DenkowskiStanislawlab3.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn.functional as F\n",
    "from torchvision import datasets, transforms\n",
    "import pytorch_lightning as pl\n",
    "import matplotlib.pylab as plt\n",
    "from torchinfo import summary\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext tensorboard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CIFARDataModule(pl.LightningDataModule):\n",
    "    def __init__(self, classes, transforms=transforms.ToTensor(), data_dir='./', batch_size=64, num_workers=0, pin_memory=True):\n",
    "        super().__init__()\n",
    "        self.pin_memory=pin_memory\n",
    "        self.data_dir = data_dir\n",
    "        self.batch_size = batch_size\n",
    "        self.transforms = transforms\n",
    "        self.classes = classes\n",
    "        self.num_workers = num_workers\n",
    "    \n",
    "    def prepare_data(self):\n",
    "        datasets.CIFAR10(root=self.data_dir, train=True, download=True)\n",
    "        datasets.CIFAR10(root=self.data_dir, train=False, download=True)\n",
    "\n",
    "    def setup(self, stage = None):\n",
    "        if stage in (None, 'fit'):\n",
    "            self.train_dset = datasets.CIFAR10(root=self.data_dir, train=True, transform=self.transforms)\n",
    "            self.val_dset = datasets.CIFAR10(root=self.data_dir, train=False, transform=self.transforms)\n",
    "\n",
    "    def train_dataloader(self):\n",
    "        return torch.utils.data.DataLoader(self.train_dset, batch_size=self.batch_size, num_workers=self.num_workers, pin_memory=self.pin_memory)\n",
    "\n",
    "    def val_dataloader(self):\n",
    "        return torch.utils.data.DataLoader(self.val_dset, batch_size=self.batch_size, num_workers=self.num_workers, pin_memory=self.pin_memory)\n",
    "    \n",
    "    def test_dataloader(self):\n",
    "        return torch.utils.data.DataLoader(torch.utils.data.Subset(self.train_dset, range(100)), batch_size=self.batch_size, num_workers=self.num_workers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n",
      "['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']\n"
     ]
    }
   ],
   "source": [
    "classes = ['airplane','automobile','bird','cat','deer','dog','frog','horse','ship','truck']\n",
    "cifardm = CIFARDataModule(classes)\n",
    "cifardm.prepare_data()\n",
    "cifardm.setup()\n",
    "test_dset = list(cifardm.test_dataloader())\n",
    "print(cifardm.classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor(1.)\n",
      "torch.Size([3, 32, 32])\n",
      "tensor(0.9961)\n",
      "torch.Size([3, 32, 32])\n",
      "tensor(1.)\n",
      "torch.Size([3, 32, 32])\n",
      "tensor(0.9176)\n",
      "torch.Size([3, 32, 32])\n",
      "tensor(0.9961)\n",
      "torch.Size([3, 32, 32])\n",
      "tensor(0.9882)\n",
      "torch.Size([3, 32, 32])\n",
      "tensor(0.9765)\n",
      "torch.Size([3, 32, 32])\n",
      "tensor(0.9725)\n",
      "torch.Size([3, 32, 32])\n",
      "tensor(0.9882)\n",
      "torch.Size([3, 32, 32])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 300x300 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(3,3))\n",
    "for ix, photo in enumerate(test_dset[0][0][:9]):\n",
    "    plt.subplot(3, 3, ix + 1)\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "    plt.grid(False)\n",
    "    print(photo.max())\n",
    "    print(photo.shape)\n",
    "    plt.imshow(photo.permute(1,2,0), cmap=plt.cm.binary)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "W PyTorchu CrossEntropyLoss, który działa właśnie jako \"categorical crossentropy\", wymagany jest numery klasy jako wartość oczekiwana, w związku z czym nie będę robił one hot encoding.  \n",
    "W pytorchu możnaby to zrobić przy pomocy torch.nn.functional.one_hot()  \n",
    "Odnośnie zdjęć, widzimy że są bardzo różnorodne, mamy 10 klas, bardzo się różniących (przynajmniej na poziomie abstrakcji).  \n",
    "Dodatkowo nasz dataloader transformuje zdjęcia tak by były znormalizowane i zawierały się w zakresie [0,1] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "class minCIFAR(pl.LightningModule):\n",
    "    def __init__(self, lr=0.01, momentum=0.9):\n",
    "        super(minCIFAR, self).__init__()\n",
    "        self.lr = lr\n",
    "        self.momentum = momentum\n",
    "        self.layer_1 = torch.nn.Sequential(\n",
    "            torch.nn.Conv2d(3,5,kernel_size=3,padding='same'),\n",
    "            torch.nn.Sigmoid()\n",
    "        )\n",
    "        self.layer_2 = torch.nn.Sequential(\n",
    "            torch.nn.Conv2d(5,5,kernel_size=3,padding='same'),\n",
    "            torch.nn.Sigmoid(),\n",
    "            torch.nn.MaxPool2d(kernel_size=8)\n",
    "        )\n",
    "        self.layer_out = torch.nn.Sequential(\n",
    "            torch.nn.Flatten(),\n",
    "            torch.nn.Linear(5*4*4,10),\n",
    "            torch.nn.Softmax(dim=1)\n",
    "        )\n",
    "        self.train_loss = []\n",
    "        self.train_accuracy = []\n",
    "        self.val_loss = []\n",
    "        self.val_accuracy = []\n",
    "    \n",
    "    def forward(self, x):\n",
    "        x = self.layer_1(x)\n",
    "        x = self.layer_2(x)\n",
    "        x = self.layer_out(x)\n",
    "        return x\n",
    "    \n",
    "    def configure_optimizers(self):\n",
    "        return torch.optim.SGD(self.parameters(), lr=self.lr, momentum=self.momentum)\n",
    "\n",
    "    def categorical_crossentropy(self, res, exp):\n",
    "        return F.cross_entropy(res, exp)\n",
    "    \n",
    "    def training_step(self, train_batch, batch_idx):\n",
    "        x,y = train_batch\n",
    "        results = self.forward(x)\n",
    "        loss = self.categorical_crossentropy(results, y)\n",
    "        # self.log('train_los', loss)\n",
    "        # self.log('train_acc', sum(y==results.argmax(dim=1))/len(y))\n",
    "        self.train_loss += [loss]\n",
    "        self.train_accuracy += [sum(y==results.argmax(dim=1))/len(y)]\n",
    "        return loss\n",
    "\n",
    "    def validation_step(self, val_batch, batch_idx):\n",
    "        x,y = val_batch\n",
    "        results = self.forward(x)\n",
    "        loss = self.categorical_crossentropy(results, y)\n",
    "        # self.log('val_loss', loss)\n",
    "        # self.log('val_acc', sum(y==results.argmax(dim=1))/len(y))\n",
    "        self.val_loss += [loss]\n",
    "        self.val_accuracy += [sum(y==results.argmax(dim=1))/len(y)]\n",
    "        return loss\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Zaimplementowałem tę architekturę jako classę pytorch lightning, korzystająć z torch'owych warstw: Sequential, Conv2d, Linear, MaxPool2d, oraz funkcji aktywacji Sigmoid, Softmax.  \n",
    "Dodatkowo nadpisałem metodę forward, która opisuje jak warstwy powinny się zachować.  \n",
    "Dodatkowo zaimplementowałem metody configure optimizers żeby wygodnie zdefinicować optymizer, categorical_crossentropy żeby mieć funkcję loss, training_step i val_step, żeby zdefinniować kolejne kroki uczenia i walidacji.   \n",
    "Na taki model składa się 1180 parametrów, co potwierdza poniższe summary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "=================================================================\n",
       "Layer (type:depth-idx)                   Param #\n",
       "=================================================================\n",
       "minCIFAR                                 --\n",
       "├─Sequential: 1-1                        --\n",
       "│    └─Conv2d: 2-1                       140\n",
       "│    └─Sigmoid: 2-2                      --\n",
       "├─Sequential: 1-2                        --\n",
       "│    └─Conv2d: 2-3                       230\n",
       "│    └─Sigmoid: 2-4                      --\n",
       "│    └─MaxPool2d: 2-5                    --\n",
       "├─Sequential: 1-3                        --\n",
       "│    └─Flatten: 2-6                      --\n",
       "│    └─Linear: 2-7                       810\n",
       "│    └─Softmax: 2-8                      --\n",
       "=================================================================\n",
       "Total params: 1,180\n",
       "Trainable params: 1,180\n",
       "Non-trainable params: 0\n",
       "================================================================="
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m=minCIFAR()\n",
    "summary(m)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([[0.1706, 0.1426, 0.0938, 0.1015, 0.0604, 0.0706, 0.0933, 0.1020, 0.0521,\n",
      "         0.1132]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1708, 0.1426, 0.0940, 0.1013, 0.0604, 0.0702, 0.0931, 0.1020, 0.0521,\n",
      "         0.1136]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1706, 0.1428, 0.0938, 0.1012, 0.0602, 0.0702, 0.0935, 0.1022, 0.0519,\n",
      "         0.1135]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1705, 0.1426, 0.0937, 0.1013, 0.0605, 0.0705, 0.0933, 0.1021, 0.0522,\n",
      "         0.1132]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1704, 0.1425, 0.0938, 0.1015, 0.0604, 0.0701, 0.0936, 0.1023, 0.0520,\n",
      "         0.1134]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1706, 0.1425, 0.0939, 0.1015, 0.0602, 0.0704, 0.0934, 0.1019, 0.0521,\n",
      "         0.1134]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1707, 0.1428, 0.0938, 0.1015, 0.0602, 0.0702, 0.0935, 0.1021, 0.0520,\n",
      "         0.1132]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1711, 0.1421, 0.0935, 0.1015, 0.0600, 0.0703, 0.0935, 0.1024, 0.0520,\n",
      "         0.1136]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1707, 0.1425, 0.0938, 0.1014, 0.0603, 0.0702, 0.0935, 0.1022, 0.0522,\n",
      "         0.1131]], grad_fn=<SoftmaxBackward0>)\n",
      "tensor([[0.1708, 0.1430, 0.0934, 0.1012, 0.0602, 0.0703, 0.0938, 0.1021, 0.0519,\n",
      "         0.1133]], grad_fn=<SoftmaxBackward0>)\n"
     ]
    }
   ],
   "source": [
    "for photo in test_dset[0][0][:10]:\n",
    "    print(m(torch.unsqueeze(photo, dim=0)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Jak widać nasz model zwrócił nam bardzo słabe wyniki - zasadniczo każda klasa ma prawie takie samo prawdopodobieństwo równe mniej więcej 10% +- 3 punkty procentowe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dataloader = cifardm.train_dataloader()\n",
    "val_dataloader = cifardm.val_dataloader()\n",
    "test_dataloader = cifardm.test_dataloader()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (mps), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "IPU available: False, using: 0 IPUs\n",
      "HPU available: False, using: 0 HPUs\n",
      "Missing logger folder: /Users/denkov/Uni/Agh/classes/MRO/lightning_logs\n",
      "\n",
      "  | Name      | Type       | Params\n",
      "-----------------------------------------\n",
      "0 | layer_1   | Sequential | 140   \n",
      "1 | layer_2   | Sequential | 230   \n",
      "2 | layer_out | Sequential | 810   \n",
      "-----------------------------------------\n",
      "1.2 K     Trainable params\n",
      "0         Non-trainable params\n",
      "1.2 K     Total params\n",
      "0.005     Total estimated model params size (MB)\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fe10aad89d5d46ba857daa2d7e94b25b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Sanity Checking: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/denkov/anaconda3/envs/MRO/lib/python3.9/site-packages/pytorch_lightning/trainer/connectors/data_connector.py:236: PossibleUserWarning: The dataloader, val_dataloader 0, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 10 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
      "  rank_zero_warn(\n",
      "/Users/denkov/anaconda3/envs/MRO/lib/python3.9/site-packages/pytorch_lightning/trainer/connectors/data_connector.py:236: PossibleUserWarning: The dataloader, train_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 10 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
      "  rank_zero_warn(\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d2c595c93a7c4b77aac3a0fc02d609b6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2fd6d98dfe54447185a8b280dc456ab2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b6491d9e1c39465598d5f34d2ccb1859",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "34e5034c90ef49d182134ca4d0aa2437",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bd487bf830324a5eb5b849124eb465c7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "57d6754c9b6244889d46668df2a37e76",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "57d5b0b46304437f974975590b1c6455",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bccbbe8c5ff74c1fae86daf53407e454",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c02c007020104fb3b152cf45f29d58b8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da8155842d544d838e9a41fc33ff828c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "983cc2e0d0c04e6cad2b65c960ddab87",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ac82d45968d24453a3641ccb7ea8fc77",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7a19309544e742eda60da6c52a7b842e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "034f21cdd0d74bf1869c6ec798993b9a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "926a850a2fae4009b942a1750213e011",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0f47bcf7bea249839f6c05b2e6d3a316",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "30da435a4dda4e3d879d85f85ba5a7a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bc3bb10b033d483bb946d4f27ef7a6b9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "95d0d7a205854dd78aa1ce094516e184",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "87ee76e352c84158a94cdd669c3de8e2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b49d2f59c75740b4b43b4f9d928bb1d7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8b3580b150bf4a7593f3739ed61da5f6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1e6fdfa3832746929f44e31d38481942",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e34deb007f274f34a7177eb0814283fe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de2ae197a3d3497ab752a0aa3b9e4d68",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dba6250c69ee4b5b9d0dbb1105b9c8c6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "90cf8995e7894a4eba3260ab3676e213",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5724354909f24043bea61e73bbf566d3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fd26127348124146a97641c4fb479ba8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f2e481248d004b67b84403b6bd1dc9b2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5233b9f2a33b40aea1589b4a4e84d0f7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "47c3a829a517485d865370348df615f2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dbee58753f204e0ca187f3e07af5cf95",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "36c52ca50bbd46faac0d3fa99535e4e8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c29a24bd240845e98dcd446b60f81a21",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ab69592d467c484db4d11b5088311311",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "711ea26cb2e84a83a171fc7609ae5eab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d8a461690a849d5b76fe5b586acf368",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b5ed78f4b5e2449397f49fe4ed8d7527",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "30be2723fece4afd8c6cbcaa9e52b4cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a502be1435a64965a073436d4a3c881d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3369b582f9644e599eead02beb43b9dc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3735cc747442449daf14c7ab239ec3e0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4c0ff563bf374709b86908df0e105f6f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2168a7705913437d82983f42fa57365d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "16c356f3c91c459ca665d3d61040eed0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bea85b295db34bd0bb0bdd9466d755dd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "74cbe276d1bd486f8f2d07048622ecab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ae7fcc6653e445659aee67b746a6b5ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6abf47d4c29b41b584928b80adaaed74",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0c52513f83f247bbb99a34584f51fe35",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "31698442803748ccbf17e7953f558a6b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "48a946ec916644e19fbf656a6dd6685b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b4bc429125ce41b0acff612a828a6b61",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f2021ad4f14745c7a6fad69402b7d2e1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "659098183fc843d19cfd9dc88b631f89",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f3dc7dedb4fc480f958dc730bd5f849d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "99bceb0236624bde8ed8ffb1e1401059",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1966c72ccde64d819e6b4d3a191ff518",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8e7c5aeb0e7c4f12b10325623a2a24a9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f09b99fe33ca4266b3838cc1707de944",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9e0871fa1b6440b8d5defe6f3dc6e3e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ae3721f247104ad2a54ea9555a238e77",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "75ea0cc00b6d43d3b6acf34be9886bf2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7c1c25b30dc042d9815e9394767d2638",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "63ab5a8ede3c422fb93b7cd64aed6c56",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4decdc8827c149379ba7df44161a7200",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f31f5cca500e4d1aae3fc557db72dfee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "90cfe3050de340d6a3ebdb280a87772b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "44868bf6cdc543e2b6e77ac2f6c599b6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8d596c2d53d149b993d830453341b2a4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f74883a583714fc0b320f25a8ca72f10",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "550a7502187a4d459001567b7ce08a80",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "af800799c5a749bfbf9d778109fb6120",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "346ef289d588437eb6a7b9a199527e50",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "298271458fe3457b863edcf135072c87",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "726efeff4e304560a1ed0857a6d097ab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "05e9cacecdcc4ecf993f7d00d093d660",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7ec3426e22424a04a4ef35a894a35903",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "624eb28687a648cba0a618a1df74bbf0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "87cff7bc2c244a2e926fa1b3a6f8e429",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de91a5ebf5a44a67846a0831d2b1cf0a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b8122bb940c64b3ba2ed2c9420a0e39e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f6e64ff7a450493297c2237d358235eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cd3845e78e3f42fb8d1ab2ad589fab3f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "18fd9d48965947abbc983ca4ad262884",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "94884a72d51b4085a8431a04dd712658",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9d771dc1b89e44cc83dbb6572e756dc3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da8cc7c13a7246fa8327291f3d1e8433",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "99614b09bf134a8993e2f4fd48ebec6f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9617a26209944f63a42207bf07fe9c8c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "386c7cb5b5f94784a91f968628cbae73",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "227b88c78fec411b8024f07976fbd3f7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "426c1578b0c0450bb669a75588ece2b0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f18590be627546158531573d8f445998",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d4f63889ac3b4cd0a86ebfc16957f3ab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e4cfa2e3ffd342beaadc734c27c88780",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "205c1a81dfa9403ea47d09a09d5e35c8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d241cb471f2c4d31af075fe4c475ee4d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "49d8a84d523c40b1b31fcbbbfd8edeaf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8933059935274489a195fb19ae9ce52e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "16ccfce806344cfab09eb5478cfdbcd4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1fceaedb9fca432a91c2554c601df6b9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "762bcf22ee2042c1a2769f38fb0e3fec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b08de9d334814e758ddfc7f00eecab1b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d12955c0f33c4335a35683e98d1b2fc0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "300923c05cb748128bf6ee7ebc8705cc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c13acc714d1641b59baca7092d26a4fb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "715559ba3b2e41fb8d467c603eca4317",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b9b4acdd88dd481a8893680bb40c4f81",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "619bcb0bb6024a16a1b3bccf38b662d9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6220557cb79c4115b8f8e78641567d68",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "276064fc6ffe41479d707baf830b4745",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dc919ffbac284395ab51aba572f6594e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "369f2b20d63e40d88cfe8a5c515cf826",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5c14d283140d42eea4d1a9174c38ba53",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7838d8a8583f4a10b218277ef8827890",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bfe940b36b0f47499403a595e74b94cc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c1ee3b374885428fba173c2102ce2b63",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "40dcb7a8b8ab44c29df2c73be7c94a2f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e612761c7e0406aae032ef703d7bcf6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a0fd332f5ad5442680d7c5bdefddeae3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "05121f7851294bbfa5aaf4a5cc88bf52",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b5e5cdf295f0494ca7495aaddd284a89",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2f2fa0aaf44c4932bcfcdfc5736f78b5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f370da8f21bf4a53a14b0daef3677b6d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "28f242fb8db84a0e9f2abd4bad64429b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1761225f8405479b9db489dd6c94f400",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9fa3325641847f4809f5c9c65abe56a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b39810f8f1af4832a9e8cf3bc98f7c98",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "84e635fb5f764a569c8824c065bdcda2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d12032cb3239428ab637c28ecdfe2cae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a732ab838fc41efa5ee74af614b75ee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de6560a33a214bc7b0393e40ffcb04f8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da18474f73ae4896b500104e65b2684d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8a5f00c83de043f887131fe9fea0dd39",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f3a82f911da84d7dbbb3dc6041d77157",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0fd3dc8147a04c4cb46eb592d62bd6a3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "93d0750370264079b59cecea439b94a8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "01382aca63904c4eb16f09c9cd4f723d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "17d0b83667b24223b29d9dfa538c6ce3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ec8821f90e04c6ab6f86e2619b0fed6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ba2a156d518a44b6a972850208c92e26",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "42ad77db7ebc479ca4a96b7ce5210da1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f7e30ec0e8664f8b9cadfb25141fcb59",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5a6ea1b5ec684387ab82796c9f8aff1e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e2b067b991542c7b0227a1c23524fc3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3f674a1a0ee7423e85280ffef5b19abc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9f98a6b1ed44875a809015bee33cb1b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cce4943a2e114229a6c6356a252e7e77",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "81e060d990d647c493aad38dcc0f6dc7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`Trainer.fit` stopped: `max_epochs=150` reached.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 1732.9184501171112\n",
      "\n"
     ]
    }
   ],
   "source": [
    "start_time = time.time()\n",
    "model = minCIFAR()\n",
    "trainer = pl.Trainer(max_epochs=150, accelerator='mps')\n",
    "trainer.fit(model, train_dataloader, val_dataloader)\n",
    "print(f\"time: {time.time()-start_time}\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Niestety nie wiem jak zrobić ładniejszy output.  \n",
    "Pojawia się informacja, że mógłbym korzystać z num_workers przy dataloaderach, ale jesli dam ich liczbę większą niż 0, to ładowanie danych trwa strasznie długo.  \n",
    "Czas trwania - mniej więcej 30 min"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_to_pltable(x):\n",
    "    results = torch.tensor(x)\n",
    "    res = []\n",
    "    for val in results.tensor_split(150):\n",
    "        res += [sum(val)/len(val)]\n",
    "    return res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/8p/_cz7tfws6qj3stcp8jwc282r0000gn/T/ipykernel_25655/1773930568.py:2: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
      "  results = torch.tensor(x)\n"
     ]
    }
   ],
   "source": [
    "train_loss = convert_to_pltable(torch.tensor(model.train_loss))\n",
    "train_accuracy = convert_to_pltable(torch.tensor(model.train_accuracy))\n",
    "val_loss = convert_to_pltable(torch.tensor(model.val_loss))\n",
    "val_accuracy = convert_to_pltable(torch.tensor(model.val_accuracy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def draw_accuracy_loss(train, val, title):\n",
    "    epochs = range(150)\n",
    "    plt.plot(epochs, train, 'r', label='training')\n",
    "    plt.plot(epochs, val, 'g', label='validation')\n",
    "    plt.title(title)\n",
    "    plt.xlabel('Epochs')\n",
    "    plt.ylabel('Value')\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "\n",
    "draw_accuracy_loss(train_loss, val_loss, 'Training and Validation loss')\n",
    "draw_accuracy_loss(train_accuracy, val_accuracy, 'Training and Validation accuracy')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.13 ('MRO')",
   "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": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "a8ef4b5be830792ced5920383c70035612d6659999c109cdd3b68c6789475da4"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}