu8gNme08ft9AmgDRs2yMcff2w++M7QMwhQAPnU0ydH9nniR0XQlo9uTBRBJ/f4nG3taqqWCecUSf9J1f6enqYaWbp2r3VC6q8UQA4j7YzpAXLAOD9zKYCcpLvt5zfrpGrSUom7Yg4vlRFF58q0bVErkpRz7/zhUlw8VtbZHqNuM74HZBR3LSWxN2y8JEnk51D3CKD6+nqBlwefnTt3ms9rr70m2I8Pvut+jYf9DIVHIG8CqOvQeuncM11OHXpfTp0IuFUI4IFzu75dLqfqtVcLiJin3de8/ViCANp3eEuiAKodl9SCvXMHS9Fls+M77KRntMiuZZUy/KfFxn1f/NPBMmpGvHgKFkC4+IfKwi3WFEFRifQvmybrEubI6k0+I/pF5vCLftxfRjyxTupsT9ThaqkaOVjOLMZUAtKplFVx1Zja1qRFNQfrZdeisTLYLu/cmnixGLXDTHlE7Vi627qLj+uM0+wAc1TG+u0LZeyV0enN6LTMqpnW9GTUxtnvLfXiFf90uEz7EJ1ivVQ/MdzjPLh8qeMRTMU5UuYJK2tk9sjBkamhopK4qSHTZqwpoaIi9QRF0g5qB2aKNe48nWqsk4XXFsnQRVbDqt8lSyeNkP4/jkw9lfQbIdPes45HpyeT2Zm6vfjEaFknY4uKpXJL9FgSASQSmboZPDdOQvkk6pdWjHOVTsH9eLCMXbRLWqQl0o5N2YvlzBurpCZuvKuX6hmj4to5zouL0lEn655w+H24VCY4U7otu5dK5Y1WWxtZJdX2FF2TXQ/FcuaVo2T2lricfMsbtq4T25K2CStZpw5wztD5NVLtlk/bhytiNamWdTKhuFgmvIfrPcY/qJ1rnMoPcU0py0jfFddnaPoZbPGU19q1a43IwZqf9evXm8+OHTu81LZt2+btRxwIIpzDJ8Q8RAXzJXcCqL3eCJ6O9y+UtiVFvp/Of/5JpD3xYoRA6tg4RNpX/7++5yHNzp0TuwXaK3veTxA/U/+5LJJ33YJEEXTkrQC79srCq3Hhh+xspUVqHu0vxVdWyrp99YJLvqWuRmbfHO/iTy6AiqW4uFgGP7pKdpkpgmqZfc9AKS4eLgs98dIi1ZP6S9GPh8q01bukrr5e9m5bGplKuKhKIqs6IgPFuSNnS/W+eqmv2yWrHh0sxT/Wu+dwtgaAie6ul1V3l0hRv1Eye9Nec+e0d1OVmS7s/2iNKb98s1SGF/eXUYtqpA537S31snf1BOlf1F8qt0RFUMYCKDdlbNlSKf2LfJhDOOj6LGNjsRT/eLBUvoey7pXqGcOluGiwjCgb6NVX3baFMgoew6d0bU0YzpGBobj4TBkxo9pMse7dNFtG9SuS4vHrIhyb6qV+9QQpKtKpoQi7ukWY4om1g7rdqyLtQM9rqZf67bPNGqvZ2yN3t9Ez4wVQdGqq5Nppsmp3XXSKt1IGF9vTZCHsTN5gfI/unT9UioonyLpocxBn8HVPqnnqXCmyp8zcCPbvuLTU/oGR9oiptWVoi8UyvGy4DLw7eq1gWu6yIikeuSoqcHC9nRmZAvfaeaR+YkKsTpbeXGxdC3Wya3VkihzrmHRNm2lrxdE2hPI21UnNi/BqDZelxgPaIuvGF0vRZZVePVTPHSX9iwbL7Fq7YD7fQ9d1UJuw0ozjJoI+C/0S+jZtH9Vzx8rA4mIZvgj9Y8Rur71Gk6pfNkKKzpkW7ZOUf5J2HhVJ/fsNlIH3aH1oHzhQqnJ4P43pLqz5SebZwTHEQVyGwiSQEwEET02QeEkQQ8v/JU7MdO17SdqW/4uv8HHPbX/nnLx6hU60t8i/vToyTgCd/7rj5dn1u3gR9Okwka5mn9qNXLBxd8k+sbxd5i7I527KuTtKLoCK5NxJ1ZFBz0u4XlaNLI4NhtumybnwFHmCKBoxOl0XsTd6p/xUtdRbXqEWHWRC2uqZ4PclyI5tC6Vq9V5ThponzvUpjwj2F+kAk6UAGpxVGSOekHMTnoKpl6VlrgAqkhHLbPG/V2ZfXCRFZUvjPAFmQFfhFIpzpJ25NrSsHSsxT49IkDCor9dKjVbSPiwqHxsTFHF8tSLjPUCmPq5dmODpbPlwQqStWdOTKe3ULEJsWzZFxOfw1yxPkzP4usmk9eBAXFp+nFtkXXmRNUhHc/sQYlO9bNhXL/XutBviXBz1Dm+qlOKiEbLU9uTgNHONqACKtKnE/iSy/1wjmqNCYuRSqbOq1btuXRju7xB1bU7xjWclFsctIoCKzqmUassmxK5fOUqKVbzWzpbBcQwifVDsmvHjLxLfzqNxHnL7wBapfujchGvNsjijr/D0QOD4PfqOfTiGOAyFSyBrAdS5Z2Yo8WKLmc5/jjGeIHh97P1hv3d+NjMvRMdtejlO/GD9z9pvnAbc8FG8AMKC6G+e9bEnTQ+Q6TRjT67oEyy61U4+uQCy1kJYFplOorjS3EnFDbBWHHytebQ4JizwtI1xteOJtaEy6omlskvH75C2OsnH/UxmRyRihJ+WP2EbLY/EdcaRDlDvmOMy9PuRdRmrE6YoNJu4gVZtjBOdURHheAjjzgvFOVLmCR9qztGtMwj5CiBn2iXG2BLiarsRMZqHLYCi7dyeDtNoUiOVxUUyaiUaTkg7vXOTf4H4gQdh8FNRb6FGd8ut+6PbbD1ALmdzPT7kLC13bajfJQvLdTrZvsYjIin4WrDbc6StxerITqco5tVqiuSFKeOSfgNlRPlsnylwB4r+TFnX0Yi+8TSRRLENRhGBZsXB17jpS4jJYvG8YhB/Ko7MaWHaT0AcnG+uJVuUOrZk+BMCB9NbbsA+ih+XSuH9zkoAtX66Ug499H9L26r/L7yQWf5DI34ggsIKHjte899K5eADxYK8cx2GvT0lTgCd9fwz8t/PNsnNM5vjPvfO/ULszxN/+9LXlLTWAJkLdELKp3S6TQBpiTpapG73uuhUWvSR4JC2ahJ+2+BOX2OHFJBxnbE9YGg6IbYZlzFNARQoImI2JgqgVG0ioMzuIOz+jk472FOQcNm37J4tA73XCojEC0y1M3MBlCBOE+zSPIK3geIHpyRNL+rZzGINkGu/fT16FsfZEMkzNv2DqcQWqbc8dMHXgl23kbbmCjAvz4QvLVK/r0aWPjFUSor6xxaJJ8SzdsRdS7rfruvoPt94Gj+xDsIJoKjHy0x5RTxZnhgySdssgvKKxPFllCcBhOktFTq4fnRKDPs49WXVU4F+zVgAnWo+Lgcf+H/k+3v+D18R1L72bOnY+F/S/v4FcUKnc8/TgkXOtqiJ+778h+Ycc+7as+PiqfhBnsgbNuQy1Bz+0hNA/zq7TPqN3ykDHjwoAyefSPr57LsufzNSPQXWsVcWTqqSGrjHTadSLGPXOn5iPClmvS/I7nDjBsvoHXZupsCwaHO2VMe57SOdy6jVLaFt9YcS3Wvc+z5TcbWrpGpRZCF09SSsZ0hcRL537cLYXW1cZxzQSfoakosyRgYHd1oHUx6JU2CWV8XY4zOwiEhcnYZqEwFljhuEdVCyxVTkPHdKpeap/lLkJ4D8vFdRr086U2CugEguWBIrLqn4QXS33FYSaT8FFpeWP2f7evSyss8zdajTWBojMi3tTZOFmgKLtJfi8ui6Lk0KaweXqXcWi9qrZFXcep/Ief1n7PLOCPyi11KSujbnarw4QW+lapcfvj+8uyzVFJg5HddNsYyaO1tGxU2H4aA///j6jsQ5t5umwGAVFjnv2bPHexJMnwDDPr4U0WoTBfo1YwFUP+c6I34gRmwRBMGTsNAZC6T/OUYginAsaL1Q586HEzBhgTTStMWP5lk/9/qE+NnumLB5gRFBP3l8ify6/GsZ8MB3ctGkhkAB9Py7bcmz9N4DNEKqliW+Bwh34JGnqyILXouKB8rYuZHFrN6iVLwrKJqL3eHGDZamgzhTzj2nxFtUW79PFwCmWAS9aaFMwPtUrEXQC7H4Om7Ror2YMpytycH4LYKOLA7FawOMDDSLoIvEXlwbWdTZXyo3RYViXGcc0En6GrJXclHGyCLoEhn6RPzC85LiYmcRdAYCKLowPnmbCCizMwiJWWNxroxduVfq6+qkviW6YLbfBFmKhctY6A5vARar2gLIPIlTJENn1khd/V6pM+tUHPHmswi6etEE30XQ2Qigli3TZGCxvoAvcretd93eUiZTbl3sjTiZvwfIb3B17bevR6+ZxbGPeoBuroo8UBC9Jk378NYJRRZBm3eFmYXS0UXQP0VdxMSTLrj3FvjqwwneQw7RBxy8BwuQTmQxeuRJKs9C/y9h6hpnxl1z+I2HFazF+3HljwigM885V0ri+hN7EXTMnJb3JpgyJ95UhGnnkTjFxSV5XwQNi0+cOGGe8MJaHwghfeQd37EPYghxGAqXQEYCCJ4XFSH2tvH10clL2l4vZtGzz1NiXd++nvTcxqWjffPMtReopbNN/tdLI+XX9+6T39x/QAZM+F4unHjMVwBdN+OktFmLhAMLEPQm6EXV8Y+e43HaZZUS9Fgy0rc73EQBVCSRR0Bjj1Vn/Bh8dC1B5DF4PE47VhZu10VAsCS1rYE8vAPhH4MPtCOuMw7oJL38nC85KmPcY/DRR6N34ekkXcwcZ6Pa4IiI6O74OsXOVJwDyuwMQiaduSMij9sXj5ClWDfslv/GSln3of1m7YhR9WsrZbB5xLtEppnHzX1sD/kYvCsg4kWGsvHfGjZxj+XH1sF4nixT7th+rJnJ9E3Q8bb5c7avR89ql7399mq8omDkbKlZjTdWW2tS4tZj4VF6vL5iXcKbzc1j8PoYvu9rLuxHwFF293UEnpW+X0LVtdueD6+SUT8uloEzo14mp/zKqP7Dad4rPgLtMgv//Z5a8+fvW0erI6+EsF/hkavH4G1oBw4cMELHfgRej2MfRBDiMBQugYwEUFvt3xPEyOFJpaGmpPzW/mC6K1WA0EEetuDCd9iS6/CPA9/Ld0c7pe5Yp3xf3yUHj3fJoYZTvp9c5515egEdROYJ8syQBPyesjHTQgnTFSETZLS+R6ADbxNyghEDA1M/wu6cVmg/VQCFsQtTsInTfGHO7P448Pgk8/DgGOIwFC6BjARQ05rKBCGCfWGCuyYI63+wJihMyCbfMOn37DgUQKej/upXjpBivJtFpzfrdsk68+6VkAtPT4fRzLOwCHTgf6vhX3zMlnXmPUrWu7n0dQ+FZXFa1oQWQPAkFfs/yZpWhoxMAiEJZCSA3PU/8MSEfSqrbfkP4xY2QwCFDcjD9QDBFgYQKCQBFLEl8JHd6D9zzWe9JZsuiVvkmwMj6t6bZv2j29y+eTYH5vWYJMxAGTDFFTddlIsSmWmc+Kkyu716U2q5yCtMGuZR+aHem7T93gIfJplkcbqVr2VIWAG098XBOX9Xj2UGv5JAAoHw6sM6lQLIgsGvvgRavH9G6SxWxf64p8t8T89+J95sm8SGhOmG7HNkCtkSwBurA+ss1zXWkiSvevEWVWdbpkI6v1v5FlLBaQsJ+BPISABlMxXFKTD/iuBeEiABEiABEiCB7iOQkQDyXQT9yJlZLIIekrLEZhH0I2cmTIHlYxF0SmMYgQRIgARIgARIoEcTyEgA5eMx+M69C5OCbFj0+wTxg/VAuX4MPqkRPEgCJEACJEACJNArCGQkgFBydx3QoQf/L2lb/W+C/9zu/yLEPwn+mal5EeKqH8UthNaXHDa+fl+CoOk8+pUce+63vuKHC6B7RRtkIUiABEiABEig2wlkLIDgeTk4vtgIExU/9r+0gNjBPzuFILL3u/8KQ8WP/XQXBA/e8nxkWj9f4YO4yJven25vL8yQBEiABEiABHoFgYwFEEqP9Td+4scWPAnfl/+L8QLhhYh+4scWQsm+c+1Pr2h/LAQJkAAJkAAJnBYCWQkgWNy5Z2achydB8Pj824uOj281IqjhlVsDPTzJxM/JD549LbCYKQmQAAmQAAmQQO8gkLUAAoZT9duk3VnXEyiElv9Q7H96ipcb6lRaMtGDY0em/lI6DnzSO8izFCRAAiRAAiRAAqeNQE4EkLEe//F9z9PmP7f7ih8VPu32P9SMlBsLnbGg2e9/fUH4HKv6rYT9VxunjSQzJgESIAESIAES6DEEcieAnCLj/3vpJ+GpMCeu+xPre+AZorfHJcPfJEACJEACJEACuSCQNwGUC+NOdxqHvqiVhk5YsVUml5TJ4rrTbNGWyVJStljM/xfubJDaL7rxPw13d36nGXWPyr6hVmpPd9vME7CtU0qkbEk3tvOQ5Tjddm1fUC5T3zlorLW/hzL/4LsytXyebA8VuW9HOvjOVClfQFK9tRX0cQEEYVMiJe4HIqNhvYw76wqZ9QWqPn8C6NCSssT8S0pk8hafJmcLoC9myRVnjZP1DYh3SBaXBZzjk0zirgZZMcp/oKl9+QopGbVCGuLyS0wh/3vyVwf+tmfLVETqFktZXNs6Wy6+uUIWf97qn2XYvXY7EJHaOVdI6fj1YppC2DRyHA+CIOE6ipY9GwGTN6Fh6sbnpsbsnyxbU/DJm11Ovo1/ny7lFUtkd1f8AVv02N/jYwX8ogAKAJO4mwIokUlv2kMBFCQ24mo5f4OvEUDq1YnL0+eHM/DFYmQ/WDesGiMlQ+fL/liiIrJDqn5TKhUfZjlgx6WZ6Y/81YG/RdkzjQgga5DtbJX9aypkwFkVUp0N0sB24F+S7t6bS3GQy7TiOPQIAXRA3nh0kkyqLJc5WzvizLdFj/09LlLQDwqgIDIJ+ymAEpD0qh0UQIECyB5w7e9wuFRL1e/Ok9KSEintXyazPs18NEsqgDoPyfopw+TsaD5VU+6KTYHpnarZWnffU1Lduwa039ZqqThrmMz/yjr+aZUM+E2V7MA0oOYXPdz6+WKpGHq2ues/e2iFrPlGRI6tkLtKKmS94thZJQPOmixbzTSiiGyZLKVhxZ5lRuSrVQdRWxavqpBh/14iJWedJ2Uv7BDN9lB1lZT1L423zSTSKrVLoueUnC3DHlwj+41tEbEzbsYsGf7vJVI2e3a85yZTpsZOSwAZG2LlSBjcLWFjjs1ZIbOi7UwZY3/M0xJJ27ShqI04PvypxV77NOft3CGLH4y1o1h7DeKRAD+tHQnlwtlx18wwqVhpSe0kx+LSsuKV/PswqVhlpZGWhdqe3brR/eoBCuYTsyvWdiJtrlTOGz1fak+ma5BP/N1LpKLyDandMkfKn/lAGq0otuixv1tRRKRZdi+bLhXjy6W8fLxMmvOBHICOigqgNRvmyCQcGzdepi7bLc3Rk5v3vStzKsdLeXm5lD84VV7fHT2yfZ6UT31dPlg2VcaPK5fy8ZNkzobINBxOPbptiUz18loi8x4rl3k6e3SsRpY86ZNm81fy7pxJMh55OXbElcU+3y/fB1HGchlfOUc2Yjr4+AcyvXyObG7TVBrlg2dUSDbLV+9Ey264bJSD6mFr/kreeDZiJ9JaNGcSp8AUYS/cUgClK4A6d0jVoFIpezXS+TbUVMmwLO7okwmgyLFZsgNzG50NUj3l4kQBZBplDrwVmOibUipXvFzrNfMdMwbIgBk7Ir9tAXRsjYw562KZXB2ZdDm0coyUGu8R7CiVaTWRU8z0WckAqdoZ+b1j5oAs1nPEhENEjJVK2cs7Imu0vpovZbpGq2GNjCkp84Tc/lfLoraJwMtVOmiyGLM7D8mK/ymVYa+gHiP8Sn83X3bUNUiDUVI5YOoKoM5WqYU9gyKiMjaIRpE7AqhkUIWsR2eOun9kgJSqELPi4UxXAJWUWW3mkQFSAh5fRORhw5oKKb1qvqCWg3lE7clwk1Au75qplVYjptdLxaABMnlLq0iyY6ZN6tRsg6wZXSJlpr5E5KvFUuYK9nTsdetGz7XaeTI+sTJG2smAB9fIIZStc78s/l1p7LrRdDPY7l5cIVPePCDSViNzxk2Vd2NaQ2zRY3+3szn43nQZ//QbUgvl1NEoNQsqpGLx7qgAKpfpK2ulsUuk49s1MrU8mr7Jq0LmbYvIrcat86Ri3DzZDoEAAVReIXM+OirQUXHHjOCYIkt2Rs/bNk8qyqMCqKtWlkwsl+lvH4ict3OJTBk3R2raRGr/ViEV82qMHdJYI/MmqkixS9IoG58rl6mrI+fD3unjpsgb34rIt2/IlHHTZc23sKhDDqyeKuVT35WDEjlnzpao56xxo1RF82zcUCXlE+dJjeFyVD54rjzCWTpkOxjN2ShHcVrbAVnzdDkFkF0Vvew7BVDcGg3cXetdoTXg2muAaqZ5A0ikLWQ3UJrBy7XBeEkOyYrflUiF506JeFC8RdBWR60DuO+6oXQarHp8cE7nVplsDzBWfoeWlUnJg+s9j4u9RgqiJyKiamX+VRWyYmWFDJgJEQVOV8h8s6YqHaM0rlUfli2Ro1YdGE/WxVKxslYOYczvbJWGY/iSjGf0/GrNC1srTXt3Ot+NnbbHpkRKr50m1dE1vbFBNJqoJWySHYMnzWsHPgIobs1NdXzcmCcvGY90CpkYN8H2hGtGpGHlXZE2lOxYnABqleqJpXLxxBVSaypWpLWhISKoEk1IvcenbmKeNXiAkvOJldGnnXwxX66A5zS1FcExjBCJDvLSITUv6SAdOcUWPfb3WIJH5d2ny2XettgeOf6V1Ow+KB0JU2AH5d2pKlY6pLmx2QgVc2ZXjcxTcWQ8QBAXGrZ7x46+N13Kn9toealwLJrmzkVSXvmGxJzLsfwOvDlFyp9+XbZ/G8mzo7lRmj2vjeYTKX/FSx/IV8cigqbjRKM0G83TLI0noiIH0evWyJToAu8OeM7mbI6ILoiel2qkQyJcPGGEc7xy7ZZF4ybJG19rviKcAoux6I3fKIDS9QBh8HEFS0mJlC3L7EmVYA+QT8dqD3xxIsAnbkatFaJlgFR9KtL6YYXnOTFJWfn5iraSqOcH014QcF/Nl2Gj10gDPDLwOGCb1aAQUgDB2LpqmfU/w+Xify8RTAFFFh1HGCXU3VnTzGCXuIg8B0wNMxXUUY+LNQUYG0SjlWXVb7JjaQkgK834ekzGI2pPhpuktmuaapdudT+21r64tDoPSfULY2T4oLPFTIEtqbVEuJ1AiO9O3XhneO08OZ+YXT7txEvDSzXtL2bx86NvyAE9E4O0tRjaFj32d40usluWjLemoGIHvCkwnZ0SiQkSRMNU1vSJ0ekqTE2FEECJQsESQMZzFJmiMtNqJs1ymf7eUZGuRqn9+zyZPml8ZApscY0c1eko2+aOo1KzrEqmwi5Mgb3zVXTKLjKdNSU6BRZJP/qEG0TkeHiaOmTzHGURKatth/k+bpHslpig06wTy6VHuO0NBCiA0hVAuGMdvthZLJx5U0gmgLrdAyQi+18ZJgNmVMv6iaVy10rruSKrUzceoEfi3CUWACycHiNVM8ui52PqYpjMf2WynK1TOFbs8F9DCiB4fCJzWCbphvWTZUBU5IDnZF+zfQaxnHmAYgIoMt0TnfqJ825EKQQN/DhsHYv7nsoDZJ+HdLx6jHg4/HlE7clwExMH0QSCvDwQyMmOOYxaj1ken2PrZfJvYtOtaZtqOFh1owmE5BMro0/bydoDFDBIl8emh2zRY3/XYkjU0xHnAeqIekuSeYDMlFKVfFCnXhVLFHieEs0ldsxMK0W9LZGjlgCCB+ixNZbnSM8X8Tw52IUpp6lRYRSLYqa2jFcqKow6jm+WeRXlsminSMfWOVL+6BLZfjxqb1zZ4DkaL3M2fCBzPPHo4xnz8qIHyEPRR75QAKUrgFq3yrRBF8u4ldH1DK21smJKVWRdSQaNJlgARdd22Os5UqwBGvduZJ1HBmbETsFC5t8MkAH2YmYc9QYGfF8hZf8+TKq2RAXSsa0ya8pi2R/NfuuUs6X0rNjggumO0rNK46fzYjmG/BZSAMHzdNYYWRF9L07DhyqARCDczr62SrYei2TZsGWWTF6yX1p9xU5kYMuKqc8ga9aVRJ+207o3C7Fb98uaB8kqVHUAACAASURBVAd4U1uxATaKxxYy+H7VLKmNLi436ViLoOOmwOzzkJRVj8E8QlZJQLQE2911PseqZfKg6NOFyY7FCaD9Mn9oqYxZGfW0NlTnWQAlay9YL6drkyLt5OKJ6+PWAF1spn0DAKXaDRFSPl0+OB4fEetlItM4ksYaoDVyAFNKXdE1QK/uTj4FtmeJVFTMk80QFF0dcvCjOVIRwgMkx7HGJmANUNtuWTSxQl74+wHpgIhpOyAfLFgiNY3Rqb3FuyNrgDqCBBCehiuXqvcja4/MWqGoADLCa2q0jB2NUrtyupTb7zjaPs8s2DZrn6I4zdqox1Q0dUjj9tdl3jtYX8Q1QPEtrvf/ogBKVwChTbhPo2ThijeDl9+UGl7+Bpf/jOHmKTC4/Ce/PE2G6xSKNZAZk1aOkfPOKpHSpzJ8Csxr65HFpqUTq+OnF5z87KfA8CRclS5sEZHW9RXxj9Sbc8fIGsuh5GUX+ktIAQQv1ruTrafAxsh87yk9+6meUjnvd1XR9Tg+d/Go5myZmnLHhKApanTAN2u7Wmtl8WjracIZ48IJoNatUmWewCuTFYcSF0GHFUAiQTxCV4pvxJg4sA5b10xp/wyfAvtmvUyOPhWH62HMK7En/6ycwn31qxucGdfOg/nEyhgVyjNmRdtcqVwxfnFWT4Fh8XPFgu2xdThaIrPQOCKMbK+P/V2jRrbWU2DjxsuUOdGnneK8JIhpT4HhnOhTXuPGS9Wbb8gcXReTxAOEVBp3vi5TzVQUnjh7QabqGiAcdJ/i0ims5t3yxnPWU2ALNseeyIoUIvK3brPM06fIxk+SqjejT611HZSN+hTZ+Eky780lMtV++qtrtyypqJAle+zE7KfAymX8k0ukJnpTJHgq7cVJ0SfgquSN5S9wEbSNrpd97+MCqJfVJotDAiTQzQT8xXM3G1EY2cWt3cEU2BRZ00vfUF4YwGlFtgQogLIlyPNJgAT6MIHcC6Bx48ZFPBDRxcIJC3YLcL+c2CwvTJojm4+YR7PkwNvTpXziEqntkh5XFj/efbiB9+qiUwD16upl4UiABPJLIPcCKL/25i918yJEfSHho9EXEuYvO6ZMAlkToADKGiETIAESIAESIAES6GkEKIB6Wo3RXhIgARIgARIggawJUABljZAJkAAJkAAJkAAJ9DQCFEA9rcZoLwmQAAmQAAmQQNYEKICyRsgESIAESIAESIAEehoBCqCeVmO0lwRIgARIgARIIGsCFEBZI2QCJEACJEACJEACPY1A0Z49e4QfMmAbYBtgG2AbYBtgG+hLbaAgPEAAzkACJEACJEACJEAC3UWAAqi7SDMfEiABEiABEiCBgiFAAVQwVUFDSIAESIAESIAEuosABVB3kWY+JEACJEACJEACBUOAAqhgqoKGkAAJkAAJkAAJdBcBCqDuIs18SIAESIAESIAECoYABVDBVAUNIQESIAESIAES6C4CFEDdRZr5kAAJkAAJkAAJFAwBCqCCqQoaQgIkQAIkQAIk0F0EKIC6izTzIQESIAESIAESKBgCFEB5qIqWlhYZP368fPzxx3lInUmSAAmQAAmQAAlkS6BXC6Ddu3enxefbb7+VcePGySWXXGI+kyZNkkOHDqWVBiKfLgF0+PBhwSdMgDgbOHCg7wfiDWVINzz99NOydOnSdE9jfBIgARIgARLodgK9WgCVlZWFBgqh88c//lFWrlwpHR0d0traKgsXLpR7771XTp48GTodRDxdAmj58uWCT7oBYihT0WPnRQFk0+B3EiABEiCBQiZAARStHXguJk+ebMSPVtjRo0fljjvukE8++cTs2rFjh9x5553Ga3LzzTfLli1b5NSpU+bY9u3bBfvgPXr44YdlxIgR3hQYPEsQUoMGDZIbb7xRtm7dqlnkdJtLAQQBOHv2bLnsssvMZ/78+dLe3m48PA899JAReRCKjz/+uDzzzDNy2223ed6kXIipnIJhYiRAAiRAAiTgEKAAigJJ5b3Yv3+/3HrrrfL3v/9durq6ZNu2bUbkfPHFF2baaeTIkfL222+bY7t27ZJrr73WCKDjx48bzxIEFs6DtwUi6uDBg05VZP8zVwIIog7i54EHHpCmpiYzDTh69Gj58MMPjTcMAuf999+XTZs2mbIdOXLEGJ+KYfYlZAokQAIkQAIkkBsCvUoAYc0Ppr2CPkFrgnTKKtn6lRUrVshjjz0mnZ2dhjxEwgsvvCALFiwwHp177rlHTpw4YY5pehA78PYEHctFFQaVVfeHycOdAquvr5dRo0YJhJwGsIHAQdi5c6cRg//93//tebmwnwJIaXFLAiRAAiRQ6AR6lQByYUMEhA2pBm9bAGiaus8VELYAwjG/xcYQVLkOufIAHTt2LG5KS+2fMGGCWRsF8Td16lTBVBimyjSkYqjxuCUBEiABEiCB002AAihaA6+++mrSNUBBHqAXX3wxqZcHHqD7779fmpub817XuRJA8ABhyqu2ttbX5n/+859y++23mynADRs2eHEogDwU/EICJEACJFDgBCiAohWkT4GtXbs27ikwrIOBR8ddA4TpIawJ+uyzz5KuAcL6GIgJfboMa4Jeeuklqaury3nTyJUAwjTfzJkz5dFHHxXYi8XOq1evNmt+8BvlgfCBEMKTc/qqAAigefPmeQvDc15AJkgCJEACJEACOSLQqwVQ0JqfIHb2e4Dw9NMTTzwh8IZoSPYUGI7hyS886YWpIQgnTH8h2E+BDRkyxDyqDlGR65DOe4DsvN0pPByznwLDk2142gtTY3g1AN6P1NbWZtZDPfvss1JVVWVE0j/+8Q9B+XSqzM6D30mABEiABEigkAiEFkB4F055ebksWrTI2K/ehlTbMIXds2dPmGiMQwIkQAIkQAIkQAI5IRBaAMG7gEXFeBIKAdswnzBWUgCFocQ4JEACJEACJEACuSIQWgAhQ4igdN+KHMZQCqAwlBiHBEiABEiABEggVwTSEkC5ytRNhwLIJcLfJEACJEACJEAC+SSQlgDCE1L6bxywwBhPQKX6hDGeAigMJcYhARIgARIgARLIFYHQAkjfsqxrgPRNw6m2YQylAApDiXFIgARIgARIgARyRSC0AEKGrgcIoijVJ4yhFEBhKDEOCZAACZAACZBArgikJYBylambDgWQS4S/SYAESIAESIAE8kkgLQH09ddf8ymwfNYG0yYBEiABEiABEugWAqEFkL4HCG8ERsBaIHxP9QlTCnqAwlBiHBIgARIgARIggVwRCC2A8P6fe++91/wbB2Se6g3QejyMoRRAYSgxDgmQAAmQAAmQQK4IhBZAucrQLx0IIH7IgG2AbYBtgG2AbYBtoLvaQMEIID9hxH0kQAIkQAIkQAIkkA8CFED5oMo0SYAESIAESIAECpoABVBBVw+NIwESIAESIAESyAcBCqB8UGWaJEACJEACJEACBU2AAqigq4fGkQAJkAAJkAAJ5IMABVA+qDJNEiABEiABEiCBgiZAAVTQ1UPjSIAESIAESIAE8kGAAigfVJkmCZAACZAACZBAQROgACro6qFxJEACJEACJEAC+SBAAZQPqkyTBEiABEiABEigoAlQABV09dA4EiABEiABEiCBfBCgAMoHVaZJAiRAAiRAAiRQ0AQogAq6emgcCZAACZAACZBAPghQAOWDKtMkARIgARIgARIoaAIUQAVdPTSOBEiABEiABEggHwQogPJBlWmSAAmQAAmQAAkUNAEKoIKuHhpHAiRAAiRAAiSQDwIUQPmgyjRJgARIgARIgAQKmgAFUEFXD40jARIgARIgARLIBwEKoHxQZZokQAIkQAIkQAIFTYACqKCrh8aRAAmQAAmQAAnkgwAFUD6oMk0SIAESIAESIIGCJkABVNDVQ+NIgARIgARIgATyQYACKB9UmSYJkAAJkAAJkEBBE6AAKujqoXEkQAIkQAIkQAL5IEABlA+qTJMESIAESIAESKCgCVAAFXT10DgSIAESIAESIIF8EKAAygdVpkkCJEACJEACJFDQBDIWQIcPH5bPPvvMfLIt4Z49e7JNgueTAAmQAAmQAAmQQGgCaQsgCJ/HH39cysrKvM+oUaNkw4YNoTN1I1IAuUT4mwRIgARIgARIIJ8E0hJAJ0+elPLycvnDH/5gBA/E0O7du2XGjBlGDGUqgiiA8lnFTJsESIAESIAESMAlkJYAWr58uRE6X3/9tZuOEUHwBGUSKIAyocZzSIAESIAESIAEMiWQlgB67LHHjNDxywyeIEyLYZtuoABKlxjjkwAJkAAJkAAJZEMgbQG0aNEi3/wwHUYB5IuGO0mABEiABEiABAqMQFoCCGt9sAbIL2D9DwQQ1gmlG+gBSpcY45MACZAACZAACWRDIC0BpNNcs2fPjssTa4Kw/sfdHxcpyQ8KoCRweIgESIAESIAESCDnBNISQMhdPT3wBEHw6CPx9957r1kf5LdAOpXVFECpCPE4CZAACZAACZBALgmkLYCQOTxBED+6KFoff8fj8fAEpSuCKIByWaVMiwRIgARIgARIIBWBjARQUKIQPpmIIAqgIKLcTwIkQAIkQAIkkA8CORVAMDATEUQBlI+qZZokQAIkQAIkQAJBBHIugJARRBCmx8I+EUYBFFQ93E8CJEACJEACJJAPAnkRQOka2tsFUG1trZx//vlSUlIiV199tRw7dswgwha/sR/HEY+BBEiABEiABEgg/wT6vACCl2r+/PkybNgwOeOMM4wYGTBggDzyyCPy+eefy6lTp7KuBQqgRIS2+IMAdD/nnXee3HffffLxxx9LV1dXYgLduMe21RWqnZ2dsn37diNe82kn0kY7Ql7Ik4EESIAESCA7An1aAH3yySdyySWXJAy+OhhDEL344otZDzgUQImN1BYVyjto++c//zn0dGpiTtnvsW21BRCESGVlpWk/aCtLly7NPrOAFBYsWOC102eeeSbrNhmQDXeTAAmQQJ8h0GcF0BdffCEDBw70BpXS0lK54oorZMiQIQLvgw7GGNiWLFmSlSeIAijxerJFBVgPGjTIsAd/fOCF0zrA9nQO+rattgBqaGiQm2++2bMT78TKNDQ3N8uYMWO8tF577TUvqfb2dhk/frx3DPEQn4EESIAESCBzAn1SAOHOfdKkSd6AcsMNN8h3333nUWxrazPTYjoldtlll8k333zjHU/3CwVQIjFbVEDgfPTRR3GRUAdPPvmkV0cQq+m+XyouwSx+2LbaAgjTo2+88YYRzFdeeaVks5YtmQCC6Vu2bJELL7xQfvWrX0l1dXUWpeGpJEACJEACINAnBZAtSH7605/K1q1bE1pDfX29DB8+3BuAV69enRAn7A47Py6CjlCzRYWfAEKsuro6ufzyy706eO+998Iiz2k821ZbAOUyk1QCKJd5MS0SIAESIIE+KoAgZnR6BVMYmMpwA7xE06dP96Zl5syZExcFHoqVK1d6i6fhLcJCauzDMTukK4DgDVH7Ro8eLZs2bZJbbrnFLNLGVN3IkSPNYli/BdpYLIuFw3/84x/lnHPOMelgOumJJ54I9GIdPnxYnnrqKW/aScuCxeFBrzKAN+aBBx7wpgsxbYjfYb00tqgIEkBuHJ0WsvlgOggLg5UPvEZ2SMdO8MS6sNtvv13AGZ9x48bJp59+Gvi0Hqa9tK7UPs0/qC7AGsw14DxNw91qmnYcv6k2sHrhhRcE3kqkAdvB5N13301ojza/sWPHGs8Vpti0vcCbhXbMxdZaQ9ySAAn0RgJ90gNkT61gAGhtbU2rbg8ePGgGF3ew0t9I0xYO2Qign/zkJ97TaZo+thisNm7cGGc38sSCYTue/R3nrF27Nm49E6Zt7LVQdnx8x/QgyqsBIgGLfTHAunHx288uPdfeuuLGnQJDXIgX2zb1ANkDONZt6aCP/FUAZWInxIKKALtsP//5z73yuh6gIAEE8YB1SzqNaqeH70gHQhXBFjduvDACCOkgPfdc/Q0RDd4abH79+/eXX/ziFwnnwu58LupWW7glARIggdNFoE8KIHvQ8rubTlYZ8O5MmDDBGzAwYD744IPmYw+ezz//vCc0shFAGMSQ7j333JOQx/XXXy+HDh0y5mLAf/nllz27MIDBk4HyXXrppd5+DJR4vB8BZcGj5jpQwgvwwQcfyFtvvWU8X7ofTzqpN6CmpiZOJFx33XUmD2w1Pp6sS+UJSiWAvv/+e4GQ1DTtNUD2AK7HdauCIV07sQYMYkrTwVobvAoBHiZb7IUVQKtWrfLET1BdwEODqdb9+/cLxJ095froo4+afzyMYwi2SLLbbFi74QHUOnT5BbXhIO+oMYh/SIAESKCHE6AASvPJnW3btgnWDWGgtMUE2gGEhd6JYzrs6NGjpnlkI4CS5QEb1CsCwXDVVVcZuzDg2k+uwTNkiwkVNLYIwYBvL+L9xz/+4U3/QSQ1NTWJ+zSSPahicMVvFRAqRIKuDztvPSfZFuIOIg/BHcAh3Pbu3Wu8HFhLk4md9rQohJC96B2eNhW3YQTQiRMn5I477vBYLF682LMd9XTNNdeYY2hHaE8IqdYABQkg+/F411tn292vXz/zT4xdfm77wno4bd8XXHCB4WoM5B8SIAES6GUEKIDSFEBz5871BraJEyd6d9VoFxAad999tzmOaYUdO3aY5pKNALIXTSMxCA3kq2LhueeeM3ngKSEIH+y3xZc5GBUNehwDMDwPsPeuu+7y0kJeeKoJXgX1Fuj52NqLku3yaRwMuGqXy0bj6DasAIL3BeuvbHtsAQTRd+TIEU3WbDOx0/YKQsip2EKCtq1hBBDEGMQDWFx88cXGw2MbuGbNGnn66afN58svvzSHMhFAmLq1ha0rOt00ly1bZvKy+bnty2bnltUuA7+TAAmQQE8n0CcFUDZrgOyBUgf7oC0GGoRcCiCkZ4swnQ6xPRh+74kJsgF3/Bjo3DJAeGABsD2VZafhxnd/+9lgXyy2qMC59nuAINAw/YQy6RSffW6yAdzl7drl/oadEIPJhIRtqysK7PagAsT2EroCwy6H/d0VK5qWxvHzALnvIdL2pudg62dfMn7Jymqny+8kQAIk0NMJ9EkBZIuFoHUOmEbBNJLeqb/zzjumru0BxR1M3d86INnCwR4QgwabZAMUjMhWAGGKx17YDDvwBBGmwdwyYOpHF1vb5XDjub/TFUDKKswFlYpPunb2FQGE6TKEZPyC2mSYemEcEiABEuhJBPqkANq9e7dgTQQGbUzl4NFnN2BQtBelQjQhYMpJB/uKigqzzgeDht8HIgrBHpCzFUCZToHZ01M6BWaMs/5g2gflfv/998Ve1KxiBgtyMaWD8oMfpt38yo198E7Y00hWNuYr4oCFssylAMrETlvY6pNkarNtaxgPUKopMKwRUm7aRjLxAKWaArOnZME5zBRYsrIqD25JgARIoDcQ6JMCyH2SCwtWsThVA46/9NJL3uBsP4FkLxLFfl3Eap8LYWCvWclGAGHA1ae2kIe90BpreiBWEDJZBA3b9V9PTJkyJe59MbZIVNHmPjWGc1paWrToZrtv3764abO4g9YPe6DF4JxLAZSJnbZXEB4y+83gEMgqmMMIoGSLoCFK8Fg6yoz6Q1tBcAXQq6++atEKfgos20XQWreamV0vblk1DrckQAIk0BsI9EkBhIpz33+DNS8Y+Nz/BYaByn4CCQNVeXm5J45wHgQUpsrwODxeCIiBzf4nqtkIIOSvjyljXYw9TQUPFTw2CPC2hHkMHgM5XhyI4IqmqVOnmgEZYsT+31P4rp4KHNMnomAbRCDsmjZtmvGYoexgqIt7TUY+f+yBFunkUgAhu3TttFnAnlw+Bo82Ai8avEz2Kwn0MXjYi/rDo+/IGx+tczxBhuC3Bgj7wz4Gj3/9oqIcbDQfCiCDl39IgAT6IIE+K4BQ16leIIfBHG+DhkfBDnjqyH7MWQcT3eZSACEtvAxR09atvTZHbYN3Id0XIWJqzP7nr5q+bvFOH/vxeAzUeEuwLYI0rm4LQQBlYidY+JXLrgPXK2JPndkLlyE2wr4IUesP3jzkpRyxRfoIQQIIx1K142QvQqQAUvrckgAJ9DUCfVoAobLhicACYLz+Xwcf/OsIeDUw3YSB1C9AFOHNwbiLxx0+Biv9dxDuedl4gDBAbd682fz7C+SDD/4Vxs6dO/3MEvz7BQzkeLxdB3Mtj/1El32y37/CAA88fg5R5Rfwnhw8Lq7/tR3s8Pg9Fo4HnWOnk28PkOaVjp2oa3jHwNdmjWkqeHBQx2EFEPIP+68w1FbEx0sowR55of7wQk2EZAIIx1GHzz77rPdWbNiPtonXGrgCnh4gJc4tCZBAXybQ5wVQIVZ+sgGqEO2lTSRAAiRAAiTQ0whQABVgjVEAFWCl0CQSIAESIIFeRYACqACrkwKoACuFJpEACZAACfQqAhRABVidFEAFWCk0iQRIgARIoFcRoAAqwOqkACrASqFJJEACJEACvYoABVCvqk4WhgRIgARIgARIIAwBCqAwlBiHBEiABEiABEigVxGgAOpV1cnCkAAJkAAJkAAJhCFAARSGEuOQAAmQAAmQAAn0KgIUQL2qOlkYEiABEiABEiCBMAQogMJQYhwSIAESIAESIIFeRYACqFdVJwtDAiRAAiRAAiQQhgAFUBhKjEMCJEACJEACJNCrCGQsgBYtWiTufxfHb+xPN+zZsyfdUxifBEiABEiABEiABDImkJEAmj17tpSVlcnWrVvjMt6wYYPZj+PpBAqgdGgxLgmQAAmQAAmQQLYE0hZAKn4gdvxCJiKIAsiPJPeRAAmQAAmQAAnki0BaAiiV+FEj0xVBFEBKjlsSIAESIAESIIHuIBBaAKUrasKKJRSSAqg7qpp5kAAJkAAJkAAJKIHQAggnhBU16YolCiCtDm5JgARIgARIgAS6g0BaAggGpRJB6YofpEkB1B1VzTxIgARIgARIgASUQNoCCCeqCHKfAsNvPB2G4+kECqB0aDEuCZAACZAACZBAtgQyEkDIFO/7OXz4cFz+fA9QHA7+IAESIAESIAESKFACGQugXJaHHqBc0mRaJEACJEACJEACqQhQAKUixOMkQAIkQAIkQAK9jgAFUK+rUhaIBEiABEiABEggFQEKoFSEeJwESIAESIAESKDXEaAA6nVVygKRAAmQAAmQAAmkIkABlIoQj5MACZAACZAACfQ6AhRAva5KRY4dOyZXX321fPTRR6Z0Bw8elN27d8upU6fyXtrXXntNxowZI83NzXnJq7OzUz799FNpaGgITP/xxx8X2JHPsGXLFrn00kvl/PPPl9ra2nxmFSptlBkfBNgzZMiQbrEL9Yz61rYWytgcRerOcubI5F6bzFdffSX4MJBATyLQpwXQyZMn5a677pKqqqpuEQfd1TBcAbRixQq5/fbbpbGxMaUJGMggnpBGJiHfAqi+vl6GDx8ua9asMeb5DYL5FkAQX7fccou8+uqr0tbWJhBlpztQAJ3uGui5+dttJ0wp/K656dOny0MPPSTt7e1hkmAcEigIAn1aAG3btk0GDBgg11xzjXz//fcFUSG5MMIVQOmkWegCyC2LX2ecbwEEvnjjOfIulGAPYn5M8mUnPUD5Itt96dptJ0yu3dm+wtjDOCSQKYE+K4AwHQTPz6xZs+S+++6T9957LylD3OXPmTNHzjnnHPnVr34lf/nLX+SKK67wXP9I7+233zbTImeccYaMHj3aE1U6YK5cudJ4L/T4kSNHvDzxfcKECVJaWmrSR9yuri5zHF6VBx54wHyGDRtmvDMHDhwweSD+eeedJ3PnzvU8Ea4Acr0ymL658sorpaSkxGw3b95sPGDoCLFPP37TGuCwdOlSGThwoKAc8IR8+eWXXjncvFAusEBceG6ef/550TLgpKamJsHdI7jig+/wzGmATU8//bSMGDHCTLV8++233vQe8lJbscVvBJyDeq2srDQ8MVWFMmvA8RkzZnh24fjHH38sb775pmEPpo8++micHXouOn9Me2m+Ot2Hunrrrbe8+r/zzjtl3759epqxza1D72D0C+pt0qRJxmawAKuWlhYvWk1NjVx33XUmb7C0XyCKMuGDkGqA2r59u6k31AnqEeXW6VEwnDhxYlxbt48jnnKCjVOnTjXeRb+2AlvC1IXWm5/tyZi45US7sdsSrld46DRdiFakp8Fmhn1B14XGt7fJ4iazA2ngekH9gf/IkSNl2rRp3rQxynTrrbfKSy+9FNcW8ZZ9vY7c9ow0g9oGyoty+/U9OAZvr7Zl2/Mb1EaSXXPa/mAPbijt/uyVV16J65+CbLIZ8zsJ5JtAnxVAuEDR0aDDef/99+X+++/3Oks/6KtWrTJiYceOHWbQfvnll00Hph3/Bx98YAaVb775xgxa6Hx1cNSORkXR0aNH5e6775bnnnvOZIUOE3ExpQIX8hdffCHXX3+9bNy40RxHp4PBBlNZSOv48eNm0EF8CJLvvvvOdKgoB4Lmp7bhfLUF9kGArF+/3pyLLYQcOlgEnGN3hGan9QdTT+i8kQ4G/cWLF8uoUaM8sWDnhcEHnSA+EEKwEww0fdiOAf/hhx8WTG3hc88995h9OIaAThWDNPhi6gnscL6WzR0E7XMwSLW2tgo635tvvtlbN4Q04fWDgAD7mTNnys9//nOTF/IAixtuuEFWr15tlTz2FXxdDxAENLh+9tlnpg5RN1dddZUpM85061DFraYKT0p5ebkZDCF68G9mwGrBggUmyueff25sRvtDG0F7RJkOHTpkjtuDuR8Tzceuf9iAtWFYL4QtgtoJMQd269atM+1j//795jgG2ssvv9yIBRwHI7RNrQ/NR7daf8nqAnlqsG1PxcSOi/byxBNPePxwfWN6W9NG3GQCyOaCtNzrQu3DNlncVHagjeOmAdc+xD/aINqiXp+w84ILLhD0LydOnDDHL7vsMlMHyhD/hghp6Dq4ZG1D+4KgvgflsduOGwhfWQAAIABJREFUWz6/NmJzVy52GphqB3sV8OCF6wllgoAOY5Omyy0J5JNAnxVAtuhBp4Q7dh0EXODq5tfOFMf1IkbHr8dtL1JdXZ0ZoPbu3Wvi3nTTTYLBSwPSGjt2rBlkkAbyR4enAR6dJ5980vxEXNyVqyjAFjbrb0RCB6TxbdtwDOfbHSwEDMQIAkQKOlYICwTYogLF7HD+wEbbTnSGGPhRTgQ3L3cxrp0+zkXnr+IL5+M79uEYAsoFFhrcsgV1xvY5ro1IEwJFAwYheMSCPCoaT7ewwRZAWv+4y9YArvAsggeCW4caT7doG/DuoN1ogD1LliwxdYTB3S4T6gBtBnWHYA9Afkw0TYgntB0VYGr7smXLTBTXTpQV9Yt8MHjBDnvNHAQkxDzq1S+49edXF8oI59u2J2OCcthx8R3XmM0P16NeYzieTADheLLrwi5bsrip7AAntG+93pAuym9fn7gxU+8wmMMbiQ++u4y0ToLaBuovWd+D9Oy2g9+p2gjK6F7XdhooI8Q52pkG/KNsiCCUO4xNeh63JJBPAn1SAGFwgsdHPSboRNCp6922CxwXrO11wHF7n35XV7Ju9QkhHLcHTJxvd3r4rufYW3QqGle/mx1RoYC7OkzX6DkaR+3RQcnOCwMepoaw9gnxcUdvCymck0wAgR0GZXhlNF8tp9qqnblfWvY+fL/jjjviBBXuatF54hgCbIT9GtyyBXXG9jluHDdN97jmqzw1b9269Ynf7iCDuLjL1zRgj37XdOwtygshAUHhBhUpytve2pw0fb/yaJpo65988ok3BaZpKS/XTpu32mF7xnSf2qH56BY2adrY59qW7HgyJm5aiKtlsbfaFpFvMgGU6rrQ8mCbLG4qO8BCbdI07X2p7HTLrfztMut32OK2VZxv54ffqANtO/idqo24deimgfTtGzYchwcRXm2cG8YmnMNAAvkm0CcFEC5Cex2Hdhi4A7TvWhS+PQj47cNxXNxYVO0XUl3w6DDGjx9v7rz8zsdxu4PCtAfupuAKx90agt2Jufa6HR46OKwh+utf/2q8DnBX4xwEdJrJBBDSQt7w1CAdtzO08/JLy96H764AAn/cIeOYlgtpanDL5ubvd44bB6zsNN3jmobNXPPH1q1P/PYTQPDIaRrIT7/bael3lDeVALKFh56nW6St6fuVR+Ph2CWXXOJNgeoAqjxcO23eGte2Q/dpfWk+uoVNmjb2ubYlO56MiZsW4trTnJq/bpFvMgGEeMmuC01Ht0FxU9kBFvkQQHadqI3Yum0V+1wb7LaD46naiFuHOMdOA+m7AggeYngScW4Ym5AmAwnkm0CfFEDw9MCVj05MA6YUsJYFrlo3YK0DXOn2FIc9MOCuHSLC9iAhbZ1mSHXBwxMFl7L9JBrOVfvQoaCD0eB25ogHF7nGsW3DOXaHhzsxvEdHvT4QHBB+OoBhm0wAIQ+kpwHTNFjrApvcvNDpQRjaU1x2+jgnzBSYnZ9bNqTh5463z3HjuGVwj6MciKM8tay6detTRYDdPvymwILSQ7p+0z14fxPqCmnBa4eP1hvOsb/b9vqVR20Hf3sA1iks5YWtbafNG+0skykwTRs2uLYhL3s60j6ejAnKbsfF9DUWB9vT2PY1hLj21BJssZmlui6UH7bJ4qayA1OJ7vQQ+GidwM5UQs0uNzgkaxtuW4X9dn4uB/xO1Ubs/BEfwWaJ890yulNgyTzi0SS5IYG8E8hIAB39Xz+WoE8mFttrLzI5P51zMMWCBYS4SN2AeXR3kNE4WHSKNRqwFYsX3UXQOD5o0CAjoNApoaPDgl50QKk6IYgQ2KRPHmER4ZQpUzzB5Q5KWFSIQf+NN94w00dYsIonwXTgsgct2G93eLqIFR0SBgg87QGPAAZaBJRP1+BA+LnhhRdeMGIPa4jwgTAMmgLDwI0F0CgXbApaBI3jKDM4ID4WRuvgjjLBfg1u2bBWQj1Gaq97jtthpzqOvBAHH7/gV59Yb4L2gSd84JX729/+Zu54da2VW4duuhBRyRZBo97AGYvQwQb5QLCruLTtdctr54V6Rn3jyT8spscCcUyjKmPXTpe3th+0b/DG2qFUi6A1bdjh2oZjGAzBCfWPtqDtKRUTOy1ta5gWxgJyLCTHNYqnNcELa0/QTpYvX27sxlN/mMbVOtZyBV0XNsNkcVPZgTLedtttxragRdDpCCDYlaxt+LVVMFfBhfMhQNH2cA2CVao2EnTNKUtdBA3+qAfULcpkL4JOJoBgA9qXvZ7L5s/vJJArAqEFUJDgSbY/rJHdKYDQwaEjtL0taifuOHFMBy3djy0uSvsxeDzhgDtOFVIQE/AA6NoYeFUgLhDCdEKw509/+pPvI9DuoIQ00UEgfwxeEAzofNTt7A5adoeHu3g8UaWPwSMNPL6v3iYMOhCBeERXF8aaQkT/wGOAQQr54tyFCxcaZmCHYOeF3+gsMSjpY/BYF2N7mDAIID0MovjgMWbs04ByIU0NbtlQLxCusEcXgbvn2AMl0kl1XOMgnl/wq0+7/lFWLFDG03wa/OpQj+kW6epj8HjVAsQJBlQE1A9Ei11vqEetN9iq9rrl1fQ1Hfsx9meffdYMfrqI1rXT5Y387PPDPAZv159rG9rTI488YtoHrp3HHnsszqOYjImblt2W9LrA+RpwzVx44YWmreBaw5SjMkO5kl0XmoYyTBY3lR3uY/CwQQUJypSuAErWNlD+ZGID5cEbnCHe8fQZvLZuHbttJOiaU5ZI0+7P3LacyiYIKLy4FU+9MpBAPgmkFEDJBE7YY6kK0J0CKJUtQcfRKWCQ04C7E6z7QIfFEEwAzMBOAzwl2tnrPm5JoC8RgICwA8SnLR7sY/xOAiSQPwKBAiisuEknXlAxeoIAwt0IpqVwdwK3rv3ysqBy9fX9mCLB6/ExRYdOX98HYnsE+jojlr9vEcBUHDyi8EbhmoDnFO/50Xd+9S0aLC0JnF4CvgIoHVGTbly/4vYEAQTRg2kvrLWBex0udL9pNL/y9eV97hur7bfz9mUuLHvfJeC+ZRlT57Z3ue+SYclJoHsJJAigMIImmYmZnN8TBFCyMvMYCZAACZAACZBAzyIQJ4BSiZd0ipZOWhRA6ZBlXBIgARIgARIggWwJeAIomWDJJpMw6VIAZUOY55IACZAACZAACaRLwAigMCIl3YTt+KnSpwCyafE7CZAACZAACZBAvgkkFUC5zDxIBCEPCqBckmZaJEACJEACJEACqQgUJRMmqU5O93hQXhRA6ZJkfBIgARIgARIggWwI+AqgbBJMda6fCKIASkWNx0mABEiABEiABHJJoCAEEEQRAwmQAAmQAAmQAAl0F4EEAdQdGft5gbojX+ZBAiRAAiRAAiRAAiBQMAKIXiA2SBIgARIgARIgge4iECeAuitT5EMvUHfSZl4kQAIkQAIkQAI2gdACqKGhQR555BG58cYb4z47d+600wv9nQIoNCpGJAESIAESIAESyDGB0AJo+vTpUlJSkvA5++yzZe3atfLRRx8lfJKJIz8BxGmwHNcukyMBEiABEiABEvAl4Akg36PWziAB5CeK7H1vv/22lUr8Vz8RFB+Dv0iABEiABEiABEgg9wTSEkCDBw+WTZs2hf4gPoRTUKAACiLD/SRAAiRAAiRAAvkkkJYAuuGGG9KyBfEpgNJCxsgkQAIkQAIkQALdQIACKEPIe/fulQsuuEDGjh0rra2tsmXLFjnjjDPkySefzDDFwj/t1KlT8uijj8pPf/pT2bZtW+Eb3EMsbGxslDvuuENuueUWqa+v7yFWd7+Z27dvl/POO0+ef/55QVvsKQH9A/qJiy++WPbv359gdnt7u/z5z3+WSy65RL7++uuE45nuwNpMXKurVq3KNAmeRwK9mkCfFUDoQDdu3CjDhw83wkU71paWllAV7gqeZcuWmQXi2PbWcPLkSbn77rsDO/J0y51tHaSbX67jo63MmTNHLrvsMlP3AwcOlCVLlkhbW5vJqrOzUyZOnCi/+tWvkv7D3+bmZikvL5cRI0YUjABC2SA0zjnnHPNZtGiRoDxBAf/OBuXE+r9XX301LhoE3p133mmOXXPNNRmX8YsvvpBBgwbJCy+8kFMBpMLeXruIcuOp12PHjsWVJZMfELUo98033yx4mtYN4PrMM8/I5ZdfbgRS2HbjpuP+xoMp6Nfeeust9xB/kwAJ2C9CTEUDU1m9ZQoMHR4GKnhs0EHcd999cuWVV5oOGh1Rso5eObmCB54fpAdh1FtDXV2d6aSDOvJ0yp2LOkgnv1zHhciZMGGCaTOXXnqpGSyxxSC6YMECk92JEyeMZ+eqq66SI0eO5NqEvKb32muvmfYMwQvPFATBp59+Gpjne++9Z8qO8sNLiPrVgGO4NnBMPaZ6rBC2Wk8oIwTr008/bcoMeyFMIVCzCeotRtph+ha1pye2m2w48VwS6G4CfdIDVFtbK+eff74RdAcPHjTMcWeGO3DbTY27WnT+6LwxuH3wwQdex24LHnVxY0oMnZ16h0aOHGnOKy0tNYMCPCga4M4PShud5NKlS80dNfJGPPsfxjY1NZn09O4cd+pBnqtkcTFIoUwq/uANg10aYAc8HMgHd/fPPfec/OIXv5Dx48cL3PY4f/Pmzb7nQyA8/PDD5tiBAwc0SW8bpg6QxiuvvCLwrGAwgp22wNRB99lnnzXHHn/8ccNBPRfg/qc//Um+//57L99kdYpBH/mgnGCB86uqqnwHLR3U0Gb0rl7LBNGAfNCWkB4++L5v3z4zwKIt4YYCXPX1EZovBtsxY8aYdoNygDs+1dXVXhnglZg0aZKxD8dQbrQT9byAG+oN4h5lQFz1ZOzYscPUIfbBduTlDvDanjEth8FYBYzN3jMm+mXu3LkmXdQVyq9t3fb+aBlxSrK6RbuDWHA5oZw6/Yr2h3aI8uu1ofWnXli0ZdQj8oVdK1eulK6uLtd0UWFve6dUhKCfQL0msxcJIl2kj3xQF7j2td1rf1BZWWk8YTgO75Iy0naD8qBdue0G02bwfiFN1CfazZQpU7zz/eoU9Y261f4MU9Zgp+IUx6+++mrjmeK0a0KT4I4+QqBPCiAMcOgU0WHZYc2aNWZgwkCFThWd2YUXXmg6DXQW/fr1k927d5s1P7iTVcGjLm7tQFevXm3SxwBUUVFhOhp0eu+//77Jzk0brm90ajU1NeY45uwRHx0iOnV0wujIkQ86TeSNjvCBBx4wgw3iqtfBLo9OrWhcCCmUW+NiChD5omzonJGPvQ4BIgxp4zjKgfLYg5h9PjpWsMJdKwQHBgx4SDA99M0339hmme9h6gB2Ir977rnHCELUh6aPRDDo4rh+Fi9ebKZtYPPUqVM98QYPH+xxudt1ivQgapEWygGBAB4QfBhg3IBBCYPLFVdcYdqE7fFAXAwwECFIb/To0Ua4YR9Ehdqrg6sO3Gg32pYQB/U1btw4UwcqRpSrHkcbQP3iN+oD4gEeDDCAENHjqF8cU9GoNkBUuEEFCMoGUQUeaHM6YAfFx/WAurI9F8gPbQzlQJ4oI0KyulXxoTaCE+rAHtA1DuoQXBFUCEJwYC0N2vKQIUNkxYoVcvvttxsBsHXrVhPX/qMCwvZOoazgp3WUzF6kpdcK6kyZ33XXXYaZeosh1nBTgOsd9QPxi4Atyoprwq/d4PoBU9iCGwKIQ8TX69ivTpEO2Ki3VuOgrSGo6LLLbA7wDwn0IQJpCSC89NB9E3Sy34hfaE+BuR2bX11jAMCAAcGjHhEMLuh00JnpIKWCR70B2pno4I5zEPRcDLCati140Cnj7gzHEdCR650t4mPqAa8fgDcHIgqd58svv2w8MJhaQeeoeZsEon/sdJGOHRcdJDpoW/BAENplhOjCIPjdd9+ZFDEY4jgGMXDE+e5x2KaeAuQJb4IbwtSBe47e8euAhLQxEIAb1jjgDly9JzoAQyzAFnxwLFmdqtdD00f+qAeUVwcq2yYIHp1GRZzrrrtO/v73v8d5GHTQUc+MehowMKHNaEA+6tnQtqSeGR3IdKDXO3mIKnAEh4ceeshrLxDoaLcQcDimrLWtqmh86aWXjChUG9wt2ttPfvITI67QnpFOULDFCAZYbbu4TiAIUBa7jG46bt36cVIOOqBrHPU2af3pTYmKCmUPwbphwwbfRchaT3r9oS3hZgjX6G233SbwYtnBtRflxLVy/fXXy6FDh8x1iRsCtQXp2oJHBa+KEf2t4lDtUdvtvPFd2wCYIvjVqQoc9dZqHO2TcE3AJi2zmwd/k0BfIOAJoFRvYYabH50q1gGF/SC+Tg+4MP3eAZTKBjeNTH5rR6oDil8aKnAwsLkfdFY6SKnosDsTt3NE+toZo8NSEaJ39DiunZV2aOoBwkACAYm7WfUwoMNybcJvHTDt8ridnn0M3hCkb69LUDtRRr0rto/bg5gy8LPFTzDYeYepA8RHuTHQYyDSfHD3jMFPB10VO4gPRpg2QlxMn2CQx4CEkKpO9bgKBaSFQSzIA6T5ff7558a7AS8MBhR7ykwFow46ylSnIZCGijYdLO22hONue9E0MUgiuBzU26C8dIv2fvjwYVPfKlBMAs4flPvdd9813j6IMpQJeanHC23KDSpG0Ab/+c9/mnaFMqMdo+4wzQqhotMxOD9Z3fpx0mtErzk3jlt/6gFCvdx7773mGoQg9At6nSgr3UIMf/zxx+aUdO3VfFxhhv1uHeK6AmfUvX1c2w2EPMS2LrZX+2C33gi4darXMvoLvzjaTrBlIIG+SiC0AMo1ID8BlOs8/NILGnz1DhFz7Tq4464OnT/uHPVz9OhR01HZd092ZwLBh7tUW2DpHR06LO3IbcHiDnoYhOB5wgJMDCD4oDPUzhSDMkSK2oQtBmI3oGMNGsC1g1TRhXPtcuhxvUPUgVoHMbUZAsW2Qz1Vri327zB1gGk0iBtMe2GdEeoF4ifIA6Dp4+4drCAwMfjhfIi9VHWqx3WAVc+JCi5NH1sMKBhwUdeoKwRMU8AbpvH9BJS2A/vOXlmo8LLrAOm6QhX1hQFQRaaKD/WEqEDG9JtdL/Aiwl5wsUWjMd76o+0T9YpBX72ATzzxhBmkNV/rFM8jAWEHwYn0IQTh/cE0KNKx6y5V3fpxctsj6hgclKXWn31dQfBhfRjaAOJiatAVQXrDgusZ03SIg8/y5cs9z08qe9U2P3Go/YHWr9su9LpSAeweB2edfgNfrFnUqVV4jFwBrPWiXiW0J23Ldr1rO/GrT02DWxLo7QT6nADSDgedIjpmBHQQGPjQCWKKSe8m0WGg80PAQHX8+HHz3R2k0JnoHZzeKeuAhA4XnjCd4nDv6HFcBxfkjXVA6IAx8CF88sknRsSoUNG8dD0R7g51isqcYP3B1IXmi934tyRYE4FOzx1YwQCDnq5zUoGj3grkAa+KChAdcHSdA9LH4Id0ECBEINj8Qpg60EFlxowZJgnloC59135Eeuedd8w6CogCBLwHBQMfBoNUdepOK6gwsT11JlERT0zZx5A+xIIKIC2j/sa57p0/9qngUOGl06c6MCkHHVz1uHoHlIt66rRt6voQDKho59i6bU/LY281v1mzZpnd+K0eOLQPrV/7HAzE4KxTZWj7+I3z0J6VrdqoeQTVrR8ne0BH3vobWwSkiWsQ1wnaITyBmBpFuSFCsODbvilR+1WgqLDX/fY2lb1aPpQfAX1AWVmZEV9gj7S1ft12oe1MBZJ7XH9rf4X+AlOeel0H1Sk4aJ+keWj5NU17utcuL7+TQF8hECeAumMKCmD9vD/dlTfyt6eYHnzwQTOwo8O+//77zZQEOk2dSsFdLDoTLATF4kYM6ipCIBLUK6N3cOqaR+eDO0oMBviuj9OiA9Pzccw9rms4sGgT6xCwqBTnw2YEDCgYWGAP7IJ99joeEyn6R+Oi49NFzBik0WlqJwgvCcqFdMBAXwOgd732Qm4c145cF+PCNpQBL3JDXHADE3TS6GA/++wz2yTve6o60EEFAgLlxBb54zuCegBUGGCfro+CSMKgjDLpIJyqTnUQ1wFVhYkKLs/w6BNM+gg81v5AJEIc2vy0XYAPFr5ClCGe65FToYk2od4Ie2DSgR72Iei6LvBAemgHyFc5YMBFe0BdoE5QN/iO8/xEo10ufNf2p+ejHaIMyAPiyi+oKIPnBpxhF+LrOiRlqzYmq1s93+Wk1wx4IWhd220beYIXRBqEOThicTLsQztAnaHd2kFvWFTY28f0ezJ7EQfiF20NzPShB+QH4aTnartVMaL56Y2EXlduu4FnEW0QdYD+BB981zbiV6eahvZJysO91lUQaTm5JYG+RqBgBFB3gocIQceornE8Xo0FwHbnaL8IDh0H7n7xWKvbuahnQe/g1H0P4XLrrbeazgrnQnRowGJmLMhFuugo9c4ZxzEA2I+Ww0bYCpv1ONZU6IAL2+3H8zUPjfvmm2+acqLTtB/NxXGIHNiGY1hDYL/ED8fx6DX2w0ZMLWBQwkCkAZ05Bjkth744DrbaL3bT+PY2VR3gOF6+h7xRVnjFYKfm7+clgNdJH0XGYOiyCapT2KWDOAYtBL3rx36/gEEFYg+DHvJCPQXxGzZsmGk7ECO2RwjpqscGW/VG2AMTBk6920d8cMEUiHL53e9+Z/JHu9PgPv6t7cdPNOo5uvVrf1iHpkIbA7YdVLTZggUixH6AQNmqjcnqFlxdTu41h/wRD+0NbQKiHt9RD1p/aNt4BYLbNm3b8V0Finqn3OP4ncxejf/ll18aRrAH7VWvSRV/QcLaFsCall53aDeYcodIQ1+CskDEwbuk4savTt0+CekiH4hltJvHHnvMTNfa04WaN7ck0JcIJAigfHtiTrf3J9+Vi7tcdMTa2ec7P6bftwjY04wQARCwOm3Zt0iwtCRAAiSQHQEKoOz4xZ2NO0XcSdp3w3ER+IMEsiAAb8BNN91k1rJMmzbNeBwgtuEdQ9tjIAESIAESCE/AVwDlywsU5P3BPHZvCEFPZPSGsrEMhUHAnd5yp90Kw0paQQIkQAKFT6AIJgYJk1yanyyP3iKAcsmLaZEACZAACZAACeSPQFIBlCtPUJD40fQpgPJXwUyZBEiABEiABEggkYARQNidSqQknhpuT5h0KYDCsWQsEiABEiABEiCB3BDwBBCSSyZW1FsTNtt00qIACkuV8UiABEiABEiABHJBIE4AIcFUwiWVEMrkfAqgXFQl0yABEiABEiABEghLIEEA4cQwIibTOH6GUQD5UeE+EiABEiABEiCBfBHwFUDILFOBk+y8oEJQAAWR4X4SIAESIAESIIF8EAgUQJpZMkET9pimFbSlAAoiw/0kQAIkQAIkQAL5IJBSAGmmYcWOHU/PTbWlAEpFiMdJgARIgARIgARySSC0ANJMbYET9F3jht1SAIUlxXgkQAIkQAIkQAK5IJC2AMpFpm4aFEAuEf4mARIgARIgARLIJwEKoHzSZdokQAIkQAIkQAIFSYACqCCrhUaRAAmQAAmQAAnkkwAFUD7pMm0SIAESIAESIIGCJEABVJDVQqNIgARIgARIgATySYACKJ90mTYJkAAJkAAJkEBBEqAAcqqlo0Nk27ZOOX78lHPE/+crr7TJpEkt/ge5lwRIgARIgARIoCAJ9EkBBNFy550npbk5UeQcO3ZKhgxpkjffbA9VYekIoA0bOuS3v22SI0cS8w2TGc4vKjoW9/nP/2yUd95pl66uMCkwDgmQAAmQAAmQAAhQAGXZDrpbANkCqq1NZOPGDunXr1HefjucYMuyuDydBEiABEiABHoFAQogpxrhnYHIgLcFAQJn/PhmmTmzVX70o+Pys581yOuvt8upqBPHFkCYPnvssRYZOfKkNDbGe3kwTWZ7bzT9urouGTu2Wf71X+tN2nPntgnS8QtBHqS//KVV7rmnWSCIcO6rr7bJL3/ZKD/4Qb1ce+0J2bMn5h6Cvf/zP83y8MOxPNesiZUH+X78cYdccUWTsffqq5tk9+5OzxyUA2W86aYTgV40LzK/kAAJkAAJkECBEqAAcirGTwBB+Kxc2S4tLacEYuGiixrlq68iokIFEATRwoVtAsHw/fcxwWEn7wqYhoZTctttJ2T69FYzHff1113yX//VJC++2OoJrGTn6zFbAGHqDjYgLUyLLVjQJrffflJOnIgIMtgLcVRd3WHKg3Lh944dEZGza1enXH55k3zySae0t4ssX94u11xzQg4ejJwPAYT469Z1mHVSKgTVFm5JgARIgARIoCcQoAByaslPAMED1Bl1guD4pZc2yaZNMQ8RRMH773fIhRc2CgREUHAFEH5DXGDdkYbNmzuMCDp8OLZPj7nnQ3x8/nmn8VhB2CA0NZ0yHz3ns886jb21tTHBBu+PCpfWVpHRo08KRBT2PfJIi/mu58OTdcstJ7zyoqyIy0ACJEACJEACPZkABZBTe34CCIO+Br/jmA6CV2jWLH/PjZ7rChiIFltcIR48S1dd1SQQLm7A+fY0Gr7/2781GEECIYOA7aJFES+Pxj377AYvPfVY2WnPmxd5kg2LwrE4XM+zt8gbASxUbNlp8DsJkAAJkAAJ9CQCFEBObfkJnFQCCNNWmFIaOrQpLQ+QnwCCpwYepiABZC+Cdkw3P5Em7Nm3r8t4dJDOJZfE0vMTQPDooIwqgFasCF5QTQHkR537SIAESIAEehoBCiCnxjIRQCqQsPh41KjYehsnabOw2hYw8KpkMwXmpo/frkDBAuZBg4IFEKb24IWCCML3hx5qNh+d8kOa9nc3fT8buI8ESIAESIAECp0ABZBTQ9kIICw0/v3vT5qFx7rGxk4eYmTw4IgYwVSVLoLGomd4X775pktuuOFE2oug7TyefbbVLKxGWvj84Q8nxZ0CwyJmrGHCE2N4h9Cvf93oeZy2bOk08bGYGsIHT5BhEfXevZE1RBRANm1+JwESIAES6KkE+qwAste36HdMD2XTZK7lAAAXsElEQVQjgNAI8PTUb37TaLZuo4DIgYcFj6f/9a+RRct4DP6uu056j8HPmdNq1vG45+K3u4bILw5EmD7iDjtefrnNiC7YhYAyIr8HHog8Bv8f/xF5oksFG7Z4txBesAguSANPfOlxCiA/6txHAiRAAiTQ0wj0SQHU0yopl/b6rQHKZfpMiwRIgARIgAR6AoE+J4Dw1mT1+PS1LRqkLYByUX68G4mBBEiABEiABHoagT4ngHpaBeXaXlsA5TptpkcCJEACJEACPYUABVBPqSnaSQIkQAIkQAIkkDMCFEA5Q8mESIAESIAESIAEegoBCqCeUlO0kwRIgARIgARIIGcEKIByhpIJkQAJkAAJkAAJ9BQCFEA9paZoJwmQAAmQAAmQQM4IUADlDCUTIgESIAESIAES6CkEKIB6Sk3RThIgARIgARIggZwRoADKGUomRAIkQAIkQAIk0FMIUAD1lJqinSRAAiRAAiRAAjkjQAGUM5RMiARIgARIgARIoKcQoADqKTVFO0mABEiABEiABHJGgAIoZyiZEAmQAAmQAAmQQE8hQAHUU2qKdpIACZAACZAACeSMAAVQzlAyIRIgARIgARIggZ5CgAKop9QU7SQBEiABEiABEsgZAQqgnKFkQiRAAiRAAiRAAj2FAAVQT6kp2kkCJEACJEACJJAzAhRADsqODpFt2zrl+PFTzhH/n6+80iaTJrX4HzwNe5ubT8k//tEp2PqFI0dOyQ03nJDPPuv0O9zt+8BO+cEm2AYbGUiABEiABEggnwT6pACCaLnzzpO+IuHYsVMyZEiTvPlmeyju6QigDRs65Le/bcpogN+7t0suuKBRNm/uSLDr3Xc75MILG+Wbb7rkiy+65D//s9GIuISIIibvfAkgiC5wBZOwgQIoLCnGIwESIAESyCUBCqAsaXaXAOrsFBk/vlkeeaRFTlkOEux/6KFm88H3VCGfHiAKoFT0eZwESIAESKBQCFAAOTUBgQAvDbw1CBA4EB4zZ7bKj350XH72swZ5/fV2T4TYAgjTZ4891iIjR56UxkZLpYiYaZ6iomOiH02/rq5Lxo5tln/913qT9ty5bYJ0/AI8PZdf3iSHD8fS/u67Lrnkkpi9mEbCb53iQlp/+UvM9unTW42HS48jn48/7pArrmgytl19dZPs3h1TUijH44+3mLKj/Ph+4kQsf7UT6Z19doNXPvWwIf9XX22TX/6yUX7wg3q59toTsmdPl55muARNgb3/fsSztX17zB7vRH4hARIgARIggSwIUAA58PwEEAb+lSvbpaXllKxZ0y4XXdQoX30VGcRVAMErs3Bhm0BAfP99bIC3k3enwBoaTsltt50QiBJ4T77+ukv+67+a5MUXWz2BZZ8P4QMBpOIJx9aubTd5YuoOwRVAy5e3G3ED0QExg7T/9/9u8ATSrl2dJs1PPumU9nYRxL/mmhNy8OApI8QeeKBZ7r+/WZA+PhB32Ocn0vw8QJhKBBOUratLZMGCNrn99pOeiAqaAoNdmNaDCGIgARIgARIggVwToAByiPoJIHiAdHoJxy+9tEk2bYp5iDCIq7cCA3dQcAUQfkNsqHjBeVjjAxFke3k0PYisJ59s8aa7dFrsqadii7BtAaSCZOnS2HqmQ4dOyXXXRRZBIz1MqcFDpAEi6ZZbTpjyIS0ILqw/0oDv2IdjbtD8IAo1NDWdEnw04Dzwq62NpOkngHbs6DSiCYLSnu7TNLglARIgARIggWwJUAA5BP0EkE7RIKrf8ZtuOmG8QrNm+XtuNAtXAOn0moorxINn6aqr/AUGjuMJLwgQTH0hLsQEBIMGWwDBVqSFczRgny6CVsGi03L2Frbic/PNJ+Km8/B0HEQbjrlB07MFUGuryKJFkSkwTR9TZSqgXAGEBei33nrCfNxpRDc//iYBEiABEiCBTAlQADnk/AROKgEEj011dYcMHdok6XiA/AQQPCMQNSoQHPOMGIGHBuuBVqxoN1No9pqcTAQQ0vELfgII3ip3Gk7P9RNAKCP47NvXZbw5tn04zxVAv/51o5lmvPfeZjOlSA+Q0uWWBEiABEgglwQogByamQggFUhY7DtqVGx9i5O08ZrYj8FDYKQzBabpzZ7dKuXlzfLHP54Ue3oLx22BoYLEjmNPgcHz5PcEmXqkkFa2U2BgY3uEsMB60KCYwHMFkL4HCGuWIJySCUrlwS0JkAAJkAAJpEuAAsghlo0Agifm978/aRb6+nkuMPgPHhwZ/DE1pIugsTAZYgXv8YEACFoEraZCmGAaCU9W2etzcNwWQPjtLoLGgmt7EfSWLZG0sFgZwgfCA4uUkS4WOmPB88MPNxtb4f3BE2tBi6BRZsTFk2IoDxY9P/tsq/FSoWz4/OEPJ43tsBMhSAAhLSyYBk/bw2VO4h8SIAESIAESyJJAnxVAuh7F3sJTkY0AQl3gaarf/KbRbN26gSiAxwWPg//1r5GFwngM/q67TnqPwc+Z0yoQR8kCjo8efdI8nq/eGo3vCqBUj8FDaGzc2GFenggWsH3dug5v8THW4UDU4Ek4fQw+2docCKr/+I9G4zmCYIJ4wfl4zB9pv/xymxGB4IQQJIBwDAIR032pBKGWnVsSIAESIAESCEugTwqgsHAYjwRIgARIgARIoHcS6HMCqF+/Ru9lfbb3h99jL2ksRBZYRM1AAiRAAiRAArki0OcEUK7AMR0SIAESIAESIIGeS4ACqOfWHS0nARIgARIgARLIkAAFUIbgeBoJkAAJkAAJkEDPJUAB1HPrjpaTAAmQAAmQAAlkSIACKENwPI0ESIAESIAESKDnEqAA6rl1R8tJgARIgARIgAQyJEABlCE4nkYCJEACJEACJNBzCVAA9dy6o+UkQAIkQAIkQAIZEqAAyhAcTyMBEiABEiABEui5BCiAem7d0XISIAESIAESIIEMCVAAZQiOp5EACZAACZAACfRcAhRAPbfuaDkJkAAJkAAJkECGBCiAMgTH00iABEiABEiABHouAQqgnlt3tJwESIAESIAESCBDAhRAGYLjaSRAAiRAAiRAAj2XAAVQz607Wk4CJEACJEACJJAhAQqgDMHxNBIgARIgARIggZ5LgAIooO46Ozvl008/lYaGhoAYIo8//ri89tprgceTHTh27JiUlZVJbW1tsmg95tgrr7TJpEktPcbedAz98ssuwScoZFt2cEMaDL2LQLbtIhWNI0dOyW9/2yQbNnSYqMjvzjtPSnPzqVSn8jgJkICI9FkB9P3338uDDz4o55xzjpxxxhkyfPhw2bJli5w6Fek86uvrzb41a9aYhgKhMmTIkDjB0p0CCILp6quvlpKSEu9z5ZVXysqVKwVi7XSHVJ39Z591yiWXNAm2GtBxowNHR57P0Nh4SkaNOhmX91/+0hpadDz+eIvcd1+ztLWJGVwwyNiCJVXZU5WtJwmgtWvb5ZFHWgSXCeqvqOiYzJ7d6lvEd9/tkB/8oD6OlW/EbtrZ3QIhqF34tcdMEFAAZUKN55BAjECfFEAHDx404mbGjBly9OhRaW1tldWrV8v5558vNTU1MTrWt0IRQB999JGxqr29XTZv3iyDBg2SVatWWZaenq9Bnb1aczoF0ObNHTJ2bLO0RsfpY8dOye9/f1L27g326qjd7hZ3131VAIEfOIInggqgq69uEjC1AzT5Qw81G4GEtlEIoVAEkNseM2VDAZQpOZ5HAhECfVIALViwQO666y45efKk1w7g+amqqpKJEycaj4p6XCA4MM1le1502gseoFmzZkllZaWUlpbKpZdearxImqjrIbJFlE6BwRacBy/U6NGj5ciRI3p63Na2xz7w3HPPeTZjPwTcddddZ+yFV2vPnj1edNgD0Yd8kB88SJ988ol3vKurS9566y3PnjvvvFP27dvnHV+xYoXcfvvt0tjY6O3TLxhcxo9vlpkzW+VHPzouP/tZg7z+ervxFOAYPAX6wW94PfQ3thhM0aHfcMMJ41H4zW8ajffg7rtPytGjscF1z54uufbaE+bYL3/ZKEuXtktXVMcsWdImN954QhoaYvHhqYDHAp4LDcjrgQeaRR1nHR0i8Aip3TNmtMpFFzV6Uwv/fzvn1hrVFcXx79Av0cc+iD4qWB/ElrTUoIJEe5FgtFSCqJFqa5EGsSS0FBFaGiJIU1CKlj70pUoeRFDSUhKqwVtzadKmD61mEjMxcZffPv3PrFme5CSZWqfMWZCcmbNv67LnrP+stfbAK3+AuBdeuF/iW+mGxWRnTcmr9b3jYu6PPnoYQRkRE2S/ejUBGRpjr/Db01OMOqb/G28UwvBwGcyNj89HoPLcc3/GPl1dxcAYKE3HgMF79+bD++9PB42R7ey6yE8kjQgGhB43bZqMuiLaYwlw+eqrk/EP+UWMJaKGrvnjdaFQthe6sP09cKbvsWPl8fAsflhjYuJxhex2fzCvbCZ+7BUdogt08Pzz9+Oe8Hr7+uvZAOBD7/RlPRF7U23YRPtG7VzT9iP7m/3Iuujk449nKtJY1649Ci+/PBn3HfPfuJFEUf0+svKl2dl/lrJsUY2urcz561wDtaqBugNARE4OHz4curq6FrWJBxwWvGgggGL9+vUR9BBFOnv2bGhqairVDWUBIFJagBHScUSiWltbI5hKS2l5fsQDciAPcg0ODobGxsYwMDAQ3xMZgp+JiYnYHX5oBxQB/rq7u2Naj/WhS5cuhc2bN4ebN2/G8T09PaGhoSGMjY3F9vHx8ShrGn88fHl4f/PNbHj48HH47rvZ6Bh/+SVxzN6RMSEO1KbA9EDHqRFRGBtLHBKRBMAKwAaAc+ZMMczOhvDTT3Nh3boHpYjEr7/OR+Agp8UazPHWW1Pxynvmee+96WAd9oULs+HFFx/E+XAKn302Ex0c/EHWkS0UAVpMdnTDHCLJaecHzAF60F13dzGmC+E9jQA/OMKhofkY1SIF9eabUxFIoKPXXy+UnCh9XnllMsqE89Xa0vHo6HwEnatW3S/ZDrCIPmQ78cA6Nt0F/wCKTz6ZieBXgJL+9MNu/CE/hF1w9G1tiX2xcUvLVLwnm2UBoE8/nYljGKvxHR1JSg6HDT/aH4ODc6GhYTL09iZ2tABBMulq9xa8oBd0rH0ivQF6sAvvd+2aCp2diV3hZevWQtQzfLDfX3ppssLurOX3I/vp7benwocfPoygB0DFGtLzzz8n87DX2fPs1cbGQvj998clW2ofWfnEr+zMusyrz9JSbFGNrqXX/JproJY1UHcAaHp6Ouzbty+zeNkDjoUAkAVS9AFA3L17N9o8CwDt3LmzIkIDcAEUDQ8PP7FnPD9ErPr6+kopMN6fPHmyAtgVCoVAFIfaJgh+ADUi1TkBfKQXaopExWIxHDx4MFNX9OfhSwRITpAH8MaNk6VIxlIB0JYthdI3XOYlXQBIAtww52uvTYYffki+AbPWjz/OPeGoxT/XixfLNSu8x6kDEP74I/nmngZo5DzkWJYCgBaTHd1kASCctggQ1tRUiLzrnq5qsxEtnPcXX8wEIj/wjIPEIYvQISAImZHN65i1ceZTU8kYLz/z/PXX4wgksaNIa/X1zcVIj1KKAgOsawGNQIH6MQ+vAQqa1/an3e8b2okyAuYg7MkeYC/Az44dhYqIEJG99vYEpFiAkIwu/wcQwLeAGC2spbHaewAREfPt2TMVQat0YfWOXpnDkt+PzEd0h/0tIsLz5ZfFCG6JXiKDCPsjI2DZ28nKl2Zn+1laqi1Wqmvxm19zDdSyBuoeAMnxK8UFOOKeBxwLASClwzCy75MFgPwpMNJfgCLm8SR+xCdXCrjPnTsXU3ZeDttPdUOeH0BTe3t7BDjMv3379hg9smuTYmNcFmU5ee/ImA+n4SNApMDkDOnDN2IcNvdwcjgD0hNEES5fflSRKvA8+poV2kmJKGLAe+9E0u7hxOTI0gBTluxZ7cxNHxHOnW/u9p7a0hyb2rgyxoIx7gESiISgQ8Z7HTPGpobSdILztHVUzCv7EY0gsqCohQUDVjbuA+xw4iKAFYCNNsj2573fN9evzwWiZUQCiYSxtgg5bFpVr2U7L6fG6UoakCgJqSg/NktvzN3amhTKaz7uaW3upe1H5LbgU2O5aq+JF3tlnLeTlS+NX/tZWootqtG1lSN/nWugVjVQdwCI9A11Pjh2T4CZWgdAAjNDQ0MxfaWibQEgirkXIg+ApAvkXggAdXR0PFMAxDdjoj4WFOH0+CaNEyTKtNAR9YGBuejY5XBJTZBysd/ivRNBd/4eTkyOTE4JZyPyjs6Pz2pnbjvfvw2Abt2aj3paKQACeAJwbNQJ2XGiArAAJFJAOFmBIdJ5REgkW5rTJWJCBIg2yOvCAyD6YAP6v/vudKnWjPus40FInPSff7RboGfb2FNEyUg/ctoPsnZPAxR2vrS1uad9w3x+P3IPObIAEHs9jdL2meRL49d+lpZiC9Zcqa7T+M3v5RqoNQ3UHQDCAKR5KBAmBWTpaQAgm3KyESIAR7UpMGp4BNgAMxRj82drdOxrD4BIkTU3N8faHwGoalJg9mHvH85pjoyHsBwodkh7aNsUASkcQv86zcX1nXfKDtbaktdEJBSV4D0OiCJegJBITprIkMjzbh3hSgEQER2lbdLmryYFBk8UygL00GlWCmy5ESBftyI9WfspNXPq1EwpHeZ1xR4A7GSlwKwu7L7B3thf6Uv4wG5y+tTr8FML8CuiQF56t4BF7bqyDnrBNpBAqPZ02t6086Xp3afA/H5kHbu/xctvvyVpPYAYYJI/QKhIr/0+svyk8WvXyrJFtboWr/k110Ata6AuARAFwPv374+/AzQ6OhoLfkdGRkJbW1s4evRofK+UkyIupKcoIOY9Bc+QBxQW4NAOoCLNRRExYIt0E0ft6af5qymCprh527ZtEcCwHtEg5ue3iwA+d+7cCXv37g1Ei8QvMnCfomnAGeNVJE0tECfI1H7+/PlY07TUImg5C9byD2ciA/qmD+iAqHXgFBEPY1SqMYAF6lqIDhCxoQ6Dhz4pChzct98mJ7+U2uE9xDdcHCR1HKRWmpsrj7pTsGqBThwUksJS6jDgA0e+WBG0HCOnl3DwOFj/TV9y4BQhrhRrI+9C8y9WBG3lYr6lFEEjA/yNjCRFzryH9zTHaB0n83v+iUDYtGEUykWAuIeDJ00jh+0BkApvrX1Jq1EYrdobeAGIUISM/enLyTvtEfpzDxDL/B98UP6NJvpzQlDt7CFOjHFyC/JySg6uFIuzt+iLjSjmJ9WqPZ2lN9a2RdD9/cnvXml82n5kXWRYrAiaNBTys8f5DHDSjIJ3QKS3k5VPbVbX9rOUZQs+j9Xo2uo2f51roFY1UJcACGMAgk6fPh3WrFkTj4yvXbs2dHZ2RmBCuwCKABCAgoJnjruTFoKyABBrHD9+PB4557TYiRMn4qkqASDAkT0Gf+DAgdL6cQHzz/OjJk56CcRQ08NvA3G8nRogjtf39vaWftwRfvlraWmJPAF2+vv7NVXgGDwRIHjlmDwF1Ldv3y61Zx2D18OeAXoACwTw8KZ+h/oKFZby8MdZcqT4q6+KcQz1Ppwq0jH4Q4emK47BU/CqY/ActbdHvO0xeF+zgoOivsNGHyQYzgDeOMnFnBxDZn3xjlxWNpzS6tUPIqBjXhyPbfeyMz/FpDpejdP289tj8JzAorBYhPPjSDlrQcynY/DMiaOiAFrE6927k1oW5KFA+h/MHnW8nAiQHCH69IR+bAQP3dpTeR4AMR5wgVNG1/wBJLknAtgcOZLsCUAhutqwoZwC5cg4qS/GSnbAtcjKTh97pBw72ToaveY+BHjGLswLKMOmqqfCpll6W+wYvN+P4perPQbv7QVovXLlUTyVB7/w9/33j0pgFv1rn3oAlPVZyrJFNbq28uWvcw3UqgbqFgDVqkGeJl8esD3NtVYyd5qTWck8gC1/1J30CPdo84ST0W8J0WZrJXzfenufljasNx1UK2/afqx2zqzx/9ZnKWudvD3XwP9ZAzkA+j9bb5m8Py0AxO+F6Nv0Sq+I4h/aK51rOeNYl8gRqRJSJkQt+E0W1ZUsZ668b/nHLnNdPFtdPIvPkmz++eflY/vLfETl3XMN/Kca+BsNjAZy4LkDIAAAAABJRU5ErkJggg==)\n",
        "\n",
        "Click on the icon and it will give you:\n",
        "\n",
        "![image.png](data:image/png;base64,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)\n",
        "\n",
        "Your Google drive is under \"gdrive\". Click on the triangle icon on left of gdrive and it will be a dropdown view of your whole drive. Navigate to the folder containing your images. Once you are there, click on the three dots on the right of the folder and select \"Copy Path\"\n",
        "\n",
        "![image.png](data:image/png;base64,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)\n",
        "\n",
        "Copy and paste this path under the **Input_Directory** location below"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h_iAN7cAthma"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "# **Setup Cellpose**\n",
        "\n",
        "\n",
        "\n",
        "Install GPU version of Cellpose and configure dependencies\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hG3LSmJmLylT",
        "cellView": "form"
      },
      "source": [
        "#@markdown ##Step 2: Click Play to Install Cellpose\n",
        "#@markdown * After cellpose installation, the runtime will restart a warning that it has **crashed**, which is fine. This is essential to configure the environment.\n",
        "\n",
        "\n",
        "#@markdown * On occasion, you may not be able to see the folders on the left panel. Just refresh your browser webpage and it should appear.\n",
        "#!pip install git+https://www.github.com/mouseland/cellpose.git #to install development version\n",
        "!pip install cellpose\n",
        "!pip install torch torchvision torchaudio\n",
        "\n",
        "#Fix opencv error: ImportError: cannot import name '_registerMatType' from 'cv2.cv2'\n",
        "!pip install \"opencv-python-headless<4.3\"\n",
        "exit(0) #Restart Runtime to use newly installed numpy \n",
        "#or may get the error: ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@markdown ##Step 3: Click Play to configure Cellpose, dependencies and check GPU access\n",
        "\n",
        "#@markdown *Models will be downloaded when running for the first time*\n",
        "#disabled installation and import of mxnet as pytorch is the default neural net\n",
        "import numpy as np\n",
        "import time, os, sys, random\n",
        "from urllib.parse import urlparse\n",
        "import skimage.io\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib as mpl\n",
        "%matplotlib inline\n",
        "mpl.rcParams['figure.dpi'] = 300\n",
        "\n",
        "from urllib.parse import urlparse\n",
        "import shutil\n",
        "\n",
        "#from tifffile import imread, imsave\n",
        "\n",
        "\n",
        "print (\"Downloading Models\")\n",
        "from cellpose import models,core\n",
        "\n",
        "\n",
        "#https://stackoverflow.com/questions/8924173/how-do-i-print-bold-text-in-python\n",
        "BOLD = '\\033[1m'\n",
        "UNDERLINE = '\\033[4m'\n",
        "END = '\\033[0m'\n",
        "\n",
        "#Check if colab notebook instance has GPU access\n",
        "if core.use_gpu()==False: \n",
        "  #Warnings from the ZeroCost StarDist notebook\n",
        "  print(BOLD+UNDERLINE+'You do not have GPU access.'+END)\n",
        "  print('Did you change your runtime ?') \n",
        "  print('If the runtime setting is correct then Google did not allocate a GPU for your session')\n",
        "  print('Expect slow performance. To access GPU try reconnecting later')\n",
        "  use_GPU=False\n",
        "else:\n",
        "  print(BOLD+UNDERLINE+\"You have access to the GPU.\"+END+\"\\nDetails are:\")\n",
        "  print(\"*************************************************\")\n",
        "  !nvidia-smi\n",
        "  use_GPU=True\n",
        "\n",
        "print(\"*************************************************\")\n",
        "print(\"Libraries imported and configured\")\n"
      ],
      "metadata": {
        "cellView": "form",
        "id": "5ydQ-fggSiUm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-lZP6alpUAfY",
        "cellView": "form"
      },
      "source": [
        "#@markdown ###Step 4: Enter Directory path containing the images: \n",
        "#@markdown ##### Existing Masks directory will be deleted. (Does not read images in subfolders)\n",
        "Input_Directory = \"/content/gdrive/MyDrive/test_imgs\" #@param {type:\"string\"}\n",
        "input_dir = os.path.join(Input_Directory, \"\") #adds separator to the end regardless if path has it or not\n",
        "\n",
        "#@markdown ###Optional: Enter image extension here to read only files/images of specified extension (.tif,.jpg..): \n",
        "#@markdown ###### Leave empty if not specifying anything\n",
        "image_format = \"tif\" #@param {type:\"string\"}\n",
        "\n",
        "##@markdown ###Tick if image is RGB: \n",
        "#RGB= False #@param {type:\"boolean\"}\n",
        "#rgb=RGB\n",
        "save_dir = input_dir+\"Masks/\"\n",
        "if not os.path.exists(save_dir):\n",
        "  os.makedirs(save_dir)\n",
        "else:\n",
        "  print(\"Existing Mask Directory found. Deleting it.\")\n",
        "  shutil.rmtree(save_dir)\n",
        "\n",
        "#@markdown ##### Save Directory will be created in the input path under Masks\n",
        "\n",
        "##@markdown ###Advanced Parameters\n",
        "#Use_Default_Advanced_Parameters = True #@param {type:\"boolean\"}\n",
        "\n",
        "\n",
        "# r=root, d=directories, f = files\n",
        "files=[]\n",
        "\n",
        "for r, d, f in os.walk(input_dir):\n",
        "    for fil in f:\n",
        "      if (image_format):\n",
        "        if fil.endswith(image_format):\n",
        "          files.append(os.path.join(r, fil))\n",
        "      else:\n",
        "        files.append(os.path.join(r, fil))\n",
        "    break #only read the root directory; can change this to include levels\n",
        "if(len(files)==0):\n",
        "  print(\"Number of images loaded: %d.\" %(len(files)))\n",
        "  print(\"Cannot read image files. Check if folder has images\")\n",
        "else:\n",
        "  print(\"Number of images loaded: %d.\" %(len(files)))\n",
        "\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "98DA8zm4A__m",
        "cellView": "form"
      },
      "source": [
        "#@markdown ###Step 5: Read and Load all the images\n",
        "#@markdown #####A random image will be shown as an example with each channel separated. If it is a single channel image, only the one channel will be shown\n",
        "#imgs = [imread(f) for f in files]\n",
        "\n",
        "imgs=[] #store all images\n",
        "#Read images\n",
        "for f in files:\n",
        "  im=skimage.io.imread(f)\n",
        "  n_dim=len(im.shape) #shape of image\n",
        "  dim=im.shape #dimensions of image\n",
        "  channel=min(dim) #channel will be dimension with min value usually\n",
        "  channel_position=dim.index(channel)\n",
        "  #if no of dim is 3 and channel is first index, swap channel to last index\n",
        "  if n_dim==3 and channel_position==0: \n",
        "    #print(dim)\n",
        "    im=im.transpose(1,2,0)\n",
        "    dim=im.shape\n",
        "    #print(\"Shape changed\")\n",
        "  #print(dim)\n",
        "  imgs.append(im)\n",
        "\n",
        "nimg = len(imgs)\n",
        "print(\"No of images loaded are: \", nimg)\n",
        "\n",
        "print(\"Example Image:\")\n",
        "\n",
        "random_idx = random.choice(range(len(imgs)))\n",
        "x=imgs[random_idx]\n",
        "n_dim=len(x.shape)\n",
        "file_name=os.path.basename(files[random_idx])\n",
        "print(file_name+\" has \"+str(n_dim)+\" dimensions/s\")\n",
        "if n_dim==3:\n",
        "  channel_image=x.shape[2]\n",
        "  fig, axs = plt.subplots(1, channel_image,figsize=(12,5))\n",
        "  print(\"Image: %s\" %(file_name))\n",
        "  for channel in range(channel_image):\n",
        "      axs[channel].imshow(x[:,:,channel])\n",
        "      axs[channel].set_title('Channel '+str(channel+1),size=5)\n",
        "      axs[channel].axis('off')\n",
        "  fig.tight_layout()\n",
        "elif n_dim==2:\n",
        "  print(\"One Channel\")\n",
        "  plt.imshow(x)\n",
        "else:\n",
        "  print(\"Channel number invalid or dimensions wrong. Image shape is: \"+str(x.shape))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3JRxBPmatrK7"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "# **Run Cellpose**\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CDIgdlDYHnq4"
      },
      "source": [
        "####**Step 6**. Choose the model based on what you want to segment, Cytoplasm to segment whole cell if there is a cytoplasmic marker. \n",
        "####Choose nucleus to segment the nuclei.\n",
        "#### If you are using cytoplasm model and have a nuclear channel, this will assist with *segmentation*\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XcYskYudMajM",
        "cellView": "form"
      },
      "source": [
        "\n",
        "\n",
        "#updated the models\n",
        "\n",
        "Model_Choice = \"Cytoplasm2\" #@param [\"Cytoplasm\",\"Cytoplasm2\", \"Cytoplasm2_Omnipose\", \"Bacteria_Omnipose\", \"Nuclei\"]\n",
        "model_choice=Model_Choice\n",
        "\n",
        "print(\"Using model \",model_choice)\n",
        "\n",
        "#@markdown If the image has only one channel, leave it as 0\n",
        "Channel_for_segmentation=\"0\" #@param[0,1,2,3]\n",
        "segment_channel=int(Channel_for_segmentation)\n",
        "\n",
        "# @markdown ###If you choose cytoplasm, tick if you have a nuclear channel\n",
        "Use_nuclear_channel= False #@param {type:\"boolean\"}\n",
        "Nuclear_channel=\"3\" #@param[1,2,3]\n",
        "nuclear_channel=int(Nuclear_channel)\n",
        "\n",
        "# @markdown ###If cells are elongated or have branches, tick this. It will use omnipose\n",
        "\n",
        "omni= False #@param {type:\"boolean\"}\n",
        "\n",
        "\n",
        "#@markdown ### Diameter of cell (pixels):\n",
        "#@markdown #### Enter 0 if you don't know and cellpose will estimate it automatically. You can define this later as well.\n",
        "Diameter =  0#@param {type:\"number\"}\n",
        "diameter=Diameter\n",
        "\n",
        "# define CHANNELS to run segementation on\n",
        "# grayscale=0, R=1, G=2, B=3\n",
        "# channels = [cytoplasm, nucleus]\n",
        "# if NUCLEUS channel does not exist, set the second channel to 0\n",
        "# channels = [0,0]\n",
        "# IF ALL YOUR IMAGES ARE THE SAME TYPE, you can give a list with 2 elements\n",
        "# channels = [0,0] # IF YOU HAVE GRAYSCALE\n",
        "# channels = [2,3] # IF YOU HAVE G=cytoplasm and B=nucleus\n",
        "# channels = [2,1] # IF YOU HAVE G=cytoplasm and R=nucleus\n",
        "\n",
        "\n",
        "if model_choice==\"Cytoplasm\":\n",
        "  model_type=\"cyto\"\n",
        "\n",
        "elif model_choice==\"Cytoplasm2\":\n",
        "  model_type=\"cyto2\"\n",
        "\n",
        "elif model_choice==\"Cytoplasm2_Omnipose\":\n",
        "  model_type=\"cyto2_omni\"\n",
        "\n",
        "elif model_choice==\"Bacteria_Omnipose\":\n",
        "  model_type=\"bact_omni\" \n",
        "  diameter = 0\n",
        "\n",
        "elif model_choice==\"Nuclei\":\n",
        "  model_type=\"nuclei\" \n",
        "\n",
        "\n",
        "# channels = [cytoplasm, nucleus]\n",
        "if model_choice not in \"Nucleus\":\n",
        "  if Use_nuclear_channel:\n",
        "    channels=[segment_channel,nuclear_channel]\n",
        "  else:\n",
        "    channels=[segment_channel,0]\n",
        "else: #nucleus\n",
        "  channels=[segment_channel,0]\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "# DEFINE CELLPOSE MODEL\n",
        "# model_type='cyto' or model_type='nuclei'\n",
        "model = models.Cellpose(gpu=use_GPU, model_type=model_type, omni = omni)\n",
        "\n",
        "# define CHANNELS to run segementation on\n",
        "# grayscale=0, R=1, G=2, B=3\n",
        "# channels = [cytoplasm, nucleus]\n",
        "# if NUCLEUS channel does not exist, set the second channel to 0\n",
        "# channels = [0,0]\n",
        "# IF ALL YOUR IMAGES ARE THE SAME TYPE, you can give a list with 2 elements\n",
        "# channels = [0,0] # IF YOU HAVE GRAYSCALE\n",
        "# channels = [2,3] # IF YOU HAVE G=cytoplasm and B=nucleus\n",
        "# channels = [2,1] # IF YOU HAVE G=cytoplasm and R=nucleus\n",
        "\n",
        "# or if you have different types of channels in each image\n",
        "#channels = [[2,3], [0,0], [0,0]]\n",
        "\n",
        "# if diameter is set to None, the size of the cells is estimated on a per image basis\n",
        "# you can set the average cell `diameter` in pixels yourself (recommended) \n",
        "# diameter can be a list or a single number for all images\n",
        "if diameter is 0:\n",
        "  diameter = None\n",
        "  print(\"Diameter is set to None. The size of the cells will be estimated on a per image basis\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_obeWmo9R4Fb"
      },
      "source": [
        "####**Step 6: Test cellpose on an image. You can tune the parameters below based on your image** \n",
        "\n",
        "The cell below gives you an opportunity to test different parameters and figure out the right combination that gives your output of interest\n",
        "\n",
        "The ***Flow_threshold parameter*** is  the  maximum  allowed  error  of  the  flows  for  each  mask.   The  default  is 0.4. \n",
        "\n",
        "*   **Increase** this threshold if cellpose is not returning as many masks as you’d expect\n",
        "\n",
        "*   **Decrease** this threshold if cellpose is returning too many ill-shaped masks. \n",
        "\n",
        "The ***Cell Probability Threshold*** determines proability that a detected object is a cell.   The  default  is 0.0. \n",
        "\n",
        "*   **Decrease** this threshold if cellpose is not returning as many masks as you’d expect or if masks are too small\n",
        "\n",
        "*   **Increase** this threshold if cellpose is returning too many masks esp from dull/dim areas. \n",
        "\n",
        "If you do not know **diameter** of the cells OR if cells are of varying diameters, enter 0 in the Diameter box and cellpose will automatically estimate the diameter\n",
        "\n",
        "For more information on the parameters: https://cellpose.readthedocs.io/_/downloads/en/latest/pdf/"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Rr0UozRm42CA",
        "cellView": "form"
      },
      "source": [
        "from skimage.util import img_as_ubyte\n",
        "\n",
        "Image_Number =  1#@param {type:\"number\"}\n",
        "Image_Number-=1 #indexing starts at zero\n",
        "#print(Image_Number)\n",
        "Diameter = 0#@param {type:\"number\"}\n",
        "#Flow_Threshold=0.4#@param {type:\"number\"}\n",
        "Flow_Threshold = 0.3 #@param {type:\"slider\", min:0.1, max:1.1, step:0.1}\n",
        "flow_threshold=Flow_Threshold\n",
        "\n",
        "#Cell_Probability_Threshold=-5#@param {type:\"number\"}\n",
        "#Using slider to restrict values\n",
        "Cell_Probability_Threshold=-1 #@param {type:\"slider\", min:-6, max:6, step:1}\n",
        "cellprob_threshold=Cell_Probability_Threshold\n",
        "\n",
        "\n",
        "diameter=Diameter\n",
        "if diameter is 0:\n",
        "  diameter = None\n",
        "if Image_Number is -1:\n",
        "  Image_Number=0\n",
        "  #print(\"Image_Number is set to zero, opening first image.\")\n",
        "try:\n",
        "    image = imgs[Image_Number]\n",
        "except IndexError as i:\n",
        "   print(\"Image number does not exist\",i)\n",
        "   print(\"Actual no of images in folder: \",len(imgs))\n",
        "print(\"Image: %s\" %(os.path.splitext(os.path.basename(files[Image_Number]))[0]))\n",
        "img1=imgs[Image_Number]\n",
        "\n",
        "import cv2\n",
        "\n",
        "masks, flows, styles, diams = model.eval(img1, diameter=diameter, flow_threshold=flow_threshold,cellprob_threshold=cellprob_threshold, channels=channels)\n",
        "\n",
        "\n",
        "\n",
        "# DISPLAY RESULTS\n",
        "from cellpose import plot\n",
        "maski = masks\n",
        "flowi = flows[0]\n",
        "\n",
        "#convert to 8-bit if not so it can display properly in the graph\n",
        "if img1.dtype!='uint8':\n",
        "  img1=img_as_ubyte(img1)\n",
        "\n",
        "fig = plt.figure(figsize=(24,8))\n",
        "plot.show_segmentation(fig, img1, maski, flowi, channels=channels)\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y4FwOpghSE8w"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "## **USE CELLPOSE TO SEGMENT CELLS**\n",
        "The values defined above will be used for segmentation.\n",
        "\n",
        "The masks will be saved automatically.\n",
        "\n",
        "Saving the flow image/s are optional. Tick if you want to save them.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XVEQ_G1sI-VB",
        "cellView": "form"
      },
      "source": [
        "#@markdown ### **Step 8. Run Cellpose on folder of images**\n",
        "\n",
        "#@markdown ###Tick if you want to save the flow image/s: \n",
        "Save_Flow= False #@param {type:\"boolean\"}\n",
        "#@markdown ##### *Flow image will be resized when saved\n",
        "save_flow=Save_Flow\n",
        "\n",
        "print(\"Running segmentation on channel %s\" %(segment_channel))\n",
        "print(\"Using the model: \",model_choice)\n",
        "if diameter is None:\n",
        "  print(\"Diameter will be estimated from the image/s\")\n",
        "else:\n",
        "  print(f\"Cellpose will use a diameter of {diameter}\")\n",
        "\n",
        "print(f\"Using a flow threshold of: {flow_threshold} and a cell probability threshold of: {cellprob_threshold}\")\n",
        "\n",
        "#if too many images, it will lead to memory error. \n",
        "#will evaluate on a per image basis\n",
        "#masks, flows, styles, diams = model.eval(imgs, diameter=diameter, flow_threshold=flow_threshold,cellprob_threshold=cellprob_threshold, channels=channels)\n",
        "\n",
        "\n",
        "#save images in folder with the diameter value used in cellpose\n",
        "print(\"Segmentation Done. Saving Masks and flows now\")\n",
        "print(\"Save Directory is: \",save_dir)\n",
        "if (not os.path.exists(save_dir)):\n",
        "    os.mkdir(save_dir)\n",
        "\n",
        "if save_flow:\n",
        "  print(\"Saving Flow\")\n",
        "  flows_save_dir=save_dir+\"flows\"+os.sep\n",
        "  print(\"Save Directory for flows is: \",flows_save_dir)\n",
        "  if (not os.path.exists(flows_save_dir)):\n",
        "      os.mkdir(flows_save_dir)\n",
        "\n",
        "\n",
        "for img_idx, img in enumerate(imgs):\n",
        "    file_name=os.path.splitext(os.path.basename(files[img_idx]))[0]\n",
        "    print(\"\\nSegmenting: \",file_name)\n",
        "    mask, flow, style, diam = model.eval(img, diameter=diameter, flow_threshold=flow_threshold,cellprob_threshold=cellprob_threshold, channels=channels)\n",
        "    #save images in folder with the diameter value used in cellpose\n",
        "    print(\"Segmentation complete . Saving Masks and flows\")\n",
        "    #Output name for masks\n",
        "    mask_output_name=save_dir+\"MASK_\"+file_name+\".tif\"\n",
        "    #Save mask as 16-bit in case this has to be used for detecting than 255 objects\n",
        "    mask=mask.astype(np.uint16)\n",
        "    #Save flow as 8-bit\n",
        "    skimage.io.imsave(mask_output_name,mask, check_contrast=False)\n",
        "    if save_flow:\n",
        "      #Output name for flows\n",
        "      flow_output_name=flows_save_dir+\"FLOWS_\"+file_name+\".tif\"\n",
        "      #Save as 8-bit\n",
        "      flow_image=flow[0].astype(np.uint8)\n",
        "      skimage.io.imsave(flow_output_name,flow_image, check_contrast=False)\n",
        "\n",
        "#Save parameters used in Cellpose\n",
        "parameters_file=save_dir+\"Cellpose_parameters_used.txt\" \n",
        "outFile=open(parameters_file, \"w\") \n",
        "outFile.write(\"CELLPOSE PARAMETERS\\n\") \n",
        "outFile.write(\"Model: \"+model_choice+\"\\n\") \n",
        "if diameter == 0:\n",
        "  diameter = \"Automatically estimated by cellpose\"\n",
        "outFile.write(\"Omni Flag: \"+str(omni)+\"\\n\") \n",
        "outFile.write(\"Diameter: \"+str(diameter)+\"\\n\") \n",
        "outFile.write(\"Flow Threshold: \"+str(flow_threshold)+\"\\n\") \n",
        "outFile.write(\"Cell probability Threshold: \"+str(cellprob_threshold)+\"\\n\") \n",
        "outFile.close() \n",
        "print(\"\\nSegmentation complete and files saved\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Drx_6GbEHIOj"
      },
      "source": [
        "---\n",
        "#**How to extract the ROIs from the mask or label maps using FIJI**\n",
        "\n",
        "The masks will be saved as a 16-bit label image, where each cell/ROI is a different colour or label. If you are using ImageJ/FIJI and want to convert this to ROIs for the ROI Manager, there are a few options to convert into ROIs:\n",
        "\n",
        "\n",
        "*   SCF plugin (Update site:\thttps://sites.imagej.net/SCF-MPI-CBG/). After [installing the plugin](https://imagej.net/How_to_follow_a_3rd_party_update_site), go to SCF-> Segmentation -> LabelMap to ROI Manager (2D). This should generate the ROIs in ROI Manager\n",
        "\n",
        "*   Another really nice plugin: [LabelsToROIs](https://github.com/ariel-waisman/LabelsToROis). It has some nice features to adjust the size of the ROIs and generate measurements  \n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MvpagyDxY1B7"
      },
      "source": [
        "###**Save Files as *.npy**\n",
        "\n",
        "If you want to save the files so as to view it in the Cellpose GUI, you can save it as an *_seg.npy files in colab that you can download and open in the GUI.\n",
        "\n",
        "The file/s will be saved in the Input_Directory\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "cellView": "form",
        "id": "s7NAuncdY618"
      },
      "source": [
        "#@markdown ### **Run this cell to save output of the model as an npy file**\n",
        "from cellpose import io\n",
        "\n",
        "#save output of model eval to be loaded in GUI\n",
        "io.masks_flows_to_seg(imgs, masks, flows, diams, files, channels)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DllTSOoD9lnU"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "**Acknowledgments:**\n",
        "\n",
        "Thanks to Lior Pytowski from University of Oxford for testing this notebook out and giving some really good suggestions\n"
      ]
    }
  ]
}