diff --git a/pythonFile/.ipynb_checkpoints/final-checkpoint.ipynb b/pythonFile/.ipynb_checkpoints/final-checkpoint.ipynb
index 7468ef6..5262cc7 100644
--- a/pythonFile/.ipynb_checkpoints/final-checkpoint.ipynb
+++ b/pythonFile/.ipynb_checkpoints/final-checkpoint.ipynb
@@ -502,7 +502,8 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# 単純マッチング"
+    "# テーブルデータ\n",
+    "## 単純マッチング"
    ]
   },
   {
@@ -632,7 +633,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# kNN"
+    "## kNN"
    ]
   },
   {
@@ -713,7 +714,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# NeuralNet"
+    "## NeuralNet"
    ]
   },
   {
@@ -780,27 +781,27 @@
       "--------epoch0--------\n",
       "train accuracy:0.05\n",
       "train loss:3.0\n",
-      "test accuracy:0.0345\n",
+      "test accuracy:0.19\n",
       "--------------------\n",
       "--------epoch50--------\n",
-      "train accuracy:0.74\n",
-      "train loss:0.8\n",
-      "test accuracy:0.397\n",
+      "train accuracy:0.785\n",
+      "train loss:0.802\n",
+      "test accuracy:0.379\n",
       "--------------------\n",
       "--------epoch100--------\n",
-      "train accuracy:1.0\n",
-      "train loss:0.0425\n",
-      "test accuracy:0.483\n",
+      "train accuracy:0.995\n",
+      "train loss:0.108\n",
+      "test accuracy:0.466\n",
       "--------------------\n",
       "--------epoch150--------\n",
       "train accuracy:1.0\n",
-      "train loss:0.0074\n",
+      "train loss:0.0162\n",
       "test accuracy:0.448\n",
       "--------------------\n",
       "--------epoch199--------\n",
       "train accuracy:1.0\n",
-      "train loss:0.00245\n",
-      "test accuracy:0.448\n",
+      "train loss:0.00594\n",
+      "test accuracy:0.466\n",
       "--------------------\n",
       "max accuracy_test: 0.4827586206896552\n"
      ]
@@ -852,7 +853,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 2 Axes>"
       ]
@@ -880,7 +881,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# LightGBM"
+    "## LightGBM"
    ]
   },
   {
@@ -895,7 +896,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -910,13 +911,13 @@
     "    'num_leaves': 15,\n",
     "    'min_data_in_leaf': 10,\n",
     "    'num_iteration': 200,\n",
-    "    'verbose': -1\n",
+    "    'verbose': -1,\n",
     "}"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 28,
    "metadata": {
     "scrolled": false
    },
@@ -1151,7 +1152,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 29,
    "metadata": {
     "scrolled": true
    },
@@ -1162,7 +1163,7 @@
        "0.5"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1173,7 +1174,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 30,
    "metadata": {
     "scrolled": false
    },
@@ -1200,12 +1201,44 @@
     "plt.tight_layout()"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 画像を直接入れる系"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "# 画像読み込み\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## ピクセルマッチング"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# ピクセルマッチングと同じ\n",
+    "result = []\n",
+    "test_min = []\n",
+    "prediction_df = pd.DataFrame(query_target_df)\n",
+    "for i in range(query_featture_df.shape[0]):\n",
+    "    minimum_id = (db_feature_df - query_featture_df.iloc[i]).abs().sum(axis=1).idxmin()\n",
+    "    test_min.append((db_feature_df - query_featture_df.iloc[i]).abs().sum(axis=1))\n",
+    "    prediction_df.loc[i, \"predict\"] = db_target_df[minimum_id]"
+   ]
   },
   {
    "cell_type": "code",
diff --git a/pythonFile/final.ipynb b/pythonFile/final.ipynb
index 7468ef6..ec48779 100644
--- a/pythonFile/final.ipynb
+++ b/pythonFile/final.ipynb
@@ -9,7 +9,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 147,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -24,14 +24,19 @@
     "# kNN\n",
     "from sklearn.neighbors import KNeighborsClassifier\n",
     "\n",
-    "# NeuralNet\n",
+    "# NeuralNet, CNN\n",
+    "from torchvision import transforms\n",
     "import torch\n",
-    "from torch import nn\n",
-    "from torch import optim\n",
+    "from torch import nn, optim\n",
+    "from torch.utils.data import TensorDataset, DataLoader\n",
     "\n",
     "# LightGBM\n",
     "import lightgbm as lgb\n",
-    "from sklearn.metrics import accuracy_score"
+    "from sklearn.metrics import accuracy_score\n",
+    "\n",
+    "# 画像読み込み\n",
+    "from pathlib import Path\n",
+    "import cv2"
    ]
   },
   {
@@ -502,27 +507,26 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# 単純マッチング"
+    "# テーブルデータ\n",
+    "## 単純マッチング"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 125,
    "metadata": {},
    "outputs": [],
    "source": [
     "result = []\n",
-    "test_min = []\n",
     "prediction_df = pd.DataFrame(query_target_df)\n",
     "for i in range(query_featture_df.shape[0]):\n",
     "    minimum_id = (db_feature_df - query_featture_df.iloc[i]).abs().sum(axis=1).idxmin()\n",
-    "    test_min.append((db_feature_df - query_featture_df.iloc[i]).abs().sum(axis=1))\n",
     "    prediction_df.loc[i, \"predict\"] = db_target_df[minimum_id]"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 126,
    "metadata": {},
    "outputs": [
     {
@@ -589,7 +593,7 @@
        "4     0.0      0.0"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 126,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -600,7 +604,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 127,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -610,7 +614,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 128,
    "metadata": {},
    "outputs": [
     {
@@ -619,7 +623,7 @@
        "0.7758620689655172"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 128,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -632,7 +636,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# kNN"
+    "## kNN"
    ]
   },
   {
@@ -713,7 +717,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# NeuralNet"
+    "## NeuralNet"
    ]
   },
   {
@@ -780,27 +784,27 @@
       "--------epoch0--------\n",
       "train accuracy:0.05\n",
       "train loss:3.0\n",
-      "test accuracy:0.0345\n",
+      "test accuracy:0.19\n",
       "--------------------\n",
       "--------epoch50--------\n",
-      "train accuracy:0.74\n",
-      "train loss:0.8\n",
-      "test accuracy:0.397\n",
+      "train accuracy:0.785\n",
+      "train loss:0.802\n",
+      "test accuracy:0.379\n",
       "--------------------\n",
       "--------epoch100--------\n",
-      "train accuracy:1.0\n",
-      "train loss:0.0425\n",
-      "test accuracy:0.483\n",
+      "train accuracy:0.995\n",
+      "train loss:0.108\n",
+      "test accuracy:0.466\n",
       "--------------------\n",
       "--------epoch150--------\n",
       "train accuracy:1.0\n",
-      "train loss:0.0074\n",
+      "train loss:0.0162\n",
       "test accuracy:0.448\n",
       "--------------------\n",
       "--------epoch199--------\n",
       "train accuracy:1.0\n",
-      "train loss:0.00245\n",
-      "test accuracy:0.448\n",
+      "train loss:0.00594\n",
+      "test accuracy:0.466\n",
       "--------------------\n",
       "max accuracy_test: 0.4827586206896552\n"
      ]
@@ -852,7 +856,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 2 Axes>"
       ]
@@ -880,7 +884,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# LightGBM"
+    "## LightGBM"
    ]
   },
   {
@@ -895,7 +899,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -910,13 +914,13 @@
     "    'num_leaves': 15,\n",
     "    'min_data_in_leaf': 10,\n",
     "    'num_iteration': 200,\n",
-    "    'verbose': -1\n",
+    "    'verbose': -1,\n",
     "}"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 28,
    "metadata": {
     "scrolled": false
    },
@@ -1151,7 +1155,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 29,
    "metadata": {
     "scrolled": true
    },
@@ -1162,7 +1166,7 @@
        "0.5"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1173,7 +1177,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 30,
    "metadata": {
     "scrolled": false
    },
@@ -1200,6 +1204,572 @@
     "plt.tight_layout()"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 画像を直接入れる系\n",
+    "こっちのデータは**上のdafatrameと同じ順番**で入ってくると仮定してTargetは同じものを使う  \n",
+    "違う順で入れる場合は注意  "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 171,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "DB_IMG_PATH = \"../input/Dlib/cutface/default/DB/jpeg/\"\n",
+    "QUERY_IMG_PATH = \"../input/Dlib/cutface/default/Query/jpeg/\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 172,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|███████████████████████████████████████████████████████████████████████████████| 200/200 [00:00<00:00, 345.15it/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "# 画像読み込み\n",
+    "p = Path(DB_IMG_PATH)\n",
+    "p = sorted(p.glob(\"*.jpg\"))\n",
+    "\n",
+    "dbImages = []\n",
+    "dbLabels = np.zeros(len(p), dtype=np.int)\n",
+    "\n",
+    "for index, filename in enumerate(tqdm(p)):\n",
+    "    # 相対パスだと参照できなかったので絶対パスでやる\n",
+    "    img = cv2.imread(str(filename.resolve()), 0)\n",
+    "    # C, H, Wの形式にする(今回はグレースケールなのでC = 1)\n",
+    "    img = img.reshape([1, img.shape[0], img.shape[1]])\n",
+    "    dbImages.append((img/225).astype(np.float32))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 173,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|█████████████████████████████████████████████████████████████████████████████████| 58/58 [00:00<00:00, 336.14it/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "# 画像読み込み\n",
+    "p = Path(QUERY_IMG_PATH)\n",
+    "p = sorted(p.glob(\"*.jpg\"))\n",
+    "\n",
+    "queryImages = []\n",
+    "\n",
+    "for index, filename in enumerate(tqdm(p)):\n",
+    "    # 相対パスだと参照できなかったので絶対パスでやる\n",
+    "    img = cv2.imread(str(filename.resolve()), 0)\n",
+    "    # C, H, Wの形式にする(今回はグレースケールなのでC = 1)\n",
+    "    img = img.reshape([1, img.shape[0], img.shape[1]])\n",
+    "    queryImages.append((img/225).astype(np.float32))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 174,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(1, 128, 128)"
+      ]
+     },
+     "execution_count": 174,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dbImages[0].shape"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## ピクセルマッチング"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 175,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "result = []\n",
+    "prediction_df = pd.DataFrame(query_target_df)\n",
+    "for queryIndex in range(len(queryImages)):\n",
+    "    distances = np.zeros(len(dbImages), dtype=np.float32)\n",
+    "    for dbIndex in range(len(dbImages)):\n",
+    "        distances[dbIndex] = (np.abs(dbImages[dbIndex] - queryImages[queryIndex])).sum()\n",
+    "    minimum_id = np.argmin(distances)\n",
+    "    prediction_df.loc[queryIndex, \"predict\"] = db_target_df[minimum_id]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 176,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "correct_num = (prediction_df[\"target\"] == prediction_df[\"predict\"]).sum()\n",
+    "accuracy = correct_num / prediction_df.shape[0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 177,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.4827586206896552"
+      ]
+     },
+     "execution_count": 177,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "accuracy"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# CNN"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 178,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "X_dbToech = torch.Tensor(dbImages)\n",
+    "y_dbTorch = torch.LongTensor(db_target_df)\n",
+    "X_queryTorch = torch.Tensor(queryImages)\n",
+    "y_queryTorch = torch.LongTensor(query_target_df)\n",
+    "\n",
+    "dbDataset = TensorDataset(X_dbToech, y_dbTorch)\n",
+    "queryDataset = TensorDataset(X_queryTorch, y_queryTorch)\n",
+    "\n",
+    "batch_size = 8\n",
+    "dbLoader = DataLoader(dbDataset, batch_size=batch_size, shuffle=True)\n",
+    "queryLoader = DataLoader(queryDataset, batch_size=batch_size, shuffle=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 182,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class FlattenLayer(nn.Module):\n",
+    "    def forward(self, x):\n",
+    "        sizes = x.size()\n",
+    "        return x.view(sizes[0], -1)\n",
+    "    \n",
+    "conv_net = nn.Sequential(\n",
+    "    nn.Conv2d(1, 32, 5),\n",
+    "    nn.MaxPool2d(2),\n",
+    "    nn.ReLU(),\n",
+    "    nn.BatchNorm2d(32),\n",
+    "    nn.Dropout2d(0.5),\n",
+    "    nn.Conv2d(32, 64, 5),\n",
+    "    nn.MaxPool2d(2),\n",
+    "    nn.ReLU(),\n",
+    "    nn.BatchNorm2d(64),\n",
+    "    nn.Dropout2d(0.5),\n",
+    "    FlattenLayer()\n",
+    ")\n",
+    "\n",
+    "test_input = torch.ones(1, 1, 128, 128)\n",
+    "conv_output_size = conv_net(test_input).size()[-1]\n",
+    "\n",
+    "mlp = nn.Sequential(\n",
+    "    nn.Linear(conv_output_size, 200),\n",
+    "    nn.ReLU(),\n",
+    "    nn.BatchNorm1d(200),\n",
+    "    nn.Dropout(0.25),\n",
+    "    nn.Linear(200, 20)\n",
+    ")\n",
+    "\n",
+    "net = nn.Sequential(\n",
+    "    conv_net,\n",
+    "    mlp\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 183,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# 評価ヘルパー\n",
+    "def eval_net(net, data_loader, device=\"cpu\"):\n",
+    "    net.eval()\n",
+    "    ys = []\n",
+    "    ypreds = []\n",
+    "    for x, y in data_loader:\n",
+    "        x = x.to(device)\n",
+    "        y = y.to(device)\n",
+    "        with torch.no_grad():\n",
+    "            _, y_pred = net(x).max(1)\n",
+    "        ys.append(y)\n",
+    "        ypreds.append(y_pred)\n",
+    "        # ミニバッチ毎の結果をまとめる\n",
+    "        ys = torch.cat(ys)\n",
+    "        ypreds = torch.cat(ypreds)\n",
+    "        acc = (ys == ypreds).float().sum() / len(ys)\n",
+    "        return acc.item()\n",
+    "\n",
+    "# 訓練ヘルパー    \n",
+    "def train_net(net, train_loader, test_loader,\n",
+    "              optimizer_cls=optim.Adam, loss_fn=nn.CrossEntropyLoss(),\n",
+    "              n_iter=10, device=\"cpu\"):\n",
+    "    train_losses = []\n",
+    "    train_acc = []\n",
+    "    val_acc = []\n",
+    "    optimizer = optimizer_cls(net.parameters())\n",
+    "    for epoch in range(n_iter):\n",
+    "        running_loss = 0.0\n",
+    "        n = 0\n",
+    "        n_acc = 0\n",
+    "        for i, (xx, yy) in enumerate(tqdm(train_loader)):\n",
+    "            xx = xx.to(device)\n",
+    "            yy = yy.to(device)\n",
+    "            h = net(xx)\n",
+    "            loss = loss_fn(h, yy)\n",
+    "            optimizer.zero_grad()\n",
+    "            loss.backward()\n",
+    "            optimizer.step()\n",
+    "            running_loss += loss.item()\n",
+    "            n += len(xx)\n",
+    "            _, y_pred = h.max(1)\n",
+    "            n_acc += (yy == y_pred).float().sum().item()\n",
+    "        train_losses.append(running_loss / i)\n",
+    "        train_acc.append(n_acc / n)\n",
+    "        val_acc.append(eval_net(net, test_loader, device))\n",
+    "        print(epoch, train_losses[-1], train_acc[-1], val_acc[-1], flush=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 184,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 48.02it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0 2.515797878305117 0.365 0.0\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.26it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1 2.1919733633597693 0.435 0.5\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 52.01it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2 0.9277823691566786 0.78 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 52.88it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "3 0.20635435481866202 0.975 0.125\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.79it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "4 0.07261284689108531 0.98 0.125\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.79it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "5 0.03575696796178818 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 52.22it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "6 0.009554455677668253 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.68it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "7 0.0050013189514478045 1.0 0.125\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 52.01it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "8 0.002986093362172445 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.68it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "9 0.002465811868508657 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.68it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "10 0.001949628194173177 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 52.44it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "11 0.001675511399904887 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.57it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "12 0.0014365812142690022 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.68it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "13 0.0012667824824651082 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.58it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "14 0.0011156325538953145 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 52.33it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "15 0.000992824633916219 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.58it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "16 0.0008824889858563741 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.90it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "17 0.0007980689406394958 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 51.37it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "18 0.0007330526908238729 1.0 0.25\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████████████████████████████████████████████████████████████████████████████| 25/25 [00:00<00:00, 52.22it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "19 0.0006666630506515503 1.0 0.25\n"
+     ]
+    }
+   ],
+   "source": [
+    "device_name = \"cuda:0\"\n",
+    "\n",
+    "net.to(device_name)\n",
+    "train_net(net, dbLoader, queryLoader, n_iter=20, device=device_name)"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,