pre_work.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1023,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feel free to change / extend / adapt this source code as needed to complete the homework, based on its requirements.\n",
    "# This code is given as a starting point.\n",
    "#\n",
    "# REFEFERENCES\n",
    "# The code is partly adapted from pytorch tutorials, including https://pytorch.org/tutorials/beginner/blitz/cifar10_tutorial.html\n",
    "\n",
    "# ---- hyper-parameters ----\n",
    "# You should tune these hyper-parameters using:\n",
    "# (i) your reasoning and observations, \n",
    "# (ii) by tuning it on the validation set, using the techniques discussed in class.\n",
    "# You definitely can add more hyper-parameters here.\n",
    "batch_size = 8\n",
    "max_num_epoch = 300\n",
    "hps = {'lr':0.01}\n",
    "\n",
    "# ---- options ----\n",
    "DEVICE_ID = 'cpu' # set to 'cpu' for cpu, 'cuda' / 'cuda:0' or similar for gpu.\n",
    "LOG_DIR = 'checkpoints'\n",
    "VISUALIZE = False # set True to visualize input, prediction and the output from the last batch\n",
    "LOAD_CHKPT = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1024,
   "metadata": {},
   "outputs": [],
   "source": [
    "# --- imports ---\n",
    "import torch\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim\n",
    "import torchvision.transforms as transforms\n",
    "import hw3utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1025,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.multiprocessing.set_start_method('spawn', force=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1026,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ---- utility functions -----\n",
    "def get_loaders(batch_size,device):\n",
    "    data_root = 'ceng483-hw3-dataset' \n",
    "    train_set = hw3utils.HW3ImageFolder(root=os.path.join(data_root,'train'),device=device)\n",
    "    train_loader = torch.utils.data.DataLoader(train_set, batch_size=batch_size, shuffle=True, num_workers=0)\n",
    "    val_set = hw3utils.HW3ImageFolder(root=os.path.join(data_root,'val'),device=device)\n",
    "    val_loader = torch.utils.data.DataLoader(val_set, batch_size=batch_size, shuffle=False, num_workers=0)\n",
    "    # Note: you may later add test_loader to here.\n",
    "    return train_loader, val_loader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1028,
   "metadata": {},
   "outputs": [],
   "source": [
    "class AccuracyLoss(nn.Module):\n",
    "    def __init__(self, error_margin=12):\n",
    "        super(AccuracyLoss, self).__init__()\n",
    "        self.error_margin = error_margin\n",
    "\n",
    "    def forward(self, ground_truth, prediction):\n",
    "        # Ensure the images have the same shape\n",
    "        assert ground_truth.shape == prediction.shape, \"Ground truth and prediction shapes must match\"\n",
    "        \n",
    "        # Convert to PyTorch tensors\n",
    "        ground_truth = 255*(ground_truth+1)/2\n",
    "        ground_truth = ground_truth.view(-1).long()\n",
    "        prediction = 255*(prediction+1)/2\n",
    "        prediction = prediction.view(-1).long()\n",
    "        \n",
    "        # Compute the accuracy\n",
    "        cur_acc = torch.sum(torch.abs(ground_truth - prediction) < self.error_margin).float() / ground_truth.shape[0]\n",
    "\n",
    "        # Compute the ratio of correctly estimated pixels to all pixels\n",
    "        return 1.0 - cur_acc  # Returning 1.0 - accuracy as a loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1029,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ---- ConvNet -----\n",
    "class Net(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(Net, self).__init__()\n",
    "\n",
    "        # Decoder\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Conv2d(1, 16, kernel_size=3, stride=1, padding=1),\n",
    "            nn.ReLU(),  # Adding ReLU activation after BatchNorm\n",
    "            nn.Conv2d(16, 16, kernel_size=3, stride=1, padding=1),\n",
    "            nn.ReLU(),\n",
    "            nn.Conv2d(16, 16, kernel_size=3, stride=1, padding=1),\n",
    "            nn.ReLU(),\n",
    "            nn.Conv2d(16, 3, kernel_size=3, stride=1, padding=1),\n",
    "            nn.Tanh()\n",
    "            )\n",
    "            \n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.model(x)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1030,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "device: cpu\n",
      "training begins\n",
      "Epoch 1 \t\t Training Loss: 0.051330359901090306 \t\t Validation Loss: 0.13192280754446983 \t\t Accuracy Loss: 3.1920020847320556\n",
      "Validation Loss Decreased(inf--->32.980702) \t Saving The Model\n",
      "Saving the model, end of epoch 1\n",
      "Epoch 2 \t\t Training Loss: 0.015466739783605066 \t\t Validation Loss: 0.11991496992111206 \t\t Accuracy Loss: 2.989938533782959\n",
      "Validation Loss Decreased(32.980702--->29.978742) \t Saving The Model\n",
      "Saving the model, end of epoch 2\n",
      "Epoch 3 \t\t Training Loss: 0.01431455910905077 \t\t Validation Loss: 0.11267004364728928 \t\t Accuracy Loss: 2.8518762226104735\n",
      "Validation Loss Decreased(29.978742--->28.167511) \t Saving The Model\n",
      "Saving the model, end of epoch 3\n",
      "Epoch 4 \t\t Training Loss: 0.01355975817743177 \t\t Validation Loss: 0.10740015482902526 \t\t Accuracy Loss: 2.7557702026367186\n",
      "Validation Loss Decreased(28.167511--->26.850039) \t Saving The Model\n",
      "Saving the model, end of epoch 4\n",
      "Epoch 5 \t\t Training Loss: 0.012966641608012704 \t\t Validation Loss: 0.10316841213405133 \t\t Accuracy Loss: 2.6832529106140135\n",
      "Validation Loss Decreased(26.850039--->25.792103) \t Saving The Model\n",
      "Saving the model, end of epoch 5\n",
      "Epoch 6 \t\t Training Loss: 0.012459647198305867 \t\t Validation Loss: 0.09951173178851605 \t\t Accuracy Loss: 2.617139387130737\n",
      "Validation Loss Decreased(25.792103--->24.877933) \t Saving The Model\n",
      "Saving the model, end of epoch 6\n",
      "Epoch 7 \t\t Training Loss: 0.012047764195795781 \t\t Validation Loss: 0.09632357709109783 \t\t Accuracy Loss: 2.5558416538238524\n",
      "Validation Loss Decreased(24.877933--->24.080894) \t Saving The Model\n",
      "Saving the model, end of epoch 7\n",
      "Epoch 8 \t\t Training Loss: 0.011674800208552528 \t\t Validation Loss: 0.09421453996002674 \t\t Accuracy Loss: 2.502662069320679\n",
      "Validation Loss Decreased(24.080894--->23.553635) \t Saving The Model\n",
      "Saving the model, end of epoch 8\n",
      "Epoch 9 \t\t Training Loss: 0.011418814026265623 \t\t Validation Loss: 0.09298711109161377 \t\t Accuracy Loss: 2.5131231117248536\n",
      "Validation Loss Decreased(23.553635--->23.246778) \t Saving The Model\n",
      "Saving the model, end of epoch 9\n",
      "Epoch 10 \t\t Training Loss: 0.011163930218226422 \t\t Validation Loss: 0.09073687244951725 \t\t Accuracy Loss: 2.4363516731262207\n",
      "Validation Loss Decreased(23.246778--->22.684218) \t Saving The Model\n",
      "Saving the model, end of epoch 10\n",
      "Epoch 11 \t\t Training Loss: 0.01098053587210802 \t\t Validation Loss: 0.08861384850740432 \t\t Accuracy Loss: 2.4146954288482667\n",
      "Validation Loss Decreased(22.684218--->22.153462) \t Saving The Model\n",
      "Saving the model, end of epoch 11\n",
      "Epoch 12 \t\t Training Loss: 0.010818623994944243 \t\t Validation Loss: 0.0876400298923254 \t\t Accuracy Loss: 2.40839041519165\n",
      "Validation Loss Decreased(22.153462--->21.910007) \t Saving The Model\n",
      "Saving the model, end of epoch 12\n",
      "Epoch 13 \t\t Training Loss: 0.010691964776424792 \t\t Validation Loss: 0.0878456368893385 \t\t Accuracy Loss: 2.3876202087402345\n",
      "Saving the model, end of epoch 13\n",
      "Epoch 14 \t\t Training Loss: 0.010582897323406875 \t\t Validation Loss: 0.08643507739901543 \t\t Accuracy Loss: 2.3651524906158445\n",
      "Validation Loss Decreased(21.910007--->21.608769) \t Saving The Model\n",
      "Saving the model, end of epoch 14\n",
      "Epoch 15 \t\t Training Loss: 0.010483010810392258 \t\t Validation Loss: 0.08596901549398898 \t\t Accuracy Loss: 2.36799312210083\n",
      "Validation Loss Decreased(21.608769--->21.492254) \t Saving The Model\n",
      "Saving the model, end of epoch 15\n",
      "Epoch 16 \t\t Training Loss: 0.010405056067733481 \t\t Validation Loss: 0.08433587181568146 \t\t Accuracy Loss: 2.355515413284302\n",
      "Validation Loss Decreased(21.492254--->21.083968) \t Saving The Model\n",
      "Saving the model, end of epoch 16\n",
      "Epoch 17 \t\t Training Loss: 0.01032599261530434 \t\t Validation Loss: 0.08345923532545567 \t\t Accuracy Loss: 2.32767791557312\n",
      "Validation Loss Decreased(21.083968--->20.864809) \t Saving The Model\n",
      "Saving the model, end of epoch 17\n",
      "Epoch 18 \t\t Training Loss: 0.010259469459279658 \t\t Validation Loss: 0.08294602172076702 \t\t Accuracy Loss: 2.3231052093505857\n",
      "Validation Loss Decreased(20.864809--->20.736505) \t Saving The Model\n",
      "Saving the model, end of epoch 18\n",
      "Epoch 19 \t\t Training Loss: 0.010201415995481059 \t\t Validation Loss: 0.08397594921290874 \t\t Accuracy Loss: 2.344384994506836\n",
      "Saving the model, end of epoch 19\n",
      "Epoch 20 \t\t Training Loss: 0.01013128348245336 \t\t Validation Loss: 0.0822039985358715 \t\t Accuracy Loss: 2.308370834350586\n",
      "Validation Loss Decreased(20.736505--->20.551000) \t Saving The Model\n",
      "Saving the model, end of epoch 20\n",
      "Early stopping after 21 epochs.\n",
      "Finished Training\n"
     ]
    }
   ],
   "source": [
    "# ---- training code -----\n",
    "device = torch.device(DEVICE_ID)\n",
    "print('device: ' + str(device))\n",
    "net = Net().to(device=device)\n",
    "criterion = nn.MSELoss()\n",
    "criteriontwo = AccuracyLoss()\n",
    "optimizer = optim.SGD(net.parameters(), lr=hps['lr'])\n",
    "train_loader, val_loader = get_loaders(batch_size,device)\n",
    "\n",
    "min_valid_loss = np.inf\n",
    "\n",
    "'''if LOAD_CHKPT:\n",
    "    print('loading the model from the checkpoint')\n",
    "    model.load_state_dict(os.path.join(LOG_DIR,'checkpoint.pt'))'''\n",
    "\n",
    "print('training begins')\n",
    "\n",
    "hist_train = []\n",
    "hist_val = []\n",
    "\n",
    "hist_acc = []\n",
    "\n",
    "early_stopping_patience = 10\n",
    "best_val_loss = float('inf')\n",
    "current_patience = 0\n",
    "improvement_threshold=0.1\n",
    "\n",
    "\n",
    "for epoch in range(max_num_epoch):  \n",
    "    running_loss = 0.0 # training loss of the network\n",
    "    valid_loss = 0.0\n",
    "    running_acc = 0.0\n",
    "    for iteri, data in enumerate(train_loader, 0):\n",
    "        inputs, targets = data # inputs: low-resolution images, targets: high-resolution images.\n",
    "\n",
    "        optimizer.zero_grad() # zero the parameter gradients\n",
    "\n",
    "        # do forward, backward, SGD step\n",
    "        preds = net(inputs)\n",
    "        loss = criterion(preds, targets)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        running_loss += loss.item()\n",
    "        \n",
    "\n",
    "        '''# print loss\n",
    "        \n",
    "        print_n = 100 # feel free to change this constant\n",
    "        if iteri % print_n == (print_n-1):    # print every print_n mini-batches\n",
    "            print('[%d, %5d] network-loss: %.3f' %\n",
    "                  (epoch + 1, iteri + 1, running_loss / 100))\n",
    "            running_loss = 0.0\n",
    "            # note: you most probably want to track the progress on the validation set as well (needs to be implemented)'''\n",
    "\n",
    "        if (iteri==0) and VISUALIZE: \n",
    "            hw3utils.visualize_batch(inputs,preds,targets)\n",
    "\n",
    "    ### CHANGE ----------------\n",
    "            \n",
    "    for iteri,data in enumerate(val_loader,0):\n",
    "        inputs,targets = data\n",
    "\n",
    "        prediction = net(inputs)\n",
    "\n",
    "        vloss = criterion(prediction,targets)\n",
    "        vacc = criteriontwo(prediction,targets)\n",
    "        valid_loss += vloss.item() * inputs.size(0)\n",
    "        running_acc += vacc.item() * inputs.size(0)\n",
    "        \n",
    "    ### CHANGE ----------------\n",
    "        \n",
    "    hist_train.append(running_loss/len(train_loader))\n",
    "    hist_val.append(valid_loss / len(val_loader))\n",
    "    hist_acc.append(running_acc/len(val_loader))\n",
    "\n",
    "    val_loss = valid_loss / len(val_loader)\n",
    "\n",
    "    if val_loss < (1 - improvement_threshold) * best_val_loss:\n",
    "        best_val_loss = val_loss\n",
    "        current_patience = 0\n",
    "    else:\n",
    "        current_patience += 1\n",
    "        if current_patience >= early_stopping_patience:\n",
    "            print(f\"Early stopping after {epoch + 1} epochs.\")\n",
    "            break\n",
    "        \n",
    "    print(f'Epoch {epoch+1} \\t\\t Training Loss: {running_loss / len(train_loader)} \\t\\t Validation Loss: {valid_loss / len(val_loader)} \\t\\t Accuracy Loss: {running_acc/len(val_loader)}')\n",
    "    if min_valid_loss > valid_loss:\n",
    "        print(f'Validation Loss Decreased({min_valid_loss:.6f}--->{valid_loss:.6f}) \\t Saving The Model')\n",
    "        min_valid_loss = valid_loss\n",
    "\n",
    "\n",
    "    print('Saving the model, end of epoch %d' % (epoch+1))\n",
    "    if not os.path.exists(LOG_DIR):\n",
    "        os.makedirs(LOG_DIR)\n",
    "    torch.save(net.state_dict(), os.path.join(LOG_DIR,'checkpoint5.pt'))\n",
    "    #hw3utils.visualize_batch(inputs,preds,targets,os.path.join(LOG_DIR,'example.png'))\n",
    "\n",
    "print('Finished Training')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1038,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.010102706802889895\n",
      "0.0822039985358715\n"
     ]
    }
   ],
   "source": [
    "print(min(hist_train))\n",
    "print(min(hist_val))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1032,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Assuming you have hist_train and hist_val lists for training and validation losses\n",
    "\n",
    "# Plotting training and validation curves\n",
    "plt.figure(figsize=(10, 6))\n",
    "\n",
    "# Plot training loss with a solid line in blue\n",
    "plt.plot(hist_train[:23], label='Training Loss', color='blue', linestyle='-')\n",
    "\n",
    "# Plot validation loss with a dashed line in orange\n",
    "plt.plot(hist_val[:23], label='Validation Loss', color='orange', linestyle='--')\n",
    "\n",
    "# Add title and labels\n",
    "plt.title('Training and Validation Loss Curves')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "\n",
    "# Add grid lines\n",
    "plt.grid(True, linestyle='--', alpha=0.6)\n",
    "\n",
    "# Add legend\n",
    "plt.legend()\n",
    "\n",
    "# Save the plot as an image (optional)\n",
    "plt.savefig('training_validation_curves.png')\n",
    "\n",
    "# Show the plot\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1039,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting training and validation curves\n",
    "plt.figure(figsize=(10, 6))\n",
    "\n",
    "# Plot training loss with a solid line in blue\n",
    "plt.plot(hist_acc, label='Training Loss', color='blue', linestyle='-')\n",
    "\n",
    "\n",
    "\n",
    "# Add title and labels\n",
    "plt.title('Validation 12-margin Error Curve')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "\n",
    "# Add grid lines\n",
    "plt.grid(True, linestyle='--', alpha=0.6)\n",
    "\n",
    "# Add legend\n",
    "plt.legend()\n",
    "\n",
    "# Save the plot as an image (optional)\n",
    "plt.savefig('training_validation_curves.png')\n",
    "\n",
    "# Show the plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1021,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_accuracy(ground_truth, prediction, error_margin=12):\n",
    "    # Ensure the images have the same shape\n",
    "    assert ground_truth.shape == prediction.shape, \"Ground truth and prediction shapes must match\"\n",
    "    prediction = prediction.reshape(-1).astype(np.int64)\n",
    "    ground_truth = ground_truth.reshape(-1).astype(np.int64)\n",
    "    cur_acc = (np.abs(ground_truth - prediction) < 12).sum() / ground_truth.shape[0]\n",
    "\n",
    "    # Compute the ratio of correctly estimated pixels to all pixels\n",
    "    \n",
    "    return cur_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1069,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9621354166666667\n",
      "0.71/1.00\n"
     ]
    }
   ],
   "source": [
    "from utils import read_image\n",
    "predictions = []\n",
    "gt = []\n",
    "# Iterate over the dataset and make predictions\n",
    "with torch.no_grad():\n",
    "    for inputs, _ in val_loader:  # Assuming your dataloader provides inputs and labels\n",
    "        outputs = net(inputs)\n",
    "        predictions.append(outputs.numpy())  # Assuming outputs is a torch tensor\n",
    "        gt.append(_.numpy())\n",
    "gt = np.concatenate(gt, axis=0)\n",
    "gt = ((gt + 1) / 2 * 255).astype(np.uint8)\n",
    "gt = np.moveaxis(gt, 1, -1)\n",
    "\n",
    "# Convert the list of predictions to a numpy array\n",
    "predictions_np = np.concatenate(predictions, axis=0)\n",
    "predictions_np = ((predictions_np + 1) / 2 * 255).astype(np.uint8)\n",
    "predictions_np = np.moveaxis(predictions_np, 1, -1)\n",
    "\n",
    "\n",
    "# Save predictions to npy file\n",
    "np.save(\"estimations.npy\", predictions_np)\n",
    "\n",
    "acc = 0\n",
    "max_acc = 0\n",
    "for i in range(2000):\n",
    "\n",
    "    ground_truth = gt[i]\n",
    "    prediction = predictions_np[i]\n",
    "    current = compute_accuracy(ground_truth,prediction)\n",
    "    acc += current\n",
    "    if current>max_acc:\n",
    "        max_acc = current\n",
    "\n",
    "print(max_acc)\n",
    "\n",
    "acc /= 2000\n",
    "print(f\"{acc:.2f}/1.00\")\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1071,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "sample = 0\n",
    "# Assuming pred_img and gt_img are your prediction and ground truth images, each represented as NumPy arrays\n",
    "\n",
    "# Create a figure with two subplots\n",
    "fig, axes = plt.subplots(1, 2, figsize=(10, 5))\n",
    "\n",
    "# Plot the prediction on the first subplot\n",
    "axes[0].imshow(predictions_np[sample], cmap='gray')  # Use 'gray' colormap for grayscale images\n",
    "axes[0].set_title('Prediction')\n",
    "\n",
    "# Plot the ground truth on the second subplot\n",
    "axes[1].imshow(gt[sample], cmap='gray')\n",
    "axes[1].set_title('Ground Truth')\n",
    "\n",
    "# Adjust layout for better visualization\n",
    "plt.tight_layout()\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1059,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skimage import io\n",
    "def preprocess_image(image_path):\n",
    "    # Read the image\n",
    "    image = io.imread(image_path)\n",
    "\n",
    "    # Convert to torch tensor and normalize\n",
    "    transform = transforms.Compose([\n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize(mean=[0.5], std=[0.5])  # Adjust normalization values if needed\n",
    "    ])\n",
    "    input_tensor = transform(image)\n",
    "\n",
    "    # Add batch dimension\n",
    "    input_tensor = input_tensor.unsqueeze(0)\n",
    "\n",
    "    return input_tensor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1065,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2000, 80, 80, 3)\n",
      "['63.jpg', '823.jpg', '1409.jpg', '189.jpg', '77.jpg', '837.jpg', '1353.jpg', '1435.jpg', '1421.jpg', '1347.jpg', '638.jpg', '1390.jpg', '604.jpg', '162.jpg', '176.jpg', '88.jpg', '610.jpg', '1384.jpg', '1179.jpg', '1637.jpg', '1151.jpg', '1145.jpg', '1623.jpg', '1810.jpg', '1804.jpg', '348.jpg', '360.jpg', '406.jpg', '1192.jpg', '1186.jpg', '412.jpg', '1838.jpg', '374.jpg', '1019.jpg', '599.jpg', '1743.jpg', '1025.jpg', '1031.jpg', '1757.jpg', '228.jpg', '1964.jpg', '1970.jpg', '214.jpg', '1780.jpg', '1958.jpg', '572.jpg', '566.jpg', '1794.jpg', '200.jpg', '957.jpg', '1569.jpg', '943.jpg', '1227.jpg', '1541.jpg', '1555.jpg', '1233.jpg', '994.jpg', '758.jpg', '980.jpg', '770.jpg', '1582.jpg', '1596.jpg', '764.jpg', '1597.jpg', '765.jpg', '771.jpg', '1583.jpg', '981.jpg', '759.jpg', '995.jpg', '1554.jpg', '1232.jpg', '1226.jpg', '1540.jpg', '1568.jpg', '942.jpg', '956.jpg', '567.jpg', '201.jpg', '1795.jpg', '1959.jpg', '1781.jpg', '215.jpg', '573.jpg', '1971.jpg', '1965.jpg', '229.jpg', '1030.jpg', '1756.jpg', '1742.jpg', '1024.jpg', '598.jpg', '1018.jpg', '413.jpg', '1187.jpg', '375.jpg', '1839.jpg', '361.jpg', '1193.jpg', '407.jpg', '349.jpg', '1805.jpg', '1811.jpg', '1144.jpg', '1622.jpg', '1636.jpg', '1150.jpg', '1178.jpg', '177.jpg', '1385.jpg', '611.jpg', '89.jpg', '605.jpg', '1391.jpg', '163.jpg', '639.jpg', '1420.jpg', '1346.jpg', '1352.jpg', '1434.jpg', '188.jpg', '836.jpg', '76.jpg', '822.jpg', '62.jpg', '1408.jpg', '74.jpg', '1378.jpg', '834.jpg', '60.jpg', '820.jpg', '48.jpg', '1344.jpg', '808.jpg', '1422.jpg', '1436.jpg', '1350.jpg', '149.jpg', '613.jpg', '1387.jpg', '175.jpg', '161.jpg', '1393.jpg', '607.jpg', '388.jpg', '1608.jpg', '1620.jpg', '1146.jpg', '1152.jpg', '1634.jpg', '1807.jpg', '439.jpg', '1813.jpg', '377.jpg', '1185.jpg', '411.jpg', '405.jpg', '1191.jpg', '363.jpg', '1768.jpg', '1754.jpg', '1032.jpg', '1026.jpg', '1998.jpg', '1740.jpg', '1973.jpg', '559.jpg', '1967.jpg', '1797.jpg', '203.jpg', '565.jpg', '571.jpg', '217.jpg', '1783.jpg', '940.jpg', '798.jpg', '954.jpg', '1218.jpg', '1230.jpg', '1556.jpg', '1542.jpg', '968.jpg', '1224.jpg', '983.jpg', '997.jpg', '767.jpg', '1595.jpg', '1581.jpg', '773.jpg', '1580.jpg', '772.jpg', '766.jpg', '1594.jpg', '996.jpg', '982.jpg', '1543.jpg', '1225.jpg', '969.jpg', '1231.jpg', '1557.jpg', '1219.jpg', '955.jpg', '799.jpg', '941.jpg', '570.jpg', '1782.jpg', '216.jpg', '202.jpg', '1796.jpg', '564.jpg', '1966.jpg', '1972.jpg', '558.jpg', '1027.jpg', '1741.jpg', '1999.jpg', '1755.jpg', '1033.jpg', '1769.jpg', '1190.jpg', '404.jpg', '362.jpg', '376.jpg', '410.jpg', '1184.jpg', '438.jpg', '1812.jpg', '1806.jpg', '1153.jpg', '1635.jpg', '1621.jpg', '1147.jpg', '1609.jpg', '389.jpg', '160.jpg', '606.jpg', '1392.jpg', '1386.jpg', '612.jpg', '174.jpg', '148.jpg', '1437.jpg', '1351.jpg', '809.jpg', '1345.jpg', '49.jpg', '1423.jpg', '821.jpg', '61.jpg', '835.jpg', '1379.jpg', '75.jpg', '1427.jpg', '1341.jpg', '1355.jpg', '59.jpg', '819.jpg', '1433.jpg', '71.jpg', '831.jpg', '1369.jpg', '65.jpg', '825.jpg', '170.jpg', '616.jpg', '1382.jpg', '1396.jpg', '602.jpg', '164.jpg', '158.jpg', '1143.jpg', '1625.jpg', '1631.jpg', '1157.jpg', '1619.jpg', '399.jpg', '1180.jpg', '414.jpg', '372.jpg', '366.jpg', '400.jpg', '1194.jpg', '428.jpg', '1802.jpg', '1816.jpg', '1037.jpg', '1989.jpg', '1751.jpg', '1745.jpg', '1023.jpg', '1779.jpg', '560.jpg', '1792.jpg', '206.jpg', '212.jpg', '1786.jpg', '574.jpg', '1976.jpg', '1962.jpg', '548.jpg', '1553.jpg', '979.jpg', '1235.jpg', '1221.jpg', '1547.jpg', '945.jpg', '1209.jpg', '951.jpg', '789.jpg', '1590.jpg', '762.jpg', '776.jpg', '1584.jpg', '986.jpg', '992.jpg', '993.jpg', '987.jpg', '777.jpg', '1585.jpg', '1591.jpg', '763.jpg', '788.jpg', '950.jpg', '1208.jpg', '944.jpg', '1220.jpg', '1546.jpg', '1552.jpg', '1234.jpg', '978.jpg', '1963.jpg', '549.jpg', '1977.jpg', '1787.jpg', '213.jpg', '575.jpg', '561.jpg', '207.jpg', '1793.jpg', '1778.jpg', '1744.jpg', '1022.jpg', '1036.jpg', '1750.jpg', '1988.jpg', '1817.jpg', '429.jpg', '1803.jpg', '367.jpg', '1195.jpg', '401.jpg', '415.jpg', '1181.jpg', '373.jpg', '398.jpg', '1618.jpg', '1630.jpg', '1156.jpg', '1142.jpg', '1624.jpg', '159.jpg', '603.jpg', '1397.jpg', '165.jpg', '171.jpg', '1383.jpg', '617.jpg', '824.jpg', '64.jpg', '1368.jpg', '830.jpg', '70.jpg', '818.jpg', '58.jpg', '1354.jpg', '1432.jpg', '1426.jpg', '1340.jpg', '1430.jpg', '8.jpg', '1356.jpg', '1342.jpg', '1424.jpg', '198.jpg', '66.jpg', '826.jpg', '72.jpg', '832.jpg', '1418.jpg', '167.jpg', '1395.jpg', '601.jpg', '99.jpg', '615.jpg', '1381.jpg', '173.jpg', '629.jpg', '1154.jpg', '1632.jpg', '1626.jpg', '1140.jpg', '1168.jpg', '403.jpg', '1197.jpg', '1829.jpg', '365.jpg', '371.jpg', '1183.jpg', '417.jpg', '1815.jpg', '359.jpg', '1801.jpg', '1020.jpg', '1746.jpg', '1752.jpg', '1034.jpg', '588.jpg', '1008.jpg', '577.jpg', '211.jpg', '1785.jpg', '1791.jpg', '205.jpg', '1949.jpg', '563.jpg', '1961.jpg', '239.jpg', '1975.jpg', '1544.jpg', '1222.jpg', '1236.jpg', '1550.jpg', '1578.jpg', '952.jpg', '946.jpg', '1587.jpg', '775.jpg', '761.jpg', '1593.jpg', '749.jpg', '991.jpg', '985.jpg', '984.jpg', '990.jpg', '748.jpg', '760.jpg', '1592.jpg', '1586.jpg', '774.jpg', '947.jpg', '1579.jpg', '953.jpg', '1237.jpg', '1551.jpg', '1545.jpg', '1223.jpg', '1974.jpg', '238.jpg', '1960.jpg', '1948.jpg', '204.jpg', '1790.jpg', '562.jpg', '576.jpg', '1784.jpg', '210.jpg', '1009.jpg', '589.jpg', '1753.jpg', '1035.jpg', '1021.jpg', '1747.jpg', '1800.jpg', '358.jpg', '1814.jpg', '370.jpg', '416.jpg', '1182.jpg', '1196.jpg', '402.jpg', '364.jpg', '1828.jpg', '1169.jpg', '1627.jpg', '1141.jpg', '1155.jpg', '1633.jpg', '628.jpg', '1380.jpg', '614.jpg', '172.jpg', '166.jpg', '98.jpg', '600.jpg', '1394.jpg', '833.jpg', '73.jpg', '1419.jpg', '199.jpg', '827.jpg', '67.jpg', '1343.jpg', '1425.jpg', '9.jpg', '1431.jpg', '1357.jpg', '698.jpg', '840.jpg', '14.jpg', '1318.jpg', '854.jpg', '1456.jpg', '1330.jpg', '28.jpg', '1324.jpg', '868.jpg', '1442.jpg', '883.jpg', '897.jpg', '129.jpg', '1495.jpg', '101.jpg', '667.jpg', '673.jpg', '115.jpg', '1481.jpg', '1668.jpg', '1132.jpg', '1654.jpg', '1640.jpg', '1898.jpg', '1126.jpg', '459.jpg', '1873.jpg', '1867.jpg', '465.jpg', '303.jpg', '1697.jpg', '1683.jpg', '317.jpg', '471.jpg', '288.jpg', '1708.jpg', '1046.jpg', '1720.jpg', '1734.jpg', '1052.jpg', '1907.jpg', '1913.jpg', '539.jpg', '511.jpg', '1085.jpg', '277.jpg', '263.jpg', '1091.jpg', '505.jpg', '934.jpg', '1278.jpg', '920.jpg', '1522.jpg', '908.jpg', '1244.jpg', '1250.jpg', '1536.jpg', '1287.jpg', '713.jpg', '707.jpg', '1293.jpg', '1292.jpg', '706.jpg', '712.jpg', '1286.jpg', '1251.jpg', '1537.jpg', '1523.jpg', '1245.jpg', '909.jpg', '921.jpg', '1279.jpg', '935.jpg', '262.jpg', '504.jpg', '1090.jpg', '1084.jpg', '510.jpg', '276.jpg', '1912.jpg', '538.jpg', '1906.jpg', '1735.jpg', '1053.jpg', '1047.jpg', '1721.jpg', '1709.jpg', '289.jpg', '316.jpg', '1682.jpg', '470.jpg', '464.jpg', '1696.jpg', '302.jpg', '1866.jpg', '458.jpg', '1872.jpg', '1899.jpg', '1641.jpg', '1127.jpg', '1133.jpg', '1655.jpg', '1669.jpg', '672.jpg', '1480.jpg', '114.jpg', '100.jpg', '1494.jpg', '666.jpg', '896.jpg', '128.jpg', '882.jpg', '869.jpg', '1325.jpg', '29.jpg', '1443.jpg', '1457.jpg', '1331.jpg', '855.jpg', '1319.jpg', '15.jpg', '841.jpg', '699.jpg', '17.jpg', '857.jpg', '843.jpg', '1469.jpg', '1441.jpg', '1327.jpg', '1333.jpg', '1455.jpg', '894.jpg', '880.jpg', '658.jpg', '116.jpg', '1482.jpg', '670.jpg', '664.jpg', '1496.jpg', '102.jpg', '1119.jpg', '499.jpg', '1125.jpg', '1643.jpg', '1657.jpg', '1131.jpg', '1864.jpg', '328.jpg', '1870.jpg', '472.jpg', '1858.jpg', '1680.jpg', '314.jpg', '300.jpg', '1694.jpg', '466.jpg', '1079.jpg', '1051.jpg', '1737.jpg', '1723.jpg', '1045.jpg', '1910.jpg', '248.jpg', '1904.jpg', '1092.jpg', '506.jpg', '260.jpg', '274.jpg', '1938.jpg', '512.jpg', '1086.jpg', '1509.jpg', '923.jpg', '937.jpg', '1535.jpg', '1253.jpg', '1247.jpg', '1521.jpg', '738.jpg', '704.jpg', '1290.jpg', '1284.jpg', '710.jpg', '711.jpg', '1285.jpg', '1291.jpg', '705.jpg', '739.jpg', '1246.jpg', '1520.jpg', '1534.jpg', '1252.jpg', '936.jpg', '1508.jpg', '922.jpg', '1939.jpg', '275.jpg', '1087.jpg', '513.jpg', '507.jpg', '1093.jpg', '261.jpg', '1905.jpg', '249.jpg', '1911.jpg', '1722.jpg', '1044.jpg', '1050.jpg', '1736.jpg', '1078.jpg', '1695.jpg', '301.jpg', '467.jpg', '473.jpg', '315.jpg', '1681.jpg', '1859.jpg', '1871.jpg', '329.jpg', '1865.jpg', '1656.jpg', '1130.jpg', '1124.jpg', '1642.jpg', '498.jpg', '1118.jpg', '665.jpg', '103.jpg', '1497.jpg', '1483.jpg', '117.jpg', '671.jpg', '659.jpg', '881.jpg', '895.jpg', '1332.jpg', '1454.jpg', '1440.jpg', '1326.jpg', '842.jpg', '1468.jpg', '856.jpg', '16.jpg', '1322.jpg', '1444.jpg', '1450.jpg', '1336.jpg', '12.jpg', '852.jpg', '1478.jpg', '846.jpg', '675.jpg', '113.jpg', '1487.jpg', '1493.jpg', '107.jpg', '661.jpg', '891.jpg', '649.jpg', '885.jpg', '1646.jpg', '1120.jpg', '1134.jpg', '1652.jpg', '488.jpg', '1108.jpg', '1685.jpg', '311.jpg', '477.jpg', '463.jpg', '1849.jpg', '305.jpg', '1691.jpg', '1861.jpg', '1875.jpg', '339.jpg', '1732.jpg', '1054.jpg', '1040.jpg', '1726.jpg', '1068.jpg', '265.jpg', '1929.jpg', '1097.jpg', '503.jpg', '517.jpg', '1083.jpg', '271.jpg', '259.jpg', '1915.jpg', '1901.jpg', '1256.jpg', '1530.jpg', '1524.jpg', '1242.jpg', '926.jpg', '1518.jpg', '932.jpg', '701.jpg', '1295.jpg', '1281.jpg', '715.jpg', '729.jpg', '728.jpg', '714.jpg', '1280.jpg', '1294.jpg', '700.jpg', '1519.jpg', '933.jpg', '927.jpg', '1525.jpg', '1243.jpg', '1257.jpg', '1531.jpg', '1900.jpg', '1914.jpg', '258.jpg', '1082.jpg', '516.jpg', '270.jpg', '1928.jpg', '264.jpg', '502.jpg', '1096.jpg', '1069.jpg', '1041.jpg', '1727.jpg', '1733.jpg', '1055.jpg', '338.jpg', '1874.jpg', '1860.jpg', '462.jpg', '1690.jpg', '304.jpg', '1848.jpg', '310.jpg', '1684.jpg', '476.jpg', '1109.jpg', '489.jpg', '1135.jpg', '1653.jpg', '1647.jpg', '1121.jpg', '884.jpg', '648.jpg', '890.jpg', '106.jpg', '1492.jpg', '660.jpg', '674.jpg', '1486.jpg', '112.jpg', '847.jpg', '853.jpg', '13.jpg', '1479.jpg', '1451.jpg', '1337.jpg', '1323.jpg', '1445.jpg', '1335.jpg', '39.jpg', '879.jpg', '1453.jpg', '1447.jpg', '1321.jpg', '1309.jpg', '845.jpg', '689.jpg', '11.jpg', '851.jpg', '662.jpg', '1490.jpg', '104.jpg', '110.jpg', '1484.jpg', '676.jpg', '886.jpg', '138.jpg', '892.jpg', '1651.jpg', '1889.jpg', '1137.jpg', '1123.jpg', '1645.jpg', '1679.jpg', '306.jpg', '1692.jpg', '460.jpg', '474.jpg', '1686.jpg', '312.jpg', '1876.jpg', '448.jpg', '1862.jpg', '1725.jpg', '1043.jpg', '1057.jpg', '1731.jpg', '1719.jpg', '299.jpg', '272.jpg', '514.jpg', '1080.jpg', '1094.jpg', '500.jpg', '266.jpg', '1902.jpg', '528.jpg', '1916.jpg', '1241.jpg', '1527.jpg', '1533.jpg', '919.jpg', '1255.jpg', '931.jpg', '925.jpg', '1269.jpg', '1282.jpg', '716.jpg', '702.jpg', '1296.jpg', '1297.jpg', '703.jpg', '717.jpg', '1283.jpg', '1268.jpg', '924.jpg', '930.jpg', '1532.jpg', '1254.jpg', '918.jpg', '1240.jpg', '1526.jpg', '1917.jpg', '1903.jpg', '529.jpg', '501.jpg', '1095.jpg', '267.jpg', '273.jpg', '1081.jpg', '515.jpg', '298.jpg', '1718.jpg', '1056.jpg', '1730.jpg', '1724.jpg', '1042.jpg', '449.jpg', '1863.jpg', '1877.jpg', '475.jpg', '313.jpg', '1687.jpg', '1693.jpg', '307.jpg', '461.jpg', '1678.jpg', '1122.jpg', '1644.jpg', '1888.jpg', '1650.jpg', '1136.jpg', '893.jpg', '887.jpg', '139.jpg', '1485.jpg', '111.jpg', '677.jpg', '663.jpg', '105.jpg', '1491.jpg', '850.jpg', '10.jpg', '688.jpg', '844.jpg', '1308.jpg', '1446.jpg', '1320.jpg', '878.jpg', '38.jpg', '1334.jpg', '1452.jpg', '21.jpg', '861.jpg', '35.jpg', '1339.jpg', '875.jpg', '1477.jpg', '685.jpg', '1311.jpg', '1305.jpg', '691.jpg', '849.jpg', '1463.jpg', '1488.jpg', '108.jpg', '120.jpg', '646.jpg', '652.jpg', '134.jpg', '1649.jpg', '1891.jpg', '1885.jpg', '1113.jpg', '487.jpg', '1675.jpg', '1661.jpg', '493.jpg', '1107.jpg', '478.jpg', '1852.jpg', '1846.jpg', '444.jpg', '322.jpg', '336.jpg', '450.jpg', '1729.jpg', '1067.jpg', '1701.jpg', '295.jpg', '281.jpg', '1715.jpg', '1073.jpg', '1098.jpg', '1926.jpg', '1932.jpg', '518.jpg', '530.jpg', '256.jpg', '242.jpg', '524.jpg', '915.jpg', '1259.jpg', '901.jpg', '1503.jpg', '929.jpg', '1265.jpg', '1271.jpg', '1517.jpg', '732.jpg', '726.jpg', '727.jpg', '733.jpg', '1270.jpg', '1516.jpg', '1502.jpg', '1264.jpg', '928.jpg', '900.jpg', '1258.jpg', '914.jpg', '243.jpg', '525.jpg', '531.jpg', '257.jpg', '1933.jpg', '519.jpg', '1099.jpg', '1927.jpg', '1714.jpg', '280.jpg', '1072.jpg', '1066.jpg', '294.jpg', '1700.jpg', '1728.jpg', '337.jpg', '451.jpg', '445.jpg', '323.jpg', '1847.jpg', '479.jpg', '1853.jpg', '1660.jpg', '1106.jpg', '492.jpg', '486.jpg', '1112.jpg', '1674.jpg', '1884.jpg', '1890.jpg', '1648.jpg', '653.jpg', '135.jpg', '121.jpg', '647.jpg', '109.jpg', '1489.jpg', '848.jpg', '690.jpg', '1304.jpg', '1462.jpg', '1476.jpg', '1310.jpg', '684.jpg', '874.jpg', '1338.jpg', '34.jpg', '860.jpg', '20.jpg', '36.jpg', '876.jpg', '22.jpg', '862.jpg', '1448.jpg', '1460.jpg', '1306.jpg', '692.jpg', '686.jpg', '1312.jpg', '1474.jpg', '679.jpg', '137.jpg', '889.jpg', '651.jpg', '645.jpg', '123.jpg', '1138.jpg', '1886.jpg', '1892.jpg', '490.jpg', '1104.jpg', '1662.jpg', '1676.jpg', '1110.jpg', '484.jpg', '1845.jpg', '309.jpg', '1851.jpg', '1689.jpg', '453.jpg', '1879.jpg', '335.jpg', '321.jpg', '447.jpg', '1058.jpg', '1070.jpg', '282.jpg', '1716.jpg', '1702.jpg', '296.jpg', '1064.jpg', '1931.jpg', '269.jpg', '1925.jpg', '527.jpg', '241.jpg', '255.jpg', '1919.jpg', '533.jpg', '1528.jpg', '902.jpg', '916.jpg', '1514.jpg', '1272.jpg', '1266.jpg', '1500.jpg', '719.jpg', '1299.jpg', '725.jpg', '731.jpg', '730.jpg', '724.jpg', '1298.jpg', '718.jpg', '1267.jpg', '1501.jpg', '1515.jpg', '1273.jpg', '917.jpg', '1529.jpg', '903.jpg', '1918.jpg', '254.jpg', '532.jpg', '526.jpg', '240.jpg', '1924.jpg', '268.jpg', '1930.jpg', '297.jpg', '1703.jpg', '1065.jpg', '1071.jpg', '1717.jpg', '283.jpg', '1059.jpg', '320.jpg', '446.jpg', '452.jpg', '334.jpg', '1878.jpg', '1688.jpg', '1850.jpg', '308.jpg', '1844.jpg', '1677.jpg', '485.jpg', '1111.jpg', '1105.jpg', '491.jpg', '1663.jpg', '1893.jpg', '1139.jpg', '1887.jpg', '644.jpg', '122.jpg', '136.jpg', '650.jpg', '888.jpg', '678.jpg', '1313.jpg', '687.jpg', '1475.jpg', '1461.jpg', '693.jpg', '1307.jpg', '863.jpg', '23.jpg', '1449.jpg', '877.jpg', '37.jpg', '1303.jpg', '697.jpg', '1465.jpg', '1471.jpg', '683.jpg', '1317.jpg', '33.jpg', '873.jpg', '1459.jpg', '27.jpg', '867.jpg', '654.jpg', '132.jpg', '126.jpg', '898.jpg', '640.jpg', '668.jpg', '1667.jpg', '495.jpg', '1101.jpg', '1115.jpg', '481.jpg', '1673.jpg', '1883.jpg', '1129.jpg', '1897.jpg', '330.jpg', '456.jpg', '442.jpg', '1868.jpg', '324.jpg', '1840.jpg', '1698.jpg', '1854.jpg', '318.jpg', '287.jpg', '1713.jpg', '1075.jpg', '1061.jpg', '1707.jpg', '293.jpg', '1049.jpg', '244.jpg', '1908.jpg', '522.jpg', '536.jpg', '250.jpg', '278.jpg', '1934.jpg', '1920.jpg', '1277.jpg', '1511.jpg', '1505.jpg', '1263.jpg', '907.jpg', '1539.jpg', '913.jpg', '720.jpg', '734.jpg', '1288.jpg', '708.jpg', '709.jpg', '1289.jpg', '735.jpg', '721.jpg', '1538.jpg', '912.jpg', '906.jpg', '1504.jpg', '1262.jpg', '1276.jpg', '1510.jpg', '1921.jpg', '1935.jpg', '279.jpg', '537.jpg', '251.jpg', '1909.jpg', '245.jpg', '523.jpg', '1048.jpg', '1060.jpg', '292.jpg', '1706.jpg', '1712.jpg', '286.jpg', '1074.jpg', '319.jpg', '1855.jpg', '1699.jpg', '1841.jpg', '443.jpg', '325.jpg', '1869.jpg', '331.jpg', '457.jpg', '1128.jpg', '1896.jpg', '1882.jpg', '480.jpg', '1114.jpg', '1672.jpg', '1666.jpg', '1100.jpg', '494.jpg', '669.jpg', '127.jpg', '641.jpg', '899.jpg', '655.jpg', '133.jpg', '866.jpg', '26.jpg', '872.jpg', '32.jpg', '1458.jpg', '1470.jpg', '1316.jpg', '682.jpg', '696.jpg', '1302.jpg', '1464.jpg', '680.jpg', '1314.jpg', '18.jpg', '858.jpg', '1472.jpg', '1466.jpg', '1300.jpg', '694.jpg', '1328.jpg', '24.jpg', '864.jpg', '30.jpg', '870.jpg', '643.jpg', '125.jpg', '131.jpg', '657.jpg', '119.jpg', '1499.jpg', '1670.jpg', '1116.jpg', '482.jpg', '496.jpg', '1102.jpg', '1664.jpg', '1894.jpg', '1658.jpg', '1880.jpg', '327.jpg', '441.jpg', '455.jpg', '333.jpg', '1857.jpg', '469.jpg', '1843.jpg', '1704.jpg', '290.jpg', '1062.jpg', '1076.jpg', '284.jpg', '1710.jpg', '1738.jpg', '253.jpg', '535.jpg', '521.jpg', '247.jpg', '1923.jpg', '509.jpg', '1089.jpg', '1937.jpg', '1260.jpg', '1506.jpg', '1512.jpg', '938.jpg', '1274.jpg', '910.jpg', '904.jpg', '1248.jpg', '737.jpg', '723.jpg', '722.jpg', '736.jpg', '1249.jpg', '905.jpg', '911.jpg', '1513.jpg', '1275.jpg', '939.jpg', '1261.jpg', '1507.jpg', '1088.jpg', '1936.jpg', '1922.jpg', '508.jpg', '520.jpg', '246.jpg', '252.jpg', '534.jpg', '1739.jpg', '1077.jpg', '1711.jpg', '285.jpg', '291.jpg', '1705.jpg', '1063.jpg', '468.jpg', '1842.jpg', '1856.jpg', '454.jpg', '332.jpg', '326.jpg', '440.jpg', '1881.jpg', '1659.jpg', '1895.jpg', '1103.jpg', '497.jpg', '1665.jpg', '1671.jpg', '483.jpg', '1117.jpg', '1498.jpg', '118.jpg', '130.jpg', '656.jpg', '642.jpg', '124.jpg', '871.jpg', '31.jpg', '865.jpg', '25.jpg', '1329.jpg', '1467.jpg', '695.jpg', '1301.jpg', '859.jpg', '19.jpg', '1315.jpg', '681.jpg', '1473.jpg', '42.jpg', '802.jpg', '1428.jpg', '4.jpg', '56.jpg', '816.jpg', '1372.jpg', '180.jpg', '1414.jpg', '1400.jpg', '194.jpg', '1366.jpg', '81.jpg', '619.jpg', '95.jpg', '1399.jpg', '625.jpg', '143.jpg', '157.jpg', '631.jpg', '1158.jpg', '1616.jpg', '382.jpg', '1170.jpg', '1164.jpg', '396.jpg', '1602.jpg', '1831.jpg', '1825.jpg', '369.jpg', '341.jpg', '427.jpg', '433.jpg', '1819.jpg', '355.jpg', '1986.jpg', '1038.jpg', '1992.jpg', '1762.jpg', '1004.jpg', '590.jpg', '584.jpg', '1010.jpg', '1776.jpg', '209.jpg', '1945.jpg', '1789.jpg', '1951.jpg', '235.jpg', '1979.jpg', '553.jpg', '547.jpg', '221.jpg', '976.jpg', '1548.jpg', '962.jpg', '792.jpg', '1206.jpg', '1560.jpg', '1574.jpg', '1212.jpg', '786.jpg', '779.jpg', '751.jpg', '989.jpg', '745.jpg', '744.jpg', '988.jpg', '750.jpg', '778.jpg', '1575.jpg', '787.jpg', '1213.jpg', '1207.jpg', '793.jpg', '1561.jpg', '1549.jpg', '963.jpg', '977.jpg', '546.jpg', '220.jpg', '1978.jpg', '234.jpg', '552.jpg', '1950.jpg', '1788.jpg', '1944.jpg', '208.jpg', '1011.jpg', '585.jpg', '1777.jpg', '1763.jpg', '591.jpg', '1005.jpg', '1993.jpg', '1987.jpg', '1039.jpg', '432.jpg', '354.jpg', '1818.jpg', '340.jpg', '426.jpg', '368.jpg', '1824.jpg', '1830.jpg', '1165.jpg', '1603.jpg', '397.jpg', '383.jpg', '1617.jpg', '1171.jpg', '1159.jpg', '156.jpg', '630.jpg', '624.jpg', '142.jpg', '1398.jpg', '94.jpg', '618.jpg', '80.jpg', '195.jpg', '1401.jpg', '1367.jpg', '1373.jpg', '1415.jpg', '181.jpg', '5.jpg', '817.jpg', '57.jpg', '803.jpg', '43.jpg', '1429.jpg', '55.jpg', '1359.jpg', '815.jpg', '7.jpg', '41.jpg', '801.jpg', '69.jpg', '1365.jpg', '829.jpg', '1403.jpg', '197.jpg', '183.jpg', '1417.jpg', '1371.jpg', '96.jpg', '168.jpg', '82.jpg', '632.jpg', '154.jpg', '140.jpg', '626.jpg', '1629.jpg', '395.jpg', '1601.jpg', '1167.jpg', '1173.jpg', '1615.jpg', '381.jpg', '1826.jpg', '1198.jpg', '418.jpg', '1832.jpg', '356.jpg', '430.jpg', '424.jpg', '342.jpg', '1991.jpg', '1749.jpg', '1985.jpg', '1775.jpg', '587.jpg', '1013.jpg', '1007.jpg', '593.jpg', '1761.jpg', '1952.jpg', '578.jpg', '1946.jpg', '222.jpg', '544.jpg', '550.jpg', '236.jpg', '961.jpg', '975.jpg', '1239.jpg', '1211.jpg', '785.jpg', '1577.jpg', '1563.jpg', '949.jpg', '791.jpg', '1205.jpg', '1588.jpg', '746.jpg', '752.jpg', '753.jpg', '747.jpg', '1589.jpg', '1562.jpg', '1204.jpg', '790.jpg', '948.jpg', '784.jpg', '1210.jpg', '1576.jpg', '1238.jpg', '974.jpg', '960.jpg', '551.jpg', '237.jpg', '223.jpg', '545.jpg', '1947.jpg', '1953.jpg', '579.jpg', '592.jpg', '1006.jpg', '1760.jpg', '1774.jpg', '1012.jpg', '586.jpg', '1984.jpg', '1748.jpg', '1990.jpg', '425.jpg', '343.jpg', '357.jpg', '431.jpg', '419.jpg', '1833.jpg', '1827.jpg', '1199.jpg', '1172.jpg', '380.jpg', '1614.jpg', '1600.jpg', '394.jpg', '1166.jpg', '1628.jpg', '141.jpg', '627.jpg', '633.jpg', '155.jpg', '83.jpg', '97.jpg', '169.jpg', '1416.jpg', '182.jpg', '1370.jpg', '828.jpg', '1364.jpg', '68.jpg', '196.jpg', '1402.jpg', '800.jpg', '40.jpg', '814.jpg', '1358.jpg', '54.jpg', '6.jpg', '1406.jpg', '192.jpg', '1360.jpg', '1374.jpg', '78.jpg', '838.jpg', '186.jpg', '1412.jpg', '2.jpg', '50.jpg', '810.jpg', '1348.jpg', '44.jpg', '804.jpg', '151.jpg', '637.jpg', '623.jpg', '145.jpg', '93.jpg', '87.jpg', '179.jpg', '1162.jpg', '390.jpg', '1604.jpg', '1610.jpg', '384.jpg', '1176.jpg', '1638.jpg', '435.jpg', '353.jpg', '347.jpg', '421.jpg', '409.jpg', '1823.jpg', '1837.jpg', '1189.jpg', '582.jpg', '1016.jpg', '1770.jpg', '1764.jpg', '1002.jpg', '596.jpg', '1994.jpg', '1980.jpg', '1758.jpg', '541.jpg', '227.jpg', '233.jpg', '555.jpg', '1957.jpg', '1943.jpg', '569.jpg', '1572.jpg', '958.jpg', '1214.jpg', '780.jpg', '794.jpg', '1200.jpg', '1566.jpg', '964.jpg', '1228.jpg', '970.jpg', '743.jpg', '757.jpg', '1599.jpg', '1598.jpg', '756.jpg', '742.jpg', '971.jpg', '1229.jpg', '965.jpg', '1201.jpg', '795.jpg', '1567.jpg', '1573.jpg', '781.jpg', '1215.jpg', '959.jpg', '1942.jpg', '568.jpg', '1956.jpg', '232.jpg', '554.jpg', '540.jpg', '226.jpg', '1759.jpg', '1981.jpg', '1995.jpg', '1765.jpg', '597.jpg', '1003.jpg', '1017.jpg', '583.jpg', '1771.jpg', '1836.jpg', '1188.jpg', '408.jpg', '1822.jpg', '346.jpg', '420.jpg', '434.jpg', '352.jpg', '1639.jpg', '385.jpg', '1611.jpg', '1177.jpg', '1163.jpg', '1605.jpg', '391.jpg', '86.jpg', '178.jpg', '92.jpg', '622.jpg', '144.jpg', '150.jpg', '636.jpg', '805.jpg', '45.jpg', '1349.jpg', '3.jpg', '811.jpg', '51.jpg', '839.jpg', '79.jpg', '1375.jpg', '1413.jpg', '187.jpg', '193.jpg', '1407.jpg', '1361.jpg', '185.jpg', '1411.jpg', '1377.jpg', '1363.jpg', '1405.jpg', '191.jpg', '47.jpg', '807.jpg', '53.jpg', '813.jpg', '1.jpg', '1439.jpg', '146.jpg', '620.jpg', '634.jpg', '152.jpg', '1388.jpg', '84.jpg', '608.jpg', '90.jpg', '1175.jpg', '1613.jpg', '387.jpg', '393.jpg', '1607.jpg', '1161.jpg', '1149.jpg', '422.jpg', '1808.jpg', '344.jpg', '350.jpg', '436.jpg', '1834.jpg', '378.jpg', '1820.jpg', '1001.jpg', '595.jpg', '1767.jpg', '1773.jpg', '581.jpg', '1015.jpg', '1983.jpg', '1997.jpg', '1029.jpg', '556.jpg', '230.jpg', '224.jpg', '1968.jpg', '542.jpg', '1798.jpg', '1940.jpg', '218.jpg', '1954.jpg', '1565.jpg', '797.jpg', '1203.jpg', '1217.jpg', '783.jpg', '1571.jpg', '1559.jpg', '973.jpg', '967.jpg', '754.jpg', '740.jpg', '998.jpg', '768.jpg', '769.jpg', '999.jpg', '741.jpg', '755.jpg', '966.jpg', '1558.jpg', '972.jpg', '782.jpg', '1216.jpg', '1570.jpg', '1564.jpg', '1202.jpg', '796.jpg', '1955.jpg', '219.jpg', '1941.jpg', '1799.jpg', '1969.jpg', '225.jpg', '543.jpg', '557.jpg', '231.jpg', '1996.jpg', '1028.jpg', '1982.jpg', '1772.jpg', '1014.jpg', '580.jpg', '594.jpg', '1000.jpg', '1766.jpg', '1821.jpg', '379.jpg', '1835.jpg', '351.jpg', '437.jpg', '423.jpg', '345.jpg', '1809.jpg', '1148.jpg', '1606.jpg', '392.jpg', '1160.jpg', '1174.jpg', '386.jpg', '1612.jpg', '91.jpg', '609.jpg', '85.jpg', '1389.jpg', '635.jpg', '153.jpg', '147.jpg', '621.jpg', '812.jpg', '52.jpg', '1438.jpg', '0.jpg', '806.jpg', '46.jpg', '1362.jpg', '190.jpg', '1404.jpg', '1410.jpg', '184.jpg', '1376.jpg']\n"
     ]
    }
   ],
   "source": [
    "folder_path = \"images_grayscale\"\n",
    "\n",
    "# Get a list of all files in the folder\n",
    "all_files = os.listdir(folder_path)\n",
    "\n",
    "# Filter out only image files (you may need to customize the image file extensions)\n",
    "image_files = [file for file in all_files if file.lower().endswith(('.png', '.jpg', '.jpeg'))]\n",
    "estimations = []\n",
    "for image_path in image_files:\n",
    "    # Preprocess the image\n",
    "    input_tensor = preprocess_image(\"images_grayscale/\"+image_path)\n",
    "\n",
    "    # Make inference\n",
    "    with torch.no_grad():\n",
    "        output = net(input_tensor)\n",
    "        estimations.append(output.numpy())\n",
    "\n",
    "# Convert the list of predictions to a numpy array\n",
    "predictions_np = np.concatenate(estimations, axis=0)\n",
    "predictions_np = ((predictions_np + 1) / 2 * 255).astype(np.uint8)\n",
    "predictions_np = np.moveaxis(predictions_np, 1, -1)\n",
    "\n",
    "\n",
    "print(predictions_np.shape)\n",
    "\n",
    "print(image_files)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1068,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.save(\"estimations.npy\", predictions_np)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "torchenv",
   "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}