ResNet34_labeled_images_classification.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tifffile\n",
    "import os\n",
    "import pandas as pd\n",
    "import glob\n",
    "import random\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision.models as models\n",
    "import torch.optim as optim\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def same_seeds(seed):\n",
    "    # Python built-in random module\n",
    "    random.seed(seed)\n",
    "    # Numpy\n",
    "    np.random.seed(seed)\n",
    "    # Torch\n",
    "    torch.manual_seed(seed)\n",
    "    if torch.cuda.is_available():\n",
    "        torch.cuda.manual_seed(seed)\n",
    "        torch.cuda.manual_seed_all(seed)\n",
    "    torch.backends.cudnn.benchmark = False\n",
    "    torch.backends.cudnn.deterministic = True\n",
    "\n",
    "same_seeds(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda:1\n"
     ]
    }
   ],
   "source": [
    "def set_device():\n",
    "    if torch.cuda.is_available():\n",
    "        device = torch.device('cuda:1')\n",
    "    else:\n",
    "        device = torch.device('cpu')\n",
    "        \n",
    "    print(device)\n",
    "\n",
    "    return device\n",
    "\n",
    "\n",
    "device = set_device()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def test_transform(img):\n",
    "    x = np.zeros(img.shape,dtype=np.float32)\n",
    "    for i in range(img.shape[0]):\n",
    "        if img[i].min() >= 0:\n",
    "            img[i]/=255\n",
    "        else:\n",
    "            img[i] = (img[i] - img[i].min())/(img[i].max()-img[i].min())\n",
    "        \n",
    "        m = img[i].mean()\n",
    "        s = img[i].std()\n",
    "        x[i] = (img[i]-m)/s\n",
    "    \n",
    "    x = torch.tensor(x,dtype=torch.float32)\n",
    "\n",
    "    return x\n",
    "\n",
    "class test_cycledataset():\n",
    "    def __init__(self,filelist,transform):\n",
    "        self.filelist = filelist\n",
    "        self.transform = transform\n",
    "\n",
    "        \n",
    "    def __getitem__(self,index):\n",
    "        if isinstance(self.filelist,list):\n",
    "            # order of channel: (nucleus, mitochondria, bright-field)\n",
    "            # It is necessary to follow the above order.\n",
    "                         \n",
    "            img = tifffile.imread(self.filelist[index])\n",
    "\n",
    "            \n",
    "            img = np.concatenate((img[0:2],np.expand_dims(img[4],axis=0)),axis=0)  #(nucleus, mitochondria, bright-field)\n",
    "            img = self.transform(img)\n",
    "            return img.to(device)\n",
    "\n",
    "        else:\n",
    "            img = tifffile.imread(self.filelist.iloc[index,0])\n",
    "\n",
    "            img = np.concatenate((img[0:2],np.expand_dims(img[4],axis=0)),axis=0)  #(nucleus, mitochondria, bright-field)\n",
    "            img = self.transform(img)\n",
    "            return img.to(device)\n",
    "                        \n",
    "    def __len__(self):\n",
    "        if isinstance(self.filelist,list): \n",
    "            return len(self.filelist)\n",
    "        else: \n",
    "            return self.filelist.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of test set:148\n"
     ]
    }
   ],
   "source": [
    "#test = glob.glob('path to the tiff files')\n",
    "test = pd.read_csv('path to the csv file of test data set')\n",
    "print(f'number of test set:{len(test)}')\n",
    "testdataset = test_cycledataset(test, test_transform)\n",
    "\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "bs = 64\n",
    "\n",
    "testdataloader = DataLoader(testdataset,batch_size=bs)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifier = models.resnet34()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifier.fc = torch.nn.Sequential(\n",
    "    torch.nn.Linear(512,2,bias=True)\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remember to change the folder name\n",
    "\n",
    "save_dir = 'path to the folder'\n",
    "if not os.path.exists(save_dir):\n",
    "    os.makedirs(save_dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "classifier.load_state_dict(torch.load('path to the trained model',map_location=device))\n",
    "classifier.to(device)\n",
    "classifier.eval()\n",
    "\n",
    "y_pred = torch.tensor([])\n",
    "\n",
    "for i, x in enumerate(testdataloader):\n",
    "    with torch.no_grad():\n",
    "        yhat = classifier(x)\n",
    "        p = torch.max(yhat,1)[1]  # return the index of maximum (max prob after sofmax)\n",
    "        y_pred = torch.cat([y_pred,p.cpu()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 1.,\n",
       "        1., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "        0., 0., 0., 0.])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "if isinstance(test,list):\n",
    "    df = pd.DataFrame({'path':test,'prediction':y_pred})\n",
    "else:\n",
    "    df = pd.DataFrame({'path':test['path'],'prediction':y_pred})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "df.to_csv(save_dir+'label_prediction.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def accuracy(df):\n",
    "    n = df.shape[0]\n",
    "    acc = 0\n",
    "    label = []\n",
    "    for i in range(n):\n",
    "        if 'rfp' in df.loc[i,'path']:\n",
    "            label.append(1)\n",
    "            if df.loc[i,'prediction'] == 1:\n",
    "                acc += 1\n",
    "        else:\n",
    "            label.append(0)\n",
    "            if df.loc[i,'prediction'] == 0:\n",
    "                acc += 1\n",
    "\n",
    "    f1score = f1_score(label,y_pred,average='micro')\n",
    "    return [acc/n, f1score, label]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test acc:0.74\n",
      "test f1 score:0.74\n"
     ]
    }
   ],
   "source": [
    "test_acc,test_f1score, label = accuracy(df)\n",
    "\n",
    "print('test acc:{:.2f}'.format(test_acc))\n",
    "print('test f1 score:{:.2f}'.format(test_f1score))\n",
    "\n",
    "\n",
    "\n",
    "with open(save_dir+'accuracy.txt','w') as file:\n",
    "    file.write('test acc:{:.2f}\\n'.format(test_acc))\n",
    "    file.write('test f1 score:{:.2f}\\n'.format(test_f1score))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns \n",
    "\n",
    "\n",
    "mat = confusion_matrix(label,df['prediction'])\n",
    "plt.figure(figsize = (5,5))\n",
    "sns.heatmap(mat,cmap='binary', square= True, annot=True, fmt='d', xticklabels=['S/G2/M','G1'], yticklabels=['S/G2/M','G1'], cbar= False)\n",
    "plt.xlabel(\"prediction\")\n",
    "plt.ylabel(\"true\")\n",
    "plt.savefig(save_dir+'confusion_matrix.png')\n",
    "plt.show()"
   ]
  },
  {
   "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}