Image Classifier Project.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Developing an AI application\n",
    "\n",
    "Going forward, AI algorithms will be incorporated into more and more everyday applications. For example, you might want to include an image classifier in a smart phone app. To do this, you'd use a deep learning model trained on hundreds of thousands of images as part of the overall application architecture. A large part of software development in the future will be using these types of models as common parts of applications. \n",
    "\n",
    "In this project, you'll train an image classifier to recognize different species of flowers. You can imagine using something like this in a phone app that tells you the name of the flower your camera is looking at. In practice you'd train this classifier, then export it for use in your application. We'll be using [this dataset](http://www.robots.ox.ac.uk/~vgg/data/flowers/102/index.html) of 102 flower categories, you can see a few examples below. \n",
    "\n",
    "<img src='assets/Flowers.png' width=500px>\n",
    "\n",
    "The project is broken down into multiple steps:\n",
    "\n",
    "* Load and preprocess the image dataset\n",
    "* Train the image classifier on your dataset\n",
    "* Use the trained classifier to predict image content\n",
    "\n",
    "We'll lead you through each part which you'll implement in Python.\n",
    "\n",
    "When you've completed this project, you'll have an application that can be trained on any set of labeled images. Here your network will be learning about flowers and end up as a command line application. But, what you do with your new skills depends on your imagination and effort in building a dataset. For example, imagine an app where you take a picture of a car, it tells you what the make and model is, then looks up information about it. Go build your own dataset and make something new.\n",
    "\n",
    "First up is importing the packages you'll need. It's good practice to keep all the imports at the beginning of your code. As you work through this notebook and find you need to import a package, make sure to add the import up here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Imports here\n",
    "%matplotlib inline \n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import torch\n",
    "import numpy as np\n",
    "from torch import nn\n",
    "from torch import optim\n",
    "from torchvision import datasets, models, transforms\n",
    "import torch.nn.functional as F\n",
    "import torch.utils.data \n",
    "import pandas as pd\n",
    "#import helper\n",
    "from collections import OrderedDict\n",
    "from PIL import Image\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the data\n",
    "\n",
    "Here you'll use `torchvision` to load the data ([documentation](http://pytorch.org/docs/0.3.0/torchvision/index.html)). The data should be included alongside this notebook, otherwise you can [download it here](https://s3.amazonaws.com/content.udacity-data.com/nd089/flower_data.tar.gz). The dataset is split into three parts, training, validation, and testing. For the training, you'll want to apply transformations such as random scaling, cropping, and flipping. This will help the network generalize leading to better performance. You'll also need to make sure the input data is resized to 224x224 pixels as required by the pre-trained networks.\n",
    "\n",
    "The validation and testing sets are used to measure the model's performance on data it hasn't seen yet. For this you don't want any scaling or rotation transformations, but you'll need to resize then crop the images to the appropriate size.\n",
    "\n",
    "The pre-trained networks you'll use were trained on the ImageNet dataset where each color channel was normalized separately. For all three sets you'll need to normalize the means and standard deviations of the images to what the network expects. For the means, it's `[0.485, 0.456, 0.406]` and for the standard deviations `[0.229, 0.224, 0.225]`, calculated from the ImageNet images.  These values will shift each color channel to be centered at 0 and range from -1 to 1.\n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dir = 'flowers'\n",
    "train_dir = data_dir + '/train'\n",
    "valid_dir = data_dir + '/valid'\n",
    "test_dir = data_dir + '/test'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: Define your transforms for the training, validation, and testing sets\n",
    "train_data_transforms = transforms.Compose ([transforms.RandomRotation (30),\n",
    "                                             transforms.RandomResizedCrop (224),\n",
    "                                             transforms.RandomHorizontalFlip (),\n",
    "                                             transforms.ToTensor (),\n",
    "                                             transforms.Normalize ([0.485, 0.456, 0.406],[0.229, 0.224, 0.225])\n",
    "                                            ])\n",
    "\n",
    "valid_data_transforms = transforms.Compose ([transforms.Resize (255),\n",
    "                                             transforms.CenterCrop (224),\n",
    "                                             transforms.ToTensor (),\n",
    "                                             transforms.Normalize ([0.485, 0.456, 0.406],[0.229, 0.224, 0.225])\n",
    "                                            ])\n",
    "\n",
    "test_data_transforms = transforms.Compose ([transforms.Resize (255),\n",
    "                                             transforms.CenterCrop (224),\n",
    "                                             transforms.ToTensor (),\n",
    "                                             transforms.Normalize ([0.485, 0.456, 0.406],[0.229, 0.224, 0.225])\n",
    "                                            ])\n",
    "\n",
    "# TODO: Load the datasets with ImageFolder\n",
    "train_image_datasets = datasets.ImageFolder (train_dir, transform = train_data_transforms)\n",
    "valid_image_datasets = datasets.ImageFolder (valid_dir, transform = valid_data_transforms)\n",
    "test_image_datasets = datasets.ImageFolder (test_dir, transform = test_data_transforms)\n",
    "\n",
    "\n",
    "# TODO: Using the image datasets and the trainforms, define the dataloaders\n",
    "train_loader = torch.utils.data.DataLoader(train_image_datasets, batch_size = 64, shuffle = True)\n",
    "valid_loader = torch.utils.data.DataLoader(valid_image_datasets, batch_size = 64, shuffle = True)\n",
    "test_loader = torch.utils.data.DataLoader(test_image_datasets, batch_size = 64, shuffle = True)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([64, 3, 224, 224])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputs, labels = next(iter(train_loader))\n",
    "inputs [0,:]\n",
    "inputs.size ()\n",
    "#plt.imshow (inputs) #helper.imshow(image[0,:]);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'1': 0,\n",
       " '10': 1,\n",
       " '100': 2,\n",
       " '101': 3,\n",
       " '102': 4,\n",
       " '11': 5,\n",
       " '12': 6,\n",
       " '13': 7,\n",
       " '14': 8,\n",
       " '15': 9,\n",
       " '16': 10,\n",
       " '17': 11,\n",
       " '18': 12,\n",
       " '19': 13,\n",
       " '2': 14,\n",
       " '20': 15,\n",
       " '21': 16,\n",
       " '22': 17,\n",
       " '23': 18,\n",
       " '24': 19,\n",
       " '25': 20,\n",
       " '26': 21,\n",
       " '27': 22,\n",
       " '28': 23,\n",
       " '29': 24,\n",
       " '3': 25,\n",
       " '30': 26,\n",
       " '31': 27,\n",
       " '32': 28,\n",
       " '33': 29,\n",
       " '34': 30,\n",
       " '35': 31,\n",
       " '36': 32,\n",
       " '37': 33,\n",
       " '38': 34,\n",
       " '39': 35,\n",
       " '4': 36,\n",
       " '40': 37,\n",
       " '41': 38,\n",
       " '42': 39,\n",
       " '43': 40,\n",
       " '44': 41,\n",
       " '45': 42,\n",
       " '46': 43,\n",
       " '47': 44,\n",
       " '48': 45,\n",
       " '49': 46,\n",
       " '5': 47,\n",
       " '50': 48,\n",
       " '51': 49,\n",
       " '52': 50,\n",
       " '53': 51,\n",
       " '54': 52,\n",
       " '55': 53,\n",
       " '56': 54,\n",
       " '57': 55,\n",
       " '58': 56,\n",
       " '59': 57,\n",
       " '6': 58,\n",
       " '60': 59,\n",
       " '61': 60,\n",
       " '62': 61,\n",
       " '63': 62,\n",
       " '64': 63,\n",
       " '65': 64,\n",
       " '66': 65,\n",
       " '67': 66,\n",
       " '68': 67,\n",
       " '69': 68,\n",
       " '7': 69,\n",
       " '70': 70,\n",
       " '71': 71,\n",
       " '72': 72,\n",
       " '73': 73,\n",
       " '74': 74,\n",
       " '75': 75,\n",
       " '76': 76,\n",
       " '77': 77,\n",
       " '78': 78,\n",
       " '79': 79,\n",
       " '8': 80,\n",
       " '80': 81,\n",
       " '81': 82,\n",
       " '82': 83,\n",
       " '83': 84,\n",
       " '84': 85,\n",
       " '85': 86,\n",
       " '86': 87,\n",
       " '87': 88,\n",
       " '88': 89,\n",
       " '89': 90,\n",
       " '9': 91,\n",
       " '90': 92,\n",
       " '91': 93,\n",
       " '92': 94,\n",
       " '93': 95,\n",
       " '94': 96,\n",
       " '95': 97,\n",
       " '96': 98,\n",
       " '97': 99,\n",
       " '98': 100,\n",
       " '99': 101}"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_image_datasets.class_to_idx\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Label mapping\n",
    "\n",
    "You'll also need to load in a mapping from category label to category name. You can find this in the file `cat_to_name.json`. It's a JSON object which you can read in with the [`json` module](https://docs.python.org/2/library/json.html). This will give you a dictionary mapping the integer encoded categories to the actual names of the flowers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'1': 'pink primrose',\n",
       " '10': 'globe thistle',\n",
       " '100': 'blanket flower',\n",
       " '101': 'trumpet creeper',\n",
       " '102': 'blackberry lily',\n",
       " '11': 'snapdragon',\n",
       " '12': \"colt's foot\",\n",
       " '13': 'king protea',\n",
       " '14': 'spear thistle',\n",
       " '15': 'yellow iris',\n",
       " '16': 'globe-flower',\n",
       " '17': 'purple coneflower',\n",
       " '18': 'peruvian lily',\n",
       " '19': 'balloon flower',\n",
       " '2': 'hard-leaved pocket orchid',\n",
       " '20': 'giant white arum lily',\n",
       " '21': 'fire lily',\n",
       " '22': 'pincushion flower',\n",
       " '23': 'fritillary',\n",
       " '24': 'red ginger',\n",
       " '25': 'grape hyacinth',\n",
       " '26': 'corn poppy',\n",
       " '27': 'prince of wales feathers',\n",
       " '28': 'stemless gentian',\n",
       " '29': 'artichoke',\n",
       " '3': 'canterbury bells',\n",
       " '30': 'sweet william',\n",
       " '31': 'carnation',\n",
       " '32': 'garden phlox',\n",
       " '33': 'love in the mist',\n",
       " '34': 'mexican aster',\n",
       " '35': 'alpine sea holly',\n",
       " '36': 'ruby-lipped cattleya',\n",
       " '37': 'cape flower',\n",
       " '38': 'great masterwort',\n",
       " '39': 'siam tulip',\n",
       " '4': 'sweet pea',\n",
       " '40': 'lenten rose',\n",
       " '41': 'barbeton daisy',\n",
       " '42': 'daffodil',\n",
       " '43': 'sword lily',\n",
       " '44': 'poinsettia',\n",
       " '45': 'bolero deep blue',\n",
       " '46': 'wallflower',\n",
       " '47': 'marigold',\n",
       " '48': 'buttercup',\n",
       " '49': 'oxeye daisy',\n",
       " '5': 'english marigold',\n",
       " '50': 'common dandelion',\n",
       " '51': 'petunia',\n",
       " '52': 'wild pansy',\n",
       " '53': 'primula',\n",
       " '54': 'sunflower',\n",
       " '55': 'pelargonium',\n",
       " '56': 'bishop of llandaff',\n",
       " '57': 'gaura',\n",
       " '58': 'geranium',\n",
       " '59': 'orange dahlia',\n",
       " '6': 'tiger lily',\n",
       " '60': 'pink-yellow dahlia',\n",
       " '61': 'cautleya spicata',\n",
       " '62': 'japanese anemone',\n",
       " '63': 'black-eyed susan',\n",
       " '64': 'silverbush',\n",
       " '65': 'californian poppy',\n",
       " '66': 'osteospermum',\n",
       " '67': 'spring crocus',\n",
       " '68': 'bearded iris',\n",
       " '69': 'windflower',\n",
       " '7': 'moon orchid',\n",
       " '70': 'tree poppy',\n",
       " '71': 'gazania',\n",
       " '72': 'azalea',\n",
       " '73': 'water lily',\n",
       " '74': 'rose',\n",
       " '75': 'thorn apple',\n",
       " '76': 'morning glory',\n",
       " '77': 'passion flower',\n",
       " '78': 'lotus lotus',\n",
       " '79': 'toad lily',\n",
       " '8': 'bird of paradise',\n",
       " '80': 'anthurium',\n",
       " '81': 'frangipani',\n",
       " '82': 'clematis',\n",
       " '83': 'hibiscus',\n",
       " '84': 'columbine',\n",
       " '85': 'desert-rose',\n",
       " '86': 'tree mallow',\n",
       " '87': 'magnolia',\n",
       " '88': 'cyclamen',\n",
       " '89': 'watercress',\n",
       " '9': 'monkshood',\n",
       " '90': 'canna lily',\n",
       " '91': 'hippeastrum',\n",
       " '92': 'bee balm',\n",
       " '93': 'ball moss',\n",
       " '94': 'foxglove',\n",
       " '95': 'bougainvillea',\n",
       " '96': 'camellia',\n",
       " '97': 'mallow',\n",
       " '98': 'mexican petunia',\n",
       " '99': 'bromelia'}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import json\n",
    "\n",
    "with open('cat_to_name.json', 'r') as f:\n",
    "    cat_to_name = json.load(f)\n",
    "    \n",
    "len (cat_to_name)\n",
    "cat_to_name"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Building and training the classifier\n",
    "\n",
    "Now that the data is ready, it's time to build and train the classifier. As usual, you should use one of the pretrained models from `torchvision.models` to get the image features. Build and train a new feed-forward classifier using those features.\n",
    "\n",
    "We're going to leave this part up to you. If you want to talk through it with someone, chat with your fellow students! You can also ask questions on the forums or join the instructors in office hours.\n",
    "\n",
    "Refer to [the rubric](https://review.udacity.com/#!/rubrics/1663/view) for guidance on successfully completing this section. Things you'll need to do:\n",
    "\n",
    "* Load a [pre-trained network](http://pytorch.org/docs/master/torchvision/models.html) (If you need a starting point, the VGG networks work great and are straightforward to use)\n",
    "* Define a new, untrained feed-forward network as a classifier, using ReLU activations and dropout\n",
    "* Train the classifier layers using backpropagation using the pre-trained network to get the features\n",
    "* Track the loss and accuracy on the validation set to determine the best hyperparameters\n",
    "\n",
    "We've left a cell open for you below, but use as many as you need. Our advice is to break the problem up into smaller parts you can run separately. Check that each part is doing what you expect, then move on to the next. You'll likely find that as you work through each part, you'll need to go back and modify your previous code. This is totally normal!\n",
    "\n",
    "When training make sure you're updating only the weights of the feed-forward network. You should be able to get the validation accuracy above 70% if you build everything right. Make sure to try different hyperparameters (learning rate, units in the classifier, epochs, etc) to find the best model. Save those hyperparameters to use as default values in the next part of the project."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://download.pytorch.org/models/alexnet-owt-4df8aa71.pth\" to /root/.torch/models/alexnet-owt-4df8aa71.pth\n",
      "100%|██████████| 244418560/244418560 [00:03<00:00, 64600296.19it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "AlexNet(\n",
       "  (features): Sequential(\n",
       "    (0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))\n",
       "    (1): ReLU(inplace)\n",
       "    (2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (3): Conv2d(64, 192, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
       "    (4): ReLU(inplace)\n",
       "    (5): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (6): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (7): ReLU(inplace)\n",
       "    (8): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (9): ReLU(inplace)\n",
       "    (10): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (11): ReLU(inplace)\n",
       "    (12): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "  )\n",
       "  (classifier): Sequential(\n",
       "    (0): Dropout(p=0.5)\n",
       "    (1): Linear(in_features=9216, out_features=4096, bias=True)\n",
       "    (2): ReLU(inplace)\n",
       "    (3): Dropout(p=0.5)\n",
       "    (4): Linear(in_features=4096, out_features=4096, bias=True)\n",
       "    (5): ReLU(inplace)\n",
       "    (6): Linear(in_features=4096, out_features=1000, bias=True)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# TODO: Build and train your network\n",
    "model = models.alexnet (pretrained = True)\n",
    "model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AlexNet(\n",
       "  (features): Sequential(\n",
       "    (0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))\n",
       "    (1): ReLU(inplace)\n",
       "    (2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (3): Conv2d(64, 192, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
       "    (4): ReLU(inplace)\n",
       "    (5): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (6): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (7): ReLU(inplace)\n",
       "    (8): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (9): ReLU(inplace)\n",
       "    (10): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (11): ReLU(inplace)\n",
       "    (12): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "  )\n",
       "  (classifier): Sequential(\n",
       "    (fc1): Linear(in_features=9216, out_features=4096, bias=True)\n",
       "    (relu1): ReLU()\n",
       "    (dropout1): Dropout(p=0.3)\n",
       "    (fc2): Linear(in_features=4096, out_features=2048, bias=True)\n",
       "    (relu2): ReLU()\n",
       "    (dropout2): Dropout(p=0.3)\n",
       "    (fc3): Linear(in_features=2048, out_features=102, bias=True)\n",
       "    (output): LogSoftmax()\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# updating classifer in the network\n",
    "for param in model.parameters(): \n",
    "    param.requires_grad = False\n",
    "\n",
    "classifier = nn.Sequential  (OrderedDict ([\n",
    "                            ('fc1', nn.Linear (9216, 4096)),\n",
    "                            ('relu1', nn.ReLU ()),\n",
    "                            ('dropout1', nn.Dropout (p = 0.3)),\n",
    "                            ('fc2', nn.Linear (4096, 2048)),\n",
    "                            ('relu2', nn.ReLU ()),\n",
    "                            ('dropout2', nn.Dropout (p = 0.3)),\n",
    "                            ('fc3', nn.Linear (2048, 102)),\n",
    "                            ('output', nn.LogSoftmax (dim =1))\n",
    "                            ]))\n",
    "model.classifier = classifier\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#initializing criterion and optimizer\n",
    "criterion = nn.NLLLoss ()\n",
    "optimizer = optim.Adam (model.classifier.parameters (), lr = 0.001)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Defining validation \n",
    "def validation(model, valid_loader, criterion):\n",
    "    model.to ('cuda')\n",
    "    \n",
    "    valid_loss = 0\n",
    "    accuracy = 0\n",
    "    for inputs, labels in valid_loader:\n",
    "        \n",
    "        inputs, labels = inputs.to('cuda'), labels.to('cuda')\n",
    "        output = model.forward(inputs)\n",
    "        valid_loss += criterion(output, labels).item()\n",
    "\n",
    "        ps = torch.exp(output)\n",
    "        equality = (labels.data == ps.max(dim=1)[1])\n",
    "        accuracy += equality.type(torch.FloatTensor).mean()\n",
    "    \n",
    "    return valid_loss, accuracy\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 1/7..  Training Loss: 4.534..  Valid Loss: 2.665..  Valid Accuracy: 34.654%\n",
      "Epoch: 1/7..  Training Loss: 2.499..  Valid Loss: 1.467..  Valid Accuracy: 58.788%\n",
      "Epoch: 2/7..  Training Loss: 0.851..  Valid Loss: 1.260..  Valid Accuracy: 64.726%\n",
      "Epoch: 2/7..  Training Loss: 1.860..  Valid Loss: 1.090..  Valid Accuracy: 68.332%\n",
      "Epoch: 2/7..  Training Loss: 1.745..  Valid Loss: 0.922..  Valid Accuracy: 74.885%\n",
      "Epoch: 3/7..  Training Loss: 1.363..  Valid Loss: 0.932..  Valid Accuracy: 73.308%\n",
      "Epoch: 3/7..  Training Loss: 1.560..  Valid Loss: 0.840..  Valid Accuracy: 76.312%\n",
      "Epoch: 4/7..  Training Loss: 0.398..  Valid Loss: 0.936..  Valid Accuracy: 73.721%\n",
      "Epoch: 4/7..  Training Loss: 1.416..  Valid Loss: 0.701..  Valid Accuracy: 79.611%\n",
      "Epoch: 4/7..  Training Loss: 1.417..  Valid Loss: 0.775..  Valid Accuracy: 79.062%\n",
      "Epoch: 5/7..  Training Loss: 0.997..  Valid Loss: 0.818..  Valid Accuracy: 77.942%\n",
      "Epoch: 5/7..  Training Loss: 1.443..  Valid Loss: 0.683..  Valid Accuracy: 81.856%\n",
      "Epoch: 6/7..  Training Loss: 0.163..  Valid Loss: 0.716..  Valid Accuracy: 80.947%\n",
      "Epoch: 6/7..  Training Loss: 1.278..  Valid Loss: 0.717..  Valid Accuracy: 80.519%\n",
      "Epoch: 6/7..  Training Loss: 1.361..  Valid Loss: 0.761..  Valid Accuracy: 79.144%\n",
      "Epoch: 7/7..  Training Loss: 0.671..  Valid Loss: 0.629..  Valid Accuracy: 82.183%\n",
      "Epoch: 7/7..  Training Loss: 1.218..  Valid Loss: 0.677..  Valid Accuracy: 82.529%\n",
      "Epoch: 7/7..  Training Loss: 1.246..  Valid Loss: 0.627..  Valid Accuracy: 83.659%\n"
     ]
    }
   ],
   "source": [
    "#training a model\n",
    "\n",
    "#change to cuda if enabled\n",
    "model.to ('cuda')\n",
    "epochs = 7\n",
    "print_every = 40\n",
    "steps = 0\n",
    "\n",
    "\n",
    "for e in range (epochs): \n",
    "    running_loss = 0\n",
    "    for ii, (inputs, labels) in enumerate (train_loader):\n",
    "        steps += 1\n",
    "    \n",
    "        inputs, labels = inputs.to('cuda'), labels.to('cuda')\n",
    "    \n",
    "        optimizer.zero_grad () #where optimizer is working on classifier paramters only\n",
    "    \n",
    "        # Forward and backward passes\n",
    "        outputs = model.forward (inputs) #calculating output\n",
    "        loss = criterion (outputs, labels) #calculating loss\n",
    "        loss.backward () \n",
    "        optimizer.step () #performs single optimization step \n",
    "    \n",
    "        running_loss += loss.item () # loss.item () returns scalar value of Loss function\n",
    "    \n",
    "        if steps % print_every == 0:\n",
    "            model.eval () #switching to evaluation mode so that dropout is turned off\n",
    "            \n",
    "            # Turn off gradients for validation, saves memory and computations\n",
    "            with torch.no_grad():\n",
    "                valid_loss, accuracy = validation(model, valid_loader, criterion)\n",
    "            \n",
    "            print(\"Epoch: {}/{}.. \".format(e+1, epochs),\n",
    "                  \"Training Loss: {:.3f}.. \".format(running_loss/print_every),\n",
    "                  \"Valid Loss: {:.3f}.. \".format(valid_loss/len(valid_loader)),\n",
    "                  \"Valid Accuracy: {:.3f}%\".format(accuracy/len(valid_loader)*100))\n",
    "            \n",
    "            running_loss = 0\n",
    "            \n",
    "            # Make sure training is back on\n",
    "            model.train()\n",
    "            \n",
    "            \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Testing your network\n",
    "\n",
    "It's good practice to test your trained network on test data, images the network has never seen either in training or validation. This will give you a good estimate for the model's performance on completely new images. Run the test images through the network and measure the accuracy, the same way you did validation. You should be able to reach around 70% accuracy on the test set if the model has been trained well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of the network on test images: 73 %\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "819"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# TODO: Do validation on the test set\n",
    "\n",
    "ts_correct = 0\n",
    "ts_total = 0\n",
    "\n",
    "with torch.no_grad ():\n",
    "    for data in test_loader:\n",
    "        inputs, labels = data\n",
    "        inputs, labels = inputs.to('cuda'), labels.to('cuda')\n",
    "        outputs = model (inputs)\n",
    "        _, predicted = torch.max (outputs.data,1)\n",
    "        ts_total += labels.size (0)\n",
    "        ts_correct += (predicted == labels).sum().item()\n",
    "\n",
    "print('Accuracy of the network on test images: %d %%' % (100 * ts_correct / ts_total))\n",
    "ts_total\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OrderedDict([('features.0.weight',\n",
       "              tensor([[[[ 0.1186,  0.0941,  0.0954,  ...,  0.0558,  0.0216,  0.0500],\n",
       "                        [ 0.0749,  0.0389,  0.0530,  ...,  0.0257, -0.0113,  0.0042],\n",
       "                        [ 0.0754,  0.0388,  0.0549,  ...,  0.0436,  0.0102,  0.0133],\n",
       "                        ...,\n",
       "                        [ 0.0932,  0.1037,  0.0675,  ..., -0.2028, -0.1284, -0.1122],\n",
       "                        [ 0.0435,  0.0649,  0.0362,  ..., -0.2025, -0.1138, -0.1072],\n",
       "                        [ 0.0474,  0.0625,  0.0248,  ..., -0.1184, -0.0956, -0.0839]],\n",
       "              \n",
       "                       [[-0.0726, -0.0580, -0.0807,  ..., -0.0006, -0.0253,  0.0255],\n",
       "                        [-0.0690, -0.0676, -0.0764,  ..., -0.0040, -0.0304,  0.0105],\n",
       "                        [-0.0995, -0.0856, -0.1052,  ..., -0.0266, -0.0228,  0.0066],\n",
       "                        ...,\n",
       "                        [-0.1512, -0.0887, -0.0967,  ...,  0.3085,  0.1810,  0.0843],\n",
       "                        [-0.1431, -0.0757, -0.0722,  ...,  0.2042,  0.1645,  0.0952],\n",
       "                        [-0.0859, -0.0401, -0.0515,  ...,  0.1635,  0.1482,  0.1020]],\n",
       "              \n",
       "                       [[-0.0236, -0.0021, -0.0278,  ...,  0.0399, -0.0071,  0.0322],\n",
       "                        [ 0.0003,  0.0225,  0.0089,  ...,  0.0188, -0.0142,  0.0183],\n",
       "                        [ 0.0054,  0.0294,  0.0003,  ...,  0.0121, -0.0025,  0.0084],\n",
       "                        ...,\n",
       "                        [-0.0628, -0.0117, -0.0621,  ...,  0.1033, -0.0095, -0.0796],\n",
       "                        [-0.0457,  0.0034, -0.0396,  ..., -0.0264, -0.0335, -0.0764],\n",
       "                        [-0.0187,  0.0114, -0.0397,  ..., -0.0686, -0.0413, -0.0555]]],\n",
       "              \n",
       "              \n",
       "                      [[[-0.0020,  0.0029,  0.0482,  ...,  0.0614,  0.0261,  0.0196],\n",
       "                        [-0.0126, -0.0049,  0.0185,  ...,  0.0539,  0.0164,  0.0238],\n",
       "                        [ 0.0037, -0.0008,  0.0264,  ..., -0.0258, -0.0618,  0.0261],\n",
       "                        ...,\n",
       "                        [-0.0108, -0.0046,  0.0151,  ...,  0.0296,  0.0053,  0.0686],\n",
       "                        [ 0.0003, -0.0149,  0.0078,  ...,  0.0272, -0.0181,  0.0525],\n",
       "                        [-0.0525, -0.0466, -0.0110,  ...,  0.0043, -0.0026,  0.0144]],\n",
       "              \n",
       "                       [[ 0.0240,  0.0227,  0.0058,  ...,  0.0072, -0.0247,  0.0447],\n",
       "                        [ 0.0269,  0.0449, -0.0011,  ...,  0.0442, -0.0212,  0.0645],\n",
       "                        [ 0.0124,  0.0102, -0.0416,  ..., -0.1213, -0.1629,  0.0263],\n",
       "                        ...,\n",
       "                        [ 0.0359,  0.0532,  0.0110,  ...,  0.0127, -0.0297,  0.0859],\n",
       "                        [ 0.0156,  0.0217, -0.0083,  ..., -0.0033, -0.0541,  0.0576],\n",
       "                        [ 0.0753,  0.0878,  0.0558,  ...,  0.0528,  0.0106,  0.0935]],\n",
       "              \n",
       "                       [[-0.0365,  0.0066, -0.0390,  ..., -0.0157, -0.0800, -0.0009],\n",
       "                        [-0.0052,  0.0574,  0.0090,  ...,  0.0742, -0.0032,  0.0428],\n",
       "                        [-0.0794, -0.0229, -0.0734,  ..., -0.0567, -0.1292,  0.0189],\n",
       "                        ...,\n",
       "                        [-0.0394,  0.0310, -0.0279,  ..., -0.0168, -0.1024,  0.0401],\n",
       "                        [-0.0608, -0.0230, -0.0768,  ..., -0.0791, -0.1619, -0.0137],\n",
       "                        [ 0.0080,  0.0470, -0.0125,  ..., -0.0470, -0.1008,  0.0198]]],\n",
       "              \n",
       "              \n",
       "                      [[[-0.0517,  0.0138,  0.0091,  ..., -0.0964, -0.1128, -0.2160],\n",
       "                        [-0.0901, -0.0131, -0.0328,  ..., -0.0753, -0.1480, -0.2997],\n",
       "                        [-0.1316, -0.0427, -0.0477,  ...,  0.2143,  0.0325, -0.1715],\n",
       "                        ...,\n",
       "                        [-0.1062, -0.0980, -0.2555,  ...,  0.1228,  0.1929,  0.1267],\n",
       "                        [-0.0808, -0.0615, -0.2231,  ...,  0.0354,  0.1053,  0.1067],\n",
       "                        [ 0.0382,  0.0500, -0.1280,  ..., -0.0329,  0.0187,  0.0471]],\n",
       "              \n",
       "                       [[ 0.0390,  0.0064, -0.0032,  ..., -0.0212,  0.0405,  0.1109],\n",
       "                        [ 0.0657,  0.0221,  0.0067,  ..., -0.0394,  0.0277,  0.1140],\n",
       "                        [ 0.0780,  0.0402,  0.0140,  ..., -0.1542, -0.0923,  0.0345],\n",
       "                        ...,\n",
       "                        [ 0.1284,  0.0944,  0.1466,  ..., -0.0601, -0.0909, -0.0611],\n",
       "                        [ 0.1268,  0.1004,  0.1375,  ..., -0.0225, -0.0667, -0.0199],\n",
       "                        [ 0.0805,  0.0782,  0.0989,  ...,  0.0093, -0.0346, -0.0124]],\n",
       "              \n",
       "                       [[ 0.0115, -0.0270,  0.0148,  ...,  0.0948,  0.1204,  0.1103],\n",
       "                        [ 0.0093, -0.0267,  0.0122,  ...,  0.0872,  0.1543,  0.1805],\n",
       "                        [ 0.0699,  0.0132,  0.0480,  ..., -0.0569,  0.0326,  0.1681],\n",
       "                        ...,\n",
       "                        [-0.0122, -0.0333,  0.1128,  ..., -0.0677, -0.1024, -0.0762],\n",
       "                        [-0.0060, -0.0286,  0.1164,  ..., -0.0068, -0.0438, -0.0311],\n",
       "                        [-0.1336, -0.1483, -0.0010,  ...,  0.0188, -0.0065, -0.0271]]],\n",
       "              \n",
       "              \n",
       "                      ...,\n",
       "              \n",
       "              \n",
       "                      [[[ 0.0091,  0.0148,  0.0047,  ...,  0.0155, -0.0006, -0.0199],\n",
       "                        [ 0.0003,  0.0212, -0.0132,  ...,  0.0024, -0.0058, -0.0204],\n",
       "                        [-0.0111,  0.0101, -0.0296,  ..., -0.0145, -0.0172, -0.0305],\n",
       "                        ...,\n",
       "                        [ 0.1001,  0.0914,  0.1308,  ...,  0.1580,  0.0904,  0.0784],\n",
       "                        [ 0.1161,  0.0818,  0.0829,  ..., -0.0602, -0.0694, -0.0502],\n",
       "                        [-0.1056, -0.1185, -0.1768,  ..., -0.2084, -0.1804, -0.1669]],\n",
       "              \n",
       "                       [[-0.0115,  0.0025, -0.0082,  ..., -0.0076, -0.0174, -0.0170],\n",
       "                        [-0.0027, -0.0114, -0.0061,  ..., -0.0282, -0.0226, -0.0226],\n",
       "                        [-0.0084, -0.0022, -0.0358,  ..., -0.0185, -0.0198, -0.0250],\n",
       "                        ...,\n",
       "                        [ 0.1296,  0.0981,  0.1489,  ...,  0.1566,  0.0795,  0.0969],\n",
       "                        [ 0.1608,  0.1052,  0.1026,  ..., -0.0652, -0.0643, -0.0391],\n",
       "                        [-0.1288, -0.1466, -0.1948,  ..., -0.2418, -0.2028, -0.1932]],\n",
       "              \n",
       "                       [[-0.0054, -0.0018,  0.0043,  ...,  0.0098,  0.0052,  0.0063],\n",
       "                        [ 0.0093,  0.0016, -0.0022,  ...,  0.0071,  0.0009, -0.0058],\n",
       "                        [-0.0014,  0.0039, -0.0077,  ...,  0.0031,  0.0150, -0.0082],\n",
       "                        ...,\n",
       "                        [ 0.0439,  0.0008,  0.0606,  ...,  0.0748,  0.0440,  0.0561],\n",
       "                        [ 0.1008,  0.0751,  0.1096,  ...,  0.0050,  0.0108,  0.0134],\n",
       "                        [-0.0888, -0.0728, -0.0929,  ..., -0.0665, -0.0390, -0.0484]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 0.0046,  0.0469, -0.0161,  ...,  0.0078, -0.0198,  0.0068],\n",
       "                        [ 0.0628,  0.0451,  0.0472,  ...,  0.0606,  0.0293,  0.0558],\n",
       "                        [ 0.0037,  0.0130,  0.0000,  ..., -0.0083, -0.0020,  0.0081],\n",
       "                        ...,\n",
       "                        [-0.0444, -0.0589, -0.0248,  ..., -0.0284, -0.0309, -0.0529],\n",
       "                        [-0.0098, -0.0432,  0.0090,  ..., -0.0426, -0.0181, -0.0280],\n",
       "                        [-0.0220, -0.0336,  0.0135,  ..., -0.0415, -0.0178, -0.0520]],\n",
       "              \n",
       "                       [[-0.0906, -0.0515, -0.1646,  ..., -0.1197, -0.1115, -0.0439],\n",
       "                        [ 0.0138,  0.0260, -0.0194,  ...,  0.0238,  0.0064,  0.0543],\n",
       "                        [-0.0932, -0.0475, -0.1127,  ..., -0.0865, -0.0740, -0.0666],\n",
       "                        ...,\n",
       "                        [ 0.0274,  0.0103,  0.0439,  ...,  0.0271,  0.0445,  0.0160],\n",
       "                        [ 0.0984,  0.0614,  0.1141,  ...,  0.0964,  0.1073,  0.0957],\n",
       "                        [-0.0151, -0.0118,  0.0486,  ...,  0.0291,  0.0563, -0.0022]],\n",
       "              \n",
       "                       [[-0.1369, -0.0793, -0.2125,  ..., -0.1363, -0.1512, -0.0639],\n",
       "                        [ 0.0157,  0.0514, -0.0182,  ...,  0.0491,  0.0196,  0.0781],\n",
       "                        [-0.1713, -0.0887, -0.1747,  ..., -0.1443, -0.1336, -0.1188],\n",
       "                        ...,\n",
       "                        [ 0.0520,  0.0159,  0.0700,  ...,  0.0444,  0.0654,  0.0269],\n",
       "                        [ 0.1314,  0.0913,  0.1623,  ...,  0.1423,  0.1550,  0.1381],\n",
       "                        [ 0.0038, -0.0204,  0.0787,  ...,  0.0860,  0.1260,  0.0388]]],\n",
       "              \n",
       "              \n",
       "                      [[[-0.0951,  0.0566,  0.1402,  ..., -0.0148,  0.0104, -0.0038],\n",
       "                        [ 0.0604,  0.1253, -0.1237,  ...,  0.0786, -0.0106, -0.0222],\n",
       "                        [ 0.1030, -0.1365, -0.1960,  ..., -0.0519, -0.0833,  0.0397],\n",
       "                        ...,\n",
       "                        [-0.1860,  0.0155,  0.3391,  ...,  0.2751,  0.1049, -0.1686],\n",
       "                        [ 0.0683,  0.1399,  0.0114,  ..., -0.0787, -0.2500, -0.0801],\n",
       "                        [ 0.0052, -0.0684, -0.0919,  ..., -0.1073,  0.0880,  0.1014]],\n",
       "              \n",
       "                       [[-0.0988,  0.0626,  0.1150,  ..., -0.0132,  0.0196, -0.0011],\n",
       "                        [ 0.0832,  0.1030, -0.1542,  ...,  0.1086,  0.0126, -0.0203],\n",
       "                        [ 0.1125, -0.1693, -0.1888,  ..., -0.0635, -0.0931,  0.0490],\n",
       "                        ...,\n",
       "                        [-0.1887,  0.0674,  0.4813,  ...,  0.3137,  0.1607, -0.1345],\n",
       "                        [ 0.1121,  0.2087,  0.0468,  ..., -0.0070, -0.2298, -0.0695],\n",
       "                        [ 0.0173, -0.0917, -0.1594,  ..., -0.0804,  0.0800,  0.1182]],\n",
       "              \n",
       "                       [[-0.1058,  0.0544,  0.1305,  ..., -0.0395,  0.0119, -0.0028],\n",
       "                        [ 0.0579,  0.1065, -0.1347,  ...,  0.1064,  0.0184, -0.0013],\n",
       "                        [ 0.1040, -0.1085, -0.1676,  ..., -0.0426, -0.0859,  0.0526],\n",
       "                        ...,\n",
       "                        [-0.1824,  0.0542,  0.3940,  ...,  0.2443,  0.1022, -0.1288],\n",
       "                        [ 0.0857,  0.1838,  0.0503,  ..., -0.0470, -0.2158, -0.0425],\n",
       "                        [ 0.0396, -0.0764, -0.1350,  ..., -0.0486,  0.1006,  0.0903]]]], device='cuda:0')),\n",
       "             ('features.0.bias',\n",
       "              tensor([-0.9705, -2.8070, -0.0371, -0.0795, -0.1159,  0.0252, -0.0752,\n",
       "                      -1.4181,  1.6454, -0.0990, -0.0161, -0.1282, -0.0658, -0.0345,\n",
       "                      -0.0743, -1.2977, -0.0505,  0.0121, -0.1013, -1.1887, -0.1380,\n",
       "                      -0.0492, -0.0789, -0.0405, -0.0958, -0.0705, -1.9374, -0.0850,\n",
       "                      -0.1388, -0.1968, -0.1279, -2.0095, -0.0476, -0.0604, -0.0351,\n",
       "                      -0.3843, -2.7823,  0.6605, -0.1655, -2.1293,  0.0543, -0.0274,\n",
       "                      -0.1703, -0.0593, -0.4215, -1.9394, -1.2094,  0.0153, -0.1081,\n",
       "                      -0.0248, -0.1503, -1.8516, -0.0928, -0.0177, -0.0700, -0.0582,\n",
       "                      -0.0630, -0.0721, -1.2678, -0.1176, -0.0441, -0.3259,  0.0507,\n",
       "                      -0.0146], device='cuda:0')),\n",
       "             ('features.3.weight',\n",
       "              tensor([[[[ 3.6245e-03,  1.4335e-03,  3.7217e-02, -2.0926e-02,  1.8121e-03],\n",
       "                        [ 2.4126e-02, -1.2056e-02,  7.1170e-02, -8.5224e-02,  1.3067e-02],\n",
       "                        [ 2.0966e-02, -1.0623e-01,  2.1572e-02, -6.9547e-02,  3.1583e-02],\n",
       "                        [-8.3392e-03, -3.9020e-02, -4.6621e-02,  2.2133e-02, -1.3252e-03],\n",
       "                        [-1.5370e-02,  8.6569e-03,  3.1479e-02,  1.6698e-02, -4.7130e-03]],\n",
       "              \n",
       "                       [[-5.4573e-03, -1.9087e-02, -3.2424e-02, -2.2006e-02, -1.2120e-02],\n",
       "                        [-7.4972e-03,  3.0946e-02,  3.1899e-02, -7.6327e-03, -1.4720e-02],\n",
       "                        [ 5.4830e-03,  8.0306e-02,  6.1262e-02,  1.3252e-02,  2.6240e-02],\n",
       "                        [-2.6938e-02,  5.9188e-03,  3.8373e-02, -6.3116e-03,  3.9863e-03],\n",
       "                        [ 2.0844e-03, -3.6292e-02, -2.1987e-03, -1.6570e-02,  5.6354e-03]],\n",
       "              \n",
       "                       [[-1.2587e-02, -5.9436e-02, -9.6281e-02,  1.6745e-02,  4.7471e-02],\n",
       "                        [-1.5036e-02, -1.1214e-01, -2.3924e-02,  3.3459e-02,  4.1306e-02],\n",
       "                        [-1.3845e-02,  4.1235e-02,  2.1580e-01,  2.9520e-02, -5.8914e-02],\n",
       "                        [ 1.2300e-02,  1.0087e-01,  2.1916e-02, -1.4008e-01, -3.1255e-02],\n",
       "                        [ 3.7845e-02,  5.2634e-02, -8.0348e-03, -8.1815e-02,  1.4355e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-1.2711e-02,  2.5915e-02,  4.1726e-02, -1.3706e-02,  1.4752e-03],\n",
       "                        [-1.0794e-02,  1.6434e-02,  8.1224e-02, -6.0144e-02, -5.1132e-02],\n",
       "                        [-1.3769e-02, -5.3764e-02,  1.1054e-01,  3.8881e-03, -5.8872e-02],\n",
       "                        [ 5.1748e-04,  2.3237e-02,  6.1570e-02,  5.5012e-02, -1.8824e-02],\n",
       "                        [-4.0024e-03,  9.4621e-03, -2.7666e-02, -3.6102e-03, -6.4362e-03]],\n",
       "              \n",
       "                       [[-9.5948e-04, -5.7705e-03,  4.5656e-02,  1.4085e-02, -4.3562e-02],\n",
       "                        [-4.4058e-03, -3.2867e-02,  4.1615e-02, -1.0995e-02,  2.3037e-02],\n",
       "                        [ 1.7953e-02,  5.5599e-03, -4.6110e-02, -6.3326e-02,  2.5256e-02],\n",
       "                        [ 5.1608e-03,  4.1085e-02,  2.1804e-02,  1.9433e-02,  3.1380e-02],\n",
       "                        [-2.9532e-02, -3.1130e-03,  5.4939e-02,  3.5771e-02, -3.1513e-03]],\n",
       "              \n",
       "                       [[ 1.7368e-03,  1.5858e-02, -3.9606e-02, -8.6650e-02,  4.2392e-02],\n",
       "                        [ 3.2755e-02,  2.0259e-02,  1.4398e-01, -1.5988e-01, -1.1522e-01],\n",
       "                        [ 1.1646e-02, -1.3281e-01,  2.5051e-01,  1.9387e-01, -1.4720e-01],\n",
       "                        [ 2.9876e-02, -7.6454e-02, -1.3301e-01,  1.2492e-01,  3.0456e-02],\n",
       "                        [ 4.8693e-02,  7.8451e-02, -3.9283e-02,  4.8439e-03,  2.0383e-02]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 2.1691e-02, -4.7307e-02,  3.3946e-02,  6.8847e-03,  2.6204e-02],\n",
       "                        [ 1.7683e-02, -5.2123e-02, -6.2835e-02, -3.3933e-02,  1.9128e-02],\n",
       "                        [ 9.2291e-03, -2.8800e-02, -1.8724e-01, -1.9551e-02, -2.1082e-02],\n",
       "                        [-1.4346e-02,  5.7543e-02, -4.1899e-02,  6.3682e-03,  4.2324e-04],\n",
       "                        [ 1.2244e-02,  2.3090e-02, -2.2817e-02,  2.3080e-03, -6.4166e-03]],\n",
       "              \n",
       "                       [[ 4.6469e-03,  7.4133e-03, -4.1122e-02,  2.9806e-02, -1.8110e-02],\n",
       "                        [ 9.6099e-03,  3.9877e-02, -4.1848e-02,  6.9251e-03, -2.5908e-02],\n",
       "                        [-1.0007e-02,  5.4747e-02,  2.6736e-02,  7.0483e-03,  2.6949e-02],\n",
       "                        [-9.5527e-03, -3.6075e-02,  1.4897e-02, -1.2595e-02, -1.8202e-02],\n",
       "                        [-1.4465e-04, -6.3873e-02,  5.8016e-02,  2.8762e-02, -9.7704e-04]],\n",
       "              \n",
       "                       [[-4.5600e-02,  9.9321e-02,  2.0830e-02, -1.7641e-02,  4.3317e-03],\n",
       "                        [-1.1306e-01,  1.0613e-01,  1.2299e-01, -1.5599e-02,  1.1943e-02],\n",
       "                        [-8.6221e-02,  5.1575e-02,  2.8114e-01,  3.8475e-03,  3.9631e-02],\n",
       "                        [ 2.7465e-02, -9.6338e-02,  1.6717e-01,  1.7882e-02, -3.1237e-02],\n",
       "                        [ 3.8818e-02, -1.5017e-01,  9.5537e-02,  8.2108e-02, -2.6467e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 2.1779e-03,  8.2149e-03,  2.2985e-02, -4.9943e-03, -2.2914e-02],\n",
       "                        [ 4.4609e-03,  4.6661e-02,  3.4394e-02,  5.2542e-02, -1.8957e-02],\n",
       "                        [-1.4202e-02,  5.9543e-02,  6.0551e-03,  3.6463e-02,  3.2430e-04],\n",
       "                        [-2.9488e-02,  4.5902e-02,  5.4980e-02, -2.6936e-03,  4.2663e-03],\n",
       "                        [-2.9711e-02, -1.6390e-02,  3.4428e-02, -9.1090e-03, -9.7971e-03]],\n",
       "              \n",
       "                       [[ 2.0002e-02, -4.4658e-03, -2.8359e-02,  7.4180e-03, -5.3517e-03],\n",
       "                        [ 1.3052e-02, -2.9312e-02, -6.5957e-02,  3.1932e-02, -2.3129e-03],\n",
       "                        [ 1.4003e-02, -5.7715e-02, -9.1541e-02,  2.8261e-03,  1.8726e-02],\n",
       "                        [-1.8143e-02, -4.5023e-02, -9.9762e-03,  1.2348e-02, -1.4161e-03],\n",
       "                        [ 3.3575e-03,  1.8285e-02,  1.2383e-02, -2.0430e-02,  2.4573e-02]],\n",
       "              \n",
       "                       [[-5.4195e-03,  4.5804e-04,  1.7469e-02, -2.2458e-03,  1.7530e-03],\n",
       "                        [-1.0281e-02,  3.1367e-02, -2.5485e-02,  1.2870e-02, -7.0466e-04],\n",
       "                        [-6.3459e-03,  7.3701e-02, -2.9366e-02,  4.9803e-02,  2.5816e-02],\n",
       "                        [-1.9934e-02,  3.8506e-02,  3.0833e-02,  2.5688e-03, -4.4185e-03],\n",
       "                        [ 1.8114e-03,  2.3094e-02,  4.6518e-02, -3.3974e-02, -1.1660e-02]]],\n",
       "              \n",
       "              \n",
       "                      [[[-1.0163e-02, -3.7849e-02, -4.6664e-02, -9.7829e-03,  3.7624e-02],\n",
       "                        [ 1.4684e-01, -6.4947e-02, -3.7485e-02, -9.2710e-03,  2.3640e-02],\n",
       "                        [ 1.8563e-02, -3.5328e-02, -9.6485e-03,  1.4210e-02,  2.7077e-02],\n",
       "                        [-6.7717e-03,  2.8039e-03,  4.6860e-03,  4.7567e-03, -6.6633e-03],\n",
       "                        [ 2.2155e-02, -1.2387e-03,  3.1133e-02,  3.1837e-02,  2.4495e-03]],\n",
       "              \n",
       "                       [[ 2.2964e-02, -3.2592e-02, -1.7921e-02,  5.2390e-03,  3.7561e-03],\n",
       "                        [ 3.8685e-02, -8.8764e-02, -5.3735e-02, -2.4427e-02,  1.7369e-02],\n",
       "                        [ 5.5512e-02, -7.1237e-02, -6.5386e-02, -3.7746e-02,  7.1429e-03],\n",
       "                        [ 3.1539e-02, -6.1858e-02, -7.4455e-02, -3.4278e-02, -9.2922e-03],\n",
       "                        [ 4.5122e-02, -8.4307e-03, -2.0509e-02, -1.4743e-02,  1.0508e-03]],\n",
       "              \n",
       "                       [[-9.5345e-03, -2.3179e-02, -1.4264e-02, -6.7151e-03, -2.8285e-02],\n",
       "                        [ 1.6202e-01, -2.8234e-02, -2.4645e-02, -1.0152e-02, -2.4914e-02],\n",
       "                        [ 8.9119e-02, -2.6843e-02, -3.7465e-02, -7.5062e-03, -1.9929e-03],\n",
       "                        [-4.5296e-02, -1.7844e-02, -5.1039e-03,  1.7445e-02,  1.3994e-02],\n",
       "                        [-2.0449e-02, -8.8112e-04,  1.3139e-02,  1.8497e-03, -5.4345e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 2.5179e-02,  8.3908e-03, -3.4752e-03, -6.7056e-03,  9.4571e-03],\n",
       "                        [ 1.4773e-01,  1.8297e-02, -8.2707e-03, -2.6735e-03, -2.9689e-03],\n",
       "                        [ 1.8945e-02, -2.4611e-02, -2.5792e-02,  5.7622e-03,  1.5490e-02],\n",
       "                        [ 1.0744e-02, -2.2880e-03,  1.3852e-02, -3.0690e-03, -1.4038e-03],\n",
       "                        [ 3.5738e-03, -1.7099e-02, -2.0962e-03, -5.2738e-04, -1.1953e-02]],\n",
       "              \n",
       "                       [[-1.8036e-02,  5.5999e-03, -4.3933e-03, -6.4853e-03,  1.1134e-02],\n",
       "                        [ 8.3012e-02, -8.5719e-02, -2.3690e-02,  2.7197e-02,  1.5338e-02],\n",
       "                        [ 1.6264e-02, -5.0213e-02, -2.9401e-02, -6.0407e-04,  1.5086e-02],\n",
       "                        [ 1.2260e-02,  6.5707e-03, -2.8775e-03, -2.0055e-02,  4.6619e-03],\n",
       "                        [ 1.4575e-02, -9.5385e-03, -2.4609e-02, -2.4521e-03,  1.0301e-03]],\n",
       "              \n",
       "                       [[-1.7248e-02,  7.3069e-03,  1.0429e-02, -1.4917e-02, -6.8504e-03],\n",
       "                        [ 6.2874e-02, -3.2599e-02,  5.7802e-03,  1.4551e-02, -6.1276e-03],\n",
       "                        [ 7.3852e-03, -6.0841e-02, -1.0494e-02,  3.6026e-03, -1.4746e-02],\n",
       "                        [-1.7262e-02,  1.5815e-03, -2.3888e-02, -9.7895e-03, -7.4106e-03],\n",
       "                        [ 6.9352e-03,  7.2060e-03, -1.9073e-02, -6.7136e-03,  6.1392e-03]]],\n",
       "              \n",
       "              \n",
       "                      ...,\n",
       "              \n",
       "              \n",
       "                      [[[ 1.7579e-02,  1.0168e-02,  4.4818e-02, -2.6664e-03, -2.2669e-02],\n",
       "                        [-1.5147e-02, -3.2701e-02,  1.4564e-02, -1.2843e-02, -3.2623e-02],\n",
       "                        [-3.0235e-02, -3.2706e-02, -3.4020e-02, -2.0842e-02, -3.2697e-02],\n",
       "                        [-4.2032e-03, -3.4967e-03, -1.0389e-02, -9.6090e-03, -3.1508e-03],\n",
       "                        [ 1.1118e-02,  8.8342e-03,  1.8196e-02,  7.5631e-03,  1.5709e-03]],\n",
       "              \n",
       "                       [[-9.1671e-03,  1.0798e-03, -2.4020e-02, -1.7457e-02, -7.8634e-03],\n",
       "                        [-1.5174e-02,  1.8126e-02,  1.3363e-02,  7.1205e-02,  5.8289e-02],\n",
       "                        [-1.0982e-02,  1.5266e-02, -2.4875e-02, -4.0705e-03, -2.4844e-02],\n",
       "                        [ 2.7485e-03,  1.6571e-02, -3.2608e-02, -5.6266e-02, -3.9944e-02],\n",
       "                        [-4.9014e-04,  1.9889e-02, -3.3673e-03, -2.2187e-02, -9.5482e-03]],\n",
       "              \n",
       "                       [[-2.6544e-02, -4.6466e-02, -1.1130e-01, -6.3725e-02,  2.4153e-02],\n",
       "                        [-3.1451e-02, -4.6735e-02, -5.0069e-02,  3.6642e-02,  8.7667e-02],\n",
       "                        [ 2.4192e-02,  2.9334e-02,  5.9852e-02,  4.2028e-02, -2.5984e-02],\n",
       "                        [ 2.5426e-02,  1.8227e-02,  4.0526e-02, -8.6688e-03, -6.0653e-02],\n",
       "                        [ 1.2664e-02,  1.5354e-02,  3.3492e-02,  1.3319e-02, -2.3908e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-1.5345e-02, -2.7450e-02, -3.7116e-02,  4.8070e-02,  1.0182e-02],\n",
       "                        [-5.4621e-02, -6.3059e-02, -5.1834e-02,  6.4004e-02,  2.5790e-02],\n",
       "                        [ 1.6780e-02,  1.3105e-02, -3.9535e-02, -1.2138e-03,  9.2518e-03],\n",
       "                        [ 1.4997e-02,  1.2259e-02, -2.3395e-02, -2.8443e-02,  2.4942e-02],\n",
       "                        [-4.0684e-03,  1.4466e-02,  6.2136e-03, -2.2554e-02, -2.0291e-03]],\n",
       "              \n",
       "                       [[-3.5230e-02,  8.0857e-03,  2.3528e-02, -1.0179e-02, -2.1804e-03],\n",
       "                        [-3.4595e-02,  3.9871e-03,  1.3592e-02, -7.9822e-02, -5.2531e-02],\n",
       "                        [-3.2725e-02,  3.4170e-02,  2.6620e-02, -2.4204e-03,  2.4924e-02],\n",
       "                        [ 1.8057e-02,  4.6391e-03, -1.2008e-02,  8.8407e-03,  1.9375e-02],\n",
       "                        [-2.9099e-02,  1.2933e-04, -1.7477e-02,  4.3010e-02,  6.6167e-02]],\n",
       "              \n",
       "                       [[ 4.6078e-02,  2.8413e-02, -2.6394e-02,  4.7513e-03, -7.6480e-02],\n",
       "                        [ 7.5960e-02,  4.4307e-02, -5.8771e-02,  1.7854e-02,  9.6082e-02],\n",
       "                        [ 5.8415e-02,  3.7583e-02, -6.7510e-02, -1.0312e-01, -2.3266e-02],\n",
       "                        [ 4.3488e-02,  6.0671e-02,  6.2699e-02,  3.7197e-03, -5.8098e-02],\n",
       "                        [ 7.3890e-05,  2.7182e-02,  4.3226e-02,  3.1200e-02, -5.9276e-03]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 2.5983e-02, -2.5757e-02, -2.7917e-02, -6.0154e-02, -2.1587e-02],\n",
       "                        [ 5.8470e-03, -4.0597e-02, -9.7084e-02, -1.1321e-01, -4.5251e-02],\n",
       "                        [-3.3056e-02, -1.7874e-02, -9.1402e-02, -1.0827e-01, -5.3423e-02],\n",
       "                        [-1.4856e-02, -3.3450e-03, -2.7355e-02, -4.8538e-02, -4.1184e-02],\n",
       "                        [ 9.0377e-04, -1.8180e-02, -3.1040e-02, -1.1854e-02, -2.6462e-02]],\n",
       "              \n",
       "                       [[ 7.2521e-02,  1.6407e-02, -1.5754e-01, -8.8864e-02, -2.1638e-02],\n",
       "                        [ 9.2281e-02,  8.5642e-02, -6.8582e-02, -7.8368e-02, -9.3343e-02],\n",
       "                        [ 1.1667e-01,  1.4296e-01,  1.4621e-02, -6.6599e-02, -1.8653e-01],\n",
       "                        [ 4.0418e-02,  8.7898e-02,  4.2521e-02, -8.9029e-03, -1.2440e-01],\n",
       "                        [ 2.0246e-02,  6.5828e-02,  9.6516e-02,  5.8696e-02, -7.2705e-02]],\n",
       "              \n",
       "                       [[-3.7801e-02, -1.1232e-02, -8.1927e-03,  1.2744e-02,  6.8491e-02],\n",
       "                        [-4.4532e-03, -9.6229e-03, -3.2608e-02,  2.9350e-03,  3.7486e-02],\n",
       "                        [ 3.3585e-02,  4.8458e-02,  2.6881e-02, -6.1178e-02, -2.1019e-02],\n",
       "                        [-1.9669e-02,  6.9724e-03,  4.2018e-02, -4.3931e-02, -7.6382e-02],\n",
       "                        [ 8.3337e-03,  2.5425e-03,  2.8063e-02, -3.4867e-02, -8.6687e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-1.2673e-02, -1.5327e-02,  1.6517e-02,  3.3334e-02,  1.3038e-02],\n",
       "                        [ 1.6805e-03, -1.8921e-02, -3.6130e-02,  3.2359e-02,  5.0130e-02],\n",
       "                        [-1.3607e-02, -3.3703e-02, -7.9256e-02, -5.1558e-03,  6.5082e-02],\n",
       "                        [-1.3499e-03,  1.9183e-02, -1.2600e-03, -1.5375e-02,  9.9117e-03],\n",
       "                        [-8.6383e-03, -6.4556e-03,  4.4472e-03, -7.7968e-04,  2.6764e-02]],\n",
       "              \n",
       "                       [[-3.1357e-02, -6.5565e-02, -4.5553e-02,  2.9694e-02,  5.0477e-02],\n",
       "                        [ 2.6388e-02, -6.5116e-03, -1.4299e-01, -2.9214e-02,  1.2134e-03],\n",
       "                        [ 5.2430e-02,  6.1108e-02, -2.8583e-02, -1.0998e-01, -3.4642e-02],\n",
       "                        [ 1.6018e-02,  3.1906e-02,  7.7014e-02, -1.1630e-02, -5.6960e-02],\n",
       "                        [ 4.9326e-02, -1.5627e-02,  4.2784e-02,  3.7805e-02, -2.6815e-02]],\n",
       "              \n",
       "                       [[ 9.0381e-03,  1.6271e-03,  9.9757e-03,  2.6084e-02,  6.8623e-03],\n",
       "                        [ 1.0622e-02, -6.1560e-03, -9.2547e-03,  1.8062e-02,  2.2674e-02],\n",
       "                        [ 2.8991e-02,  1.2231e-04, -1.6760e-02, -1.9807e-03,  3.1453e-02],\n",
       "                        [ 1.8964e-02,  2.8631e-02,  5.8559e-04, -1.0603e-02,  1.1077e-02],\n",
       "                        [ 5.4998e-03, -6.3534e-03,  2.9878e-03, -2.7331e-02, -2.4437e-02]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 7.0352e-03, -2.0708e-02, -1.9059e-02, -2.8221e-03,  9.0732e-03],\n",
       "                        [-2.2202e-02, -3.6482e-02, -3.3435e-02, -2.0294e-02, -1.1695e-02],\n",
       "                        [-7.7194e-03, -3.2950e-02, -5.5740e-02, -3.5089e-02, -1.5053e-02],\n",
       "                        [-1.2823e-02, -3.5408e-02, -4.2076e-02, -1.6718e-02, -1.8323e-02],\n",
       "                        [-3.9433e-03,  5.0854e-03, -2.3092e-03,  1.1569e-02,  1.1675e-02]],\n",
       "              \n",
       "                       [[ 1.8536e-02,  2.3997e-02,  1.8607e-02, -2.1316e-03,  3.0036e-02],\n",
       "                        [-9.1898e-03, -2.0234e-02, -2.0921e-02, -1.5293e-02, -4.1515e-03],\n",
       "                        [-2.5399e-02, -3.6147e-02,  1.1215e-02, -1.3978e-02,  1.0484e-02],\n",
       "                        [-1.7011e-03, -1.7085e-02,  1.9866e-02,  6.1436e-03,  4.9555e-03],\n",
       "                        [ 1.0626e-02,  1.2917e-02,  4.2516e-02,  1.5874e-02,  1.7050e-02]],\n",
       "              \n",
       "                       [[-4.0153e-03,  1.6083e-02,  8.7844e-03, -9.1274e-03,  2.2305e-02],\n",
       "                        [-2.0949e-02, -3.9927e-03, -3.2194e-02, -2.1123e-02, -1.3385e-03],\n",
       "                        [-4.2372e-03,  1.4911e-02,  3.8126e-03, -5.2412e-03,  9.8842e-03],\n",
       "                        [ 1.2249e-02,  3.7495e-03, -3.4932e-03,  6.0049e-03,  1.4841e-02],\n",
       "                        [ 1.3200e-02, -1.9996e-03,  2.0933e-02,  1.6309e-04, -8.3707e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-6.4706e-03, -3.2926e-02, -1.5733e-02, -1.7831e-02,  2.1256e-03],\n",
       "                        [-5.4066e-02, -5.4719e-02, -4.5444e-02, -5.6256e-02, -3.6743e-02],\n",
       "                        [ 2.8305e-02,  2.9538e-02,  4.4844e-02,  2.2081e-02,  3.5385e-02],\n",
       "                        [ 1.6821e-02,  1.6557e-02,  5.4560e-02,  1.7084e-02,  2.3373e-02],\n",
       "                        [-4.0657e-03, -8.0451e-03,  4.4215e-04,  7.5198e-03,  1.0919e-02]],\n",
       "              \n",
       "                       [[ 2.0678e-02,  2.0736e-02,  7.6578e-02,  2.9091e-02,  4.1559e-02],\n",
       "                        [-1.0118e-03,  5.5359e-02,  1.3228e-01,  6.7964e-02,  4.1530e-02],\n",
       "                        [-3.9346e-03,  3.5696e-02,  1.2877e-01,  6.1861e-02,  2.9380e-02],\n",
       "                        [ 2.5981e-02,  3.8598e-02,  1.3250e-01,  3.2665e-02, -1.8039e-02],\n",
       "                        [-1.9937e-02, -1.3913e-02, -1.0549e-02, -1.3653e-02, -2.4947e-03]],\n",
       "              \n",
       "                       [[ 7.0426e-02,  2.1494e-02, -9.1289e-03, -9.1144e-03, -5.7207e-03],\n",
       "                        [ 3.0881e-02, -1.0100e-02, -5.8352e-02, -3.8790e-02, -2.0214e-02],\n",
       "                        [-7.5705e-03, -4.3871e-02, -8.5624e-02, -5.4874e-02, -1.8760e-02],\n",
       "                        [-5.6932e-03, -4.2488e-02, -7.0337e-02, -3.9515e-02,  2.8042e-03],\n",
       "                        [ 4.9093e-02,  8.0856e-03, -2.0497e-02,  1.7022e-02,  2.4330e-02]]]], device='cuda:0')),\n",
       "             ('features.3.bias',\n",
       "              tensor([-0.1093, -0.1726, -0.0517, -0.0655, -0.0394,  0.1742, -0.1846,\n",
       "                       0.0955, -0.0251, -0.0301,  0.1365, -0.3227, -0.1483,  0.1279,\n",
       "                       0.0041, -0.0573, -0.0021,  0.2031,  0.3310,  0.0731,  0.4980,\n",
       "                      -0.0448, -0.0822,  0.0149, -0.0464,  0.0575,  0.0190,  0.0007,\n",
       "                      -0.1818, -0.0001, -0.1316, -0.1274,  0.0060, -0.4274,  0.1165,\n",
       "                      -0.1637, -0.0603, -0.2311,  0.2419, -0.0786,  0.0208, -0.0250,\n",
       "                      -0.0447,  0.0283, -0.1746,  0.0716, -0.0574,  0.1207, -0.0581,\n",
       "                      -0.1735, -0.1278,  0.0537,  0.0453,  0.0950,  0.0104,  0.1388,\n",
       "                      -0.1366,  0.2429,  0.1451,  0.4810, -0.1179, -0.1593,  0.1602,\n",
       "                       0.2256,  0.0960,  0.0299, -0.1758, -0.0950,  0.2345, -0.2452,\n",
       "                       0.1025, -0.2432, -0.0570, -0.0278, -0.3461,  0.0407, -0.2309,\n",
       "                       0.1912,  0.1629,  0.2960, -0.1370,  0.0052, -0.1660, -0.0770,\n",
       "                      -0.1169,  0.0683, -0.0984, -0.0913, -0.1649,  0.0625,  0.1023,\n",
       "                      -0.0058, -0.1724, -0.0086, -0.1200, -0.1863, -0.0538, -0.0674,\n",
       "                      -0.1480,  0.0262,  0.0400, -0.2199, -0.1498, -0.0831, -0.0679,\n",
       "                      -0.0790, -0.0421,  0.3717, -0.1835,  0.0612, -0.4772, -0.2170,\n",
       "                      -0.0705, -0.1000, -0.0447,  0.0866, -0.1361, -0.1826, -0.0379,\n",
       "                      -0.2928, -0.1391,  0.1413,  0.2949,  0.2761, -0.0446,  0.3673,\n",
       "                       0.0878, -0.0209,  0.2094,  0.4994,  0.1499, -0.1327,  0.2152,\n",
       "                       0.2325,  0.1449, -0.0823, -0.0327, -0.1276,  0.1118,  0.3876,\n",
       "                      -0.0379, -0.1795,  0.3152, -0.0400,  0.3261,  0.0109, -0.4202,\n",
       "                      -0.1933,  0.0464,  0.0145,  0.0418,  0.1487, -0.1557, -0.1616,\n",
       "                       0.4913, -0.1458,  0.1304, -0.0325, -0.0334, -0.3072,  0.3035,\n",
       "                      -0.0829,  0.1503, -0.0212,  0.1771,  0.2058,  0.1064, -0.1446,\n",
       "                      -0.1605, -0.0268, -0.0673,  0.4874,  0.1741,  0.1841, -0.0894,\n",
       "                       0.1441,  0.0530, -0.1310, -0.0846, -0.0370, -0.2381, -0.0806,\n",
       "                      -0.0975,  0.1490, -0.0031,  0.2869,  0.0939,  0.1201, -0.1244,\n",
       "                       0.0724,  0.0138,  0.0730], device='cuda:0')),\n",
       "             ('features.6.weight',\n",
       "              tensor([[[[ 2.4858e-02,  1.3116e-02,  2.8198e-02],\n",
       "                        [ 4.2541e-02,  5.7339e-02, -6.0905e-03],\n",
       "                        [-4.1912e-03,  9.3096e-03, -1.5442e-02]],\n",
       "              \n",
       "                       [[-3.8954e-03, -7.8586e-02, -5.1808e-02],\n",
       "                        [ 2.6484e-02, -4.9877e-02, -1.7763e-03],\n",
       "                        [ 8.2902e-03, -4.9339e-02,  3.1145e-02]],\n",
       "              \n",
       "                       [[ 1.6455e-02, -1.2150e-02,  1.7428e-02],\n",
       "                        [ 5.2012e-02, -6.7141e-03,  2.7325e-02],\n",
       "                        [ 7.5568e-03, -4.2402e-02, -2.7909e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-1.7289e-04,  2.4099e-02,  3.6948e-02],\n",
       "                        [-2.4145e-02, -3.5276e-02,  3.6910e-03],\n",
       "                        [ 1.9541e-02, -3.0342e-02, -3.5262e-02]],\n",
       "              \n",
       "                       [[ 1.3239e-02, -1.8624e-02, -5.3330e-02],\n",
       "                        [ 1.7639e-04, -1.4714e-02, -2.2829e-02],\n",
       "                        [-6.7702e-03,  2.3287e-02,  1.3873e-02]],\n",
       "              \n",
       "                       [[ 3.0512e-02,  7.5860e-03,  4.9459e-04],\n",
       "                        [-4.5703e-03, -1.2827e-02, -6.5061e-03],\n",
       "                        [-9.8111e-03, -1.5570e-02,  1.9379e-02]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 1.3660e-02, -3.9180e-02,  2.0014e-02],\n",
       "                        [ 5.7389e-02,  1.1934e-02,  2.3058e-02],\n",
       "                        [-2.2237e-02, -1.8879e-02, -2.0941e-02]],\n",
       "              \n",
       "                       [[-1.3402e-02,  1.9384e-02,  2.0018e-02],\n",
       "                        [ 2.2285e-03, -1.5328e-02, -2.0604e-02],\n",
       "                        [ 5.2206e-02, -4.0713e-02, -1.6011e-02]],\n",
       "              \n",
       "                       [[ 8.2728e-03, -1.2311e-02, -2.6974e-02],\n",
       "                        [ 2.5519e-03, -5.6930e-03, -5.1476e-02],\n",
       "                        [-2.8822e-02, -8.5134e-02, -6.5895e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 1.1531e-02,  2.4226e-02,  6.0344e-04],\n",
       "                        [-3.3641e-02,  7.7504e-02, -1.8122e-02],\n",
       "                        [ 2.2435e-02,  6.3408e-02, -1.0774e-03]],\n",
       "              \n",
       "                       [[-1.6893e-02,  5.7512e-03,  3.7639e-02],\n",
       "                        [-1.6503e-02, -3.2492e-03,  3.6293e-02],\n",
       "                        [ 3.2918e-02, -1.2321e-02,  1.4753e-03]],\n",
       "              \n",
       "                       [[ 5.8365e-02,  6.4994e-02,  6.6423e-02],\n",
       "                        [-1.3995e-02, -2.4027e-02, -1.2127e-02],\n",
       "                        [ 2.6365e-03, -1.4931e-02, -7.9207e-03]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 8.0889e-03,  9.6816e-03, -1.2593e-02],\n",
       "                        [ 1.1684e-02,  4.7408e-02,  4.9439e-02],\n",
       "                        [-1.1663e-02,  7.7298e-02,  2.3240e-02]],\n",
       "              \n",
       "                       [[ 9.5861e-03, -3.5039e-02, -1.5836e-02],\n",
       "                        [-2.2685e-02,  1.6363e-02,  1.8725e-02],\n",
       "                        [-1.7577e-02, -5.9963e-03, -1.2350e-02]],\n",
       "              \n",
       "                       [[-3.9020e-03, -5.0362e-03,  2.8650e-03],\n",
       "                        [ 2.3688e-02,  3.0239e-03, -5.0350e-02],\n",
       "                        [-3.0330e-02, -2.0917e-02, -9.9929e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 1.3766e-02, -4.8649e-03,  3.4593e-03],\n",
       "                        [-1.7299e-03,  9.8713e-03,  3.1695e-03],\n",
       "                        [-1.5713e-02,  2.4070e-02,  4.3523e-02]],\n",
       "              \n",
       "                       [[ 1.7398e-02,  3.9648e-02,  5.0599e-02],\n",
       "                        [-1.8082e-02,  2.3186e-02,  7.5082e-02],\n",
       "                        [-1.4652e-02, -1.4806e-03,  2.4491e-02]],\n",
       "              \n",
       "                       [[-3.6486e-03,  1.6976e-02,  9.5258e-03],\n",
       "                        [ 8.3009e-03,  3.0063e-03, -3.5994e-03],\n",
       "                        [-2.0521e-04, -9.6109e-04, -7.6914e-03]]],\n",
       "              \n",
       "              \n",
       "                      ...,\n",
       "              \n",
       "              \n",
       "                      [[[-2.1159e-02,  3.1025e-02, -5.1508e-04],\n",
       "                        [-2.0021e-02, -9.5897e-03,  2.7262e-02],\n",
       "                        [-9.7968e-03, -3.9145e-02,  9.2344e-03]],\n",
       "              \n",
       "                       [[-2.2890e-02, -6.2061e-03, -1.8220e-02],\n",
       "                        [ 2.6104e-02,  1.8208e-02, -3.5678e-02],\n",
       "                        [-1.9037e-02, -2.5568e-02, -3.2400e-03]],\n",
       "              \n",
       "                       [[ 4.5357e-04,  1.1460e-02,  3.0362e-02],\n",
       "                        [ 2.1339e-02,  1.1392e-02,  1.3384e-02],\n",
       "                        [ 2.2492e-02, -1.0258e-02, -5.8449e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-2.7361e-02,  1.4307e-02,  1.8851e-02],\n",
       "                        [ 3.9169e-02, -4.1835e-02, -1.2617e-03],\n",
       "                        [-2.4972e-02, -6.0201e-02,  1.1093e-02]],\n",
       "              \n",
       "                       [[ 1.3269e-02,  2.8252e-02, -2.0582e-02],\n",
       "                        [ 1.2612e-02,  5.3777e-02, -2.9175e-02],\n",
       "                        [ 4.3714e-02,  1.5523e-02, -3.4617e-02]],\n",
       "              \n",
       "                       [[-1.4312e-02, -1.1627e-02, -8.0607e-03],\n",
       "                        [-1.7464e-02,  9.4876e-03, -3.3993e-04],\n",
       "                        [-4.5948e-03,  2.4349e-02,  2.9515e-02]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 2.3024e-02,  3.7754e-02, -7.9412e-03],\n",
       "                        [ 2.3338e-02, -3.2373e-02, -4.1358e-02],\n",
       "                        [ 6.4814e-02,  2.6370e-02,  4.9119e-02]],\n",
       "              \n",
       "                       [[ 4.7991e-02, -2.4855e-02,  1.2162e-02],\n",
       "                        [-2.7339e-02,  3.9421e-02, -1.8773e-02],\n",
       "                        [-5.8830e-02,  2.6058e-02,  1.4781e-03]],\n",
       "              \n",
       "                       [[ 8.6110e-03, -2.6908e-02,  1.1525e-02],\n",
       "                        [ 5.6151e-02, -1.3285e-02, -7.1971e-03],\n",
       "                        [ 1.1242e-03, -4.7446e-02, -3.6017e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 3.5633e-02, -8.5117e-03,  6.2245e-02],\n",
       "                        [ 1.0972e-02,  1.9737e-02,  4.8234e-02],\n",
       "                        [-1.1555e-02,  4.3559e-03,  5.2946e-02]],\n",
       "              \n",
       "                       [[-1.0731e-03,  2.2912e-02,  2.2005e-02],\n",
       "                        [ 1.3288e-02,  4.9365e-02,  1.2220e-02],\n",
       "                        [ 1.0511e-02,  1.7318e-02, -1.8007e-02]],\n",
       "              \n",
       "                       [[-2.6693e-02, -1.8063e-02, -2.4829e-02],\n",
       "                        [-3.3943e-02, -1.4893e-03, -1.1168e-02],\n",
       "                        [-1.3225e-02, -4.6098e-03,  3.0331e-03]]],\n",
       "              \n",
       "              \n",
       "                      [[[-6.0866e-03, -2.5078e-02, -4.7683e-02],\n",
       "                        [ 2.4006e-02,  5.2576e-02, -2.1802e-02],\n",
       "                        [ 2.5606e-02,  1.1889e-02, -2.4519e-03]],\n",
       "              \n",
       "                       [[ 2.5286e-02, -2.8539e-02,  8.4982e-03],\n",
       "                        [ 6.8764e-03, -8.9434e-03,  3.4527e-03],\n",
       "                        [ 2.7720e-02,  1.2911e-02,  1.2079e-02]],\n",
       "              \n",
       "                       [[ 2.8175e-02,  1.3563e-02,  6.9976e-02],\n",
       "                        [ 2.7799e-02, -3.6929e-02,  1.0927e-01],\n",
       "                        [ 2.3212e-02, -3.4578e-02, -4.8821e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-3.3232e-04,  1.7463e-02, -9.6818e-03],\n",
       "                        [ 9.2093e-03,  1.5052e-02,  8.6647e-03],\n",
       "                        [-1.8870e-02, -2.1472e-02, -1.5119e-02]],\n",
       "              \n",
       "                       [[ 4.3916e-02, -5.5188e-05, -7.7297e-03],\n",
       "                        [-8.0442e-03, -3.4422e-02, -2.2207e-02],\n",
       "                        [-1.4105e-02, -2.7909e-02, -2.4631e-02]],\n",
       "              \n",
       "                       [[-2.6292e-02,  1.1666e-02, -1.6261e-02],\n",
       "                        [-5.7774e-02, -5.8361e-02, -4.8310e-02],\n",
       "                        [-4.1695e-02, -6.6255e-02, -5.7636e-02]]]], device='cuda:0')),\n",
       "             ('features.6.bias',\n",
       "              tensor([ 0.0253,  0.1194, -0.3336,  0.1127, -0.1158, -0.0597,  0.1142,\n",
       "                      -0.0951,  0.0315, -0.0111, -0.0494,  0.0910, -0.0521,  0.0366,\n",
       "                      -0.1202, -0.1342, -0.0050, -0.0014, -0.0763, -0.0491, -0.1255,\n",
       "                       0.2337, -0.0815,  0.0245, -0.0112,  0.1324,  0.2578, -0.0244,\n",
       "                       0.0732,  0.2082,  0.0599, -0.0111,  0.0706,  0.0054,  0.0139,\n",
       "                      -0.0690,  0.0956,  0.0821, -0.0892,  0.0236,  0.0209,  0.2095,\n",
       "                       0.0570,  0.1006, -0.1517,  0.0514,  0.0610,  0.1612,  0.0464,\n",
       "                      -0.2936, -0.1922, -0.0330, -0.2284,  0.0523, -0.0420, -0.2286,\n",
       "                      -0.2237, -0.1040, -0.0395,  0.0231,  0.0852, -0.0038,  0.1000,\n",
       "                       0.2224,  0.1404, -0.0833, -0.0153,  0.0008,  0.0200, -0.1466,\n",
       "                      -0.0525, -0.0818,  0.3813, -0.0347, -0.0427,  0.0190,  0.3016,\n",
       "                       0.1282,  0.1380,  0.1206, -0.0347,  0.1001, -0.0796, -0.0751,\n",
       "                       0.1023,  0.3221,  0.2122,  0.3072, -0.0082,  0.1320, -0.0683,\n",
       "                       0.1409, -0.0630, -0.0823,  0.0330, -0.1891, -0.0166, -0.1837,\n",
       "                       0.2002,  0.0698, -0.0130,  0.1311,  0.1935,  0.3463, -0.0420,\n",
       "                      -0.0163,  0.1196, -0.0683, -0.1502,  0.0099, -0.3099,  0.0900,\n",
       "                      -0.0811, -0.0524, -0.0902,  0.0507, -0.0269,  0.1488, -0.0608,\n",
       "                      -0.0707, -0.0344,  0.1997,  0.5527, -0.0715, -0.0095,  0.1007,\n",
       "                      -0.0095, -0.0257,  0.0978,  0.0113,  0.5273,  0.2715,  0.1356,\n",
       "                       0.2296, -0.1263,  0.0138, -0.0224, -0.1892, -0.0455, -0.0298,\n",
       "                       0.0467, -0.0517, -0.1754, -0.3434, -0.0057, -0.2355,  0.0537,\n",
       "                       0.3491, -0.2342, -0.1125,  0.0314, -0.0967, -0.1656, -0.0023,\n",
       "                      -0.2594,  0.0508,  0.0898, -0.0107,  0.1284,  0.0355,  0.0339,\n",
       "                      -0.0055, -0.0951, -0.0310,  0.0320,  0.0527,  0.1263,  0.1594,\n",
       "                       0.0359,  0.0353,  0.3697,  0.0179, -0.0228, -0.0956,  0.1729,\n",
       "                       0.0517,  0.1532, -0.0122,  0.2510, -0.0161,  0.0224, -0.0286,\n",
       "                       0.1197, -0.0255, -0.0419,  0.2101, -0.1348, -0.1648, -0.0257,\n",
       "                      -0.1292, -0.1646,  0.0539,  0.1147, -0.1193, -0.0591, -0.1505,\n",
       "                      -0.0270,  0.1071, -0.0881,  0.1148,  0.1745,  0.1580,  0.0525,\n",
       "                      -0.0288,  0.2443,  0.1458,  0.0111,  0.2188,  0.3987,  0.3208,\n",
       "                      -0.0872, -0.0773,  0.0551, -0.0742,  0.2994, -0.1120,  0.0204,\n",
       "                      -0.1653, -0.0840,  0.0919, -0.0163, -0.0135,  0.0341, -0.1523,\n",
       "                       0.0800, -0.0155,  0.0905, -0.1243,  0.0409,  0.0651, -0.0374,\n",
       "                      -0.0253,  0.2886, -0.1779,  0.0454,  0.0804, -0.1186,  0.1268,\n",
       "                       0.0133,  0.0045,  0.0838,  0.1505, -0.0749, -0.0511, -0.1585,\n",
       "                       0.1111,  0.0204,  0.0199,  0.2497,  0.2134,  0.1192,  0.1521,\n",
       "                       0.1012, -0.0202,  0.0189, -0.1558,  0.2609, -0.0290,  0.0654,\n",
       "                       0.0936, -0.0335, -0.1045,  0.1431,  0.0810, -0.1630,  0.1833,\n",
       "                      -0.0632,  0.0344,  0.0221, -0.1021,  0.0873,  0.0188, -0.1501,\n",
       "                       0.2394,  0.2239, -0.0240,  0.2311, -0.1096, -0.1695, -0.0364,\n",
       "                      -0.0508,  0.0767,  0.2064,  0.0014,  0.1128, -0.2035,  0.0806,\n",
       "                       0.2042,  0.3692,  0.1227, -0.0238, -0.0731, -0.0158,  0.1277,\n",
       "                      -0.0579,  0.2867, -0.0043, -0.0000,  0.2161,  0.1552, -0.0162,\n",
       "                      -0.0972, -0.1005, -0.0200, -0.1522, -0.0596,  0.0031, -0.3413,\n",
       "                       0.1624,  0.1747,  0.2099, -0.0636,  0.0547, -0.2457,  0.1255,\n",
       "                      -0.0449, -0.0594,  0.0294,  0.1510, -0.0400,  0.1121, -0.1552,\n",
       "                       0.5047, -0.0473, -0.0753,  0.0426, -0.1338, -0.4201,  0.3820,\n",
       "                       0.0894,  0.0530,  0.0582,  0.0429,  0.0251,  0.1971,  0.1012,\n",
       "                      -0.1119, -0.0281,  0.3542,  0.0238,  0.1228, -0.1242,  0.0309,\n",
       "                      -0.1154, -0.0282,  0.0237,  0.0159,  0.1002,  0.1117,  0.0085,\n",
       "                      -0.1309,  0.0925, -0.0054,  0.0893,  0.0534,  0.0665,  0.0354,\n",
       "                       0.0448, -0.0876,  0.1360,  0.1878, -0.0374,  0.0346,  0.0386,\n",
       "                      -0.1951, -0.0843, -0.0676, -0.0114,  0.1083, -0.1093, -0.0453,\n",
       "                       0.0213,  0.1211,  0.1122, -0.0952,  0.5096, -0.0033,  0.0549,\n",
       "                       0.0672,  0.1091,  0.0186, -0.0293,  0.0110, -0.0115], device='cuda:0')),\n",
       "             ('features.8.weight',\n",
       "              tensor([[[[-1.9727e-03, -8.1454e-03, -1.1364e-02],\n",
       "                        [-1.9331e-02,  7.0742e-04,  1.1423e-02],\n",
       "                        [-5.4075e-02, -1.1616e-03, -2.4432e-02]],\n",
       "              \n",
       "                       [[ 3.4995e-02,  1.3325e-02,  2.6029e-02],\n",
       "                        [-2.8163e-02, -6.2435e-03, -2.6879e-02],\n",
       "                        [ 3.4512e-03,  1.8146e-02,  1.4671e-02]],\n",
       "              \n",
       "                       [[-5.7246e-02, -4.7419e-02,  1.8569e-03],\n",
       "                        [-4.0166e-02, -4.6209e-02, -2.5661e-02],\n",
       "                        [-5.1450e-02, -4.8978e-02,  2.5351e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-1.8442e-02, -2.3369e-02,  9.7001e-03],\n",
       "                        [-4.4311e-02, -7.6195e-03, -1.7835e-02],\n",
       "                        [-5.1811e-02, -3.5133e-02, -4.5530e-02]],\n",
       "              \n",
       "                       [[-3.6594e-03, -1.1409e-03, -4.4672e-02],\n",
       "                        [-5.2415e-02, -3.1765e-02, -5.2399e-02],\n",
       "                        [-3.1243e-03, -1.1095e-02, -4.4262e-02]],\n",
       "              \n",
       "                       [[-1.9879e-02, -1.5260e-03,  1.5899e-02],\n",
       "                        [ 5.0753e-03, -1.4877e-02, -2.3683e-02],\n",
       "                        [ 2.5881e-02,  3.3199e-02,  8.0742e-03]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 2.1040e-02,  2.1385e-02,  5.2782e-02],\n",
       "                        [-5.6299e-03,  2.3969e-02,  3.3785e-02],\n",
       "                        [-9.1422e-03,  3.4282e-02,  2.3602e-02]],\n",
       "              \n",
       "                       [[-2.3864e-02, -1.8330e-02, -8.2681e-03],\n",
       "                        [ 3.1609e-02,  1.3647e-02,  4.5288e-02],\n",
       "                        [-3.5653e-02,  2.4665e-02,  1.0097e-02]],\n",
       "              \n",
       "                       [[ 1.3495e-03,  3.9706e-02,  2.5353e-02],\n",
       "                        [ 3.0785e-02,  1.1350e-02, -3.0964e-03],\n",
       "                        [-2.8046e-02,  2.2734e-03, -1.8373e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-1.9495e-02, -4.1632e-03, -1.7916e-02],\n",
       "                        [ 1.9192e-02,  2.6518e-02,  2.5553e-03],\n",
       "                        [ 1.3746e-02, -1.0167e-02,  8.5529e-03]],\n",
       "              \n",
       "                       [[ 2.7528e-02,  2.5611e-02,  2.5169e-02],\n",
       "                        [-1.0245e-02, -2.6295e-02, -1.2180e-02],\n",
       "                        [-3.3978e-02,  6.6900e-03,  3.8497e-02]],\n",
       "              \n",
       "                       [[-2.1693e-03,  2.1725e-02,  1.1239e-02],\n",
       "                        [ 1.2229e-02, -2.1039e-02,  7.5299e-03],\n",
       "                        [-2.8304e-03, -5.2147e-03, -1.2766e-02]]],\n",
       "              \n",
       "              \n",
       "                      [[[-3.2854e-02, -2.4809e-02, -1.8468e-02],\n",
       "                        [-5.4112e-03,  6.1196e-03,  1.2564e-02],\n",
       "                        [ 4.7752e-02,  1.8786e-02,  4.1288e-02]],\n",
       "              \n",
       "                       [[-7.3721e-03, -1.5722e-03,  2.6161e-03],\n",
       "                        [ 4.6837e-03, -4.8271e-03,  7.7925e-03],\n",
       "                        [-6.7745e-03,  2.7660e-03,  1.4312e-02]],\n",
       "              \n",
       "                       [[ 4.7401e-04, -2.2783e-02, -2.7205e-03],\n",
       "                        [-1.5211e-02, -2.4859e-02,  2.3492e-02],\n",
       "                        [ 5.2511e-03, -2.0474e-02,  6.4651e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-3.5232e-02, -1.4924e-02,  1.7090e-02],\n",
       "                        [-2.3447e-03, -8.3897e-03, -9.8155e-04],\n",
       "                        [-2.0932e-02,  3.6793e-03, -2.2268e-02]],\n",
       "              \n",
       "                       [[-4.1739e-03, -4.7361e-03, -9.9844e-03],\n",
       "                        [-6.9427e-03, -2.8120e-02, -3.8678e-02],\n",
       "                        [ 4.7891e-03,  3.3325e-02,  1.1383e-02]],\n",
       "              \n",
       "                       [[-4.2148e-02,  1.9920e-04, -3.4742e-02],\n",
       "                        [-1.1400e-04,  1.3599e-03, -9.2337e-03],\n",
       "                        [ 7.1887e-02,  5.0243e-02,  5.7336e-02]]],\n",
       "              \n",
       "              \n",
       "                      ...,\n",
       "              \n",
       "              \n",
       "                      [[[ 4.1041e-02,  3.8478e-02,  1.6273e-02],\n",
       "                        [ 7.2963e-03, -1.7565e-02, -4.6216e-02],\n",
       "                        [-2.0127e-02, -2.6285e-02, -4.7344e-02]],\n",
       "              \n",
       "                       [[ 1.2926e-02,  2.2733e-02,  2.5179e-02],\n",
       "                        [ 1.4479e-02,  2.5440e-02,  2.9994e-02],\n",
       "                        [-2.4523e-03,  1.5145e-02,  1.1197e-02]],\n",
       "              \n",
       "                       [[-1.9929e-02, -1.1274e-02, -1.0443e-02],\n",
       "                        [ 1.2676e-02,  1.2854e-02,  7.3260e-03],\n",
       "                        [ 1.6059e-02,  5.4710e-03, -6.3094e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 3.0193e-02,  5.0501e-03, -1.1663e-03],\n",
       "                        [ 2.0806e-02,  3.3353e-04, -1.8671e-02],\n",
       "                        [-1.4426e-02, -3.5664e-02, -1.7456e-02]],\n",
       "              \n",
       "                       [[ 5.9916e-02,  8.5456e-03,  7.5015e-03],\n",
       "                        [ 6.1019e-02,  1.6958e-02, -3.3683e-02],\n",
       "                        [ 1.1938e-02,  2.5057e-02, -1.4893e-02]],\n",
       "              \n",
       "                       [[-4.5513e-02, -5.0426e-02, -4.0966e-02],\n",
       "                        [-1.9976e-02, -2.8136e-02, -1.9349e-02],\n",
       "                        [-1.7795e-02, -2.2510e-02, -4.6813e-02]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 7.6864e-03, -2.4943e-02,  1.5596e-02],\n",
       "                        [-2.4905e-02,  4.3126e-03,  4.9446e-02],\n",
       "                        [ 4.5479e-03,  4.5500e-02,  4.5485e-02]],\n",
       "              \n",
       "                       [[ 1.0515e-02,  2.4689e-02, -5.7297e-03],\n",
       "                        [ 4.6129e-02,  3.0522e-02,  3.4931e-02],\n",
       "                        [-3.5357e-02,  4.2213e-02, -2.5364e-03]],\n",
       "              \n",
       "                       [[ 2.4712e-02,  3.6402e-02,  6.1931e-04],\n",
       "                        [-2.2153e-03, -2.0391e-02, -1.0495e-03],\n",
       "                        [-1.8919e-02, -2.0683e-02, -2.5401e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-1.2330e-02, -3.4162e-02, -3.6028e-04],\n",
       "                        [-6.7105e-03, -2.4533e-02, -8.1501e-03],\n",
       "                        [-1.2227e-02,  1.3290e-02,  1.3857e-02]],\n",
       "              \n",
       "                       [[ 9.8366e-03,  2.5934e-02,  1.7975e-02],\n",
       "                        [ 3.7568e-02,  4.7179e-02,  3.4257e-03],\n",
       "                        [ 2.3867e-02, -1.8648e-03, -1.2810e-02]],\n",
       "              \n",
       "                       [[ 8.7276e-03,  2.9231e-02, -7.5102e-03],\n",
       "                        [ 3.0996e-03, -1.1793e-02, -1.5828e-02],\n",
       "                        [-1.2800e-02, -5.2393e-02,  3.5761e-03]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 5.0408e-04,  4.8085e-02,  2.7914e-02],\n",
       "                        [ 6.0406e-04, -2.9416e-02, -8.9835e-03],\n",
       "                        [ 9.5136e-03,  2.7638e-03,  1.5747e-02]],\n",
       "              \n",
       "                       [[ 1.6563e-02,  3.2774e-02,  1.8208e-02],\n",
       "                        [-5.4806e-03,  2.7492e-02,  1.1908e-02],\n",
       "                        [-5.4084e-02, -1.8953e-02, -3.8581e-02]],\n",
       "              \n",
       "                       [[-3.8647e-02, -3.5310e-02,  1.4264e-02],\n",
       "                        [-5.7655e-02, -4.8306e-02,  2.8926e-02],\n",
       "                        [-3.6082e-02,  3.1528e-03,  6.1847e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 1.3315e-02,  2.5940e-02,  4.9885e-02],\n",
       "                        [ 3.3364e-03, -4.9974e-02,  9.1023e-04],\n",
       "                        [ 1.5312e-02,  1.8448e-02,  1.0342e-04]],\n",
       "              \n",
       "                       [[-3.2358e-02, -1.6284e-02,  2.3682e-02],\n",
       "                        [-1.1630e-03, -1.3672e-02,  3.6939e-03],\n",
       "                        [-2.3587e-02,  4.0935e-03,  3.6314e-02]],\n",
       "              \n",
       "                       [[-2.9316e-02, -1.7552e-02, -2.6831e-02],\n",
       "                        [-3.4132e-02,  4.5964e-02, -1.1251e-02],\n",
       "                        [ 2.7400e-04,  7.0052e-02,  6.7779e-03]]]], device='cuda:0')),\n",
       "             ('features.8.bias',\n",
       "              tensor([-0.0629,  0.1260,  0.2991,  0.1123,  0.2853,  0.1280,  0.1828,\n",
       "                      -0.0310,  0.5452,  0.1565, -0.1342, -0.1711,  0.2740,  0.2277,\n",
       "                       0.2672,  0.2619,  0.0456, -1.4264,  0.6095,  0.0824,  0.0964,\n",
       "                       0.1568, -0.2905,  0.1264,  0.2099,  0.0725, -0.0003, -0.2209,\n",
       "                       0.1117,  0.4803,  0.0355, -0.2749, -0.5027,  0.0007, -0.1227,\n",
       "                       0.1488, -0.0082,  0.1321,  0.2992, -0.1647,  0.3246,  0.5349,\n",
       "                       0.2450,  0.5994,  0.0338,  0.0664,  0.4898,  0.4211,  0.6347,\n",
       "                      -0.0947, -0.0785,  0.2570,  0.0767, -0.2393,  0.1710,  0.4168,\n",
       "                       0.2698, -0.4600, -0.0845,  0.2778,  0.2675,  0.3321,  0.0507,\n",
       "                      -0.0863, -0.2101, -0.0522, -0.0411, -0.1392, -0.3049,  0.3017,\n",
       "                       0.2358, -0.3874, -0.0601,  0.9037, -0.1400,  0.5320,  0.2318,\n",
       "                       0.2069,  0.2338, -0.3639,  0.2204, -0.5408, -0.6493,  0.0985,\n",
       "                       0.3403,  0.2848, -0.1167, -0.1494,  0.2674, -0.1469,  0.0529,\n",
       "                       0.2966,  0.0148,  0.1732,  0.1521,  0.0111, -0.0244, -0.0926,\n",
       "                      -0.0547,  0.1777,  0.0076, -0.0156, -0.0200,  0.0301,  0.3428,\n",
       "                       0.1848, -0.1893,  0.1192,  0.0438,  0.2409, -0.1295,  0.2866,\n",
       "                      -0.3398,  0.1724, -0.0140, -0.0332, -0.2137, -0.3905,  0.6983,\n",
       "                       0.0287,  0.0748,  0.1588,  0.1620,  0.0387,  0.1498, -0.0706,\n",
       "                       0.2821,  0.3019, -0.4792,  0.0948, -0.2582, -0.3050,  0.3716,\n",
       "                       0.0425,  0.2296,  0.0718,  0.1077,  0.2120,  0.2802,  0.3751,\n",
       "                      -0.1298,  0.2235,  0.0156,  0.2990,  0.5827,  0.4477,  0.3193,\n",
       "                       0.1551,  0.1608,  0.2476, -0.6709,  0.2488,  0.1474,  0.0933,\n",
       "                       0.2399, -0.2536, -0.0582,  0.2241,  0.1592, -0.0568,  0.4653,\n",
       "                      -0.2117, -0.3184, -1.2303,  0.1207, -0.0101,  0.3314,  0.8853,\n",
       "                      -0.0505,  0.4009, -0.1292, -0.0665,  0.2693,  0.2968, -0.5151,\n",
       "                      -0.1394,  0.0340,  0.5053,  0.2388,  0.0904,  0.2861, -0.0537,\n",
       "                      -0.2304, -0.0695,  0.0628, -0.0835, -0.1461,  0.1807, -0.1780,\n",
       "                      -0.0978, -0.0064,  0.8115,  0.2005,  0.3071, -0.2360,  0.0732,\n",
       "                       0.1117,  0.2837,  0.0670,  0.1611, -0.6772,  0.1382,  0.1734,\n",
       "                      -0.0442, -0.2506, -0.2652,  0.1673, -0.1684, -0.0851,  0.1246,\n",
       "                       0.0421,  0.1100,  0.4331,  0.1476, -0.0224,  0.2177,  0.0591,\n",
       "                       0.2993,  0.0848, -0.2806,  0.0374, -0.1994, -0.4867,  0.2039,\n",
       "                       0.1982,  0.1383,  0.0302, -0.2072,  0.7092, -0.1091,  0.4816,\n",
       "                       0.2920,  0.0479, -0.7954, -0.0441,  0.1333, -0.1962,  0.3378,\n",
       "                       0.4062, -0.1133,  0.0517,  0.0774,  0.2679,  0.3282,  0.1893,\n",
       "                      -0.0216,  0.0531, -0.2430,  0.1592,  0.5749,  0.0773, -0.0072,\n",
       "                       0.3115,  0.0332,  0.2763,  0.1434], device='cuda:0')),\n",
       "             ('features.10.weight',\n",
       "              tensor([[[[ 4.5008e-03, -7.7347e-03, -1.5003e-02],\n",
       "                        [-3.0285e-02, -4.4114e-02, -1.7563e-02],\n",
       "                        [-1.4331e-02, -3.6714e-02, -5.2003e-02]],\n",
       "              \n",
       "                       [[ 3.3324e-02,  9.8262e-03, -6.4676e-03],\n",
       "                        [ 1.4188e-02,  2.4336e-02,  3.1040e-02],\n",
       "                        [-1.9439e-02,  1.9590e-02, -1.6543e-02]],\n",
       "              \n",
       "                       [[ 3.3328e-05, -6.3628e-04,  1.5421e-02],\n",
       "                        [ 1.1260e-02,  5.7387e-04,  1.7457e-02],\n",
       "                        [ 7.1211e-03,  1.0711e-03,  2.1856e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-1.6804e-02, -2.1512e-02, -1.0703e-02],\n",
       "                        [-1.0698e-02, -5.3842e-02, -1.9335e-02],\n",
       "                        [ 1.6292e-02,  4.4939e-04,  6.5307e-05]],\n",
       "              \n",
       "                       [[ 1.2114e-03,  1.1926e-03, -1.7091e-03],\n",
       "                        [-1.7243e-02, -2.7680e-02,  1.9918e-05],\n",
       "                        [ 5.2268e-04, -1.6247e-03, -4.3989e-03]],\n",
       "              \n",
       "                       [[ 1.9892e-02,  2.9338e-02,  2.0883e-02],\n",
       "                        [ 1.3522e-02,  3.1604e-02,  1.2854e-02],\n",
       "                        [ 5.2760e-03,  1.2341e-02,  7.5441e-03]]],\n",
       "              \n",
       "              \n",
       "                      [[[-1.4208e-02,  2.3224e-03, -2.5112e-02],\n",
       "                        [-7.6076e-03,  9.4883e-03, -4.6517e-03],\n",
       "                        [ 1.3253e-02,  1.1511e-02, -1.4017e-02]],\n",
       "              \n",
       "                       [[ 2.1794e-02,  1.0763e-02,  1.1629e-02],\n",
       "                        [ 4.1082e-02, -7.8142e-03,  1.9052e-02],\n",
       "                        [ 6.6561e-02,  3.6654e-02,  2.0193e-03]],\n",
       "              \n",
       "                       [[-1.8703e-02, -4.9244e-02, -2.9182e-02],\n",
       "                        [-1.5992e-02,  1.6744e-04,  4.5855e-03],\n",
       "                        [-6.2261e-03,  4.1283e-03,  8.5156e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-5.8011e-02, -7.3355e-02, -6.5635e-02],\n",
       "                        [-5.9603e-02, -5.7514e-02, -4.2397e-02],\n",
       "                        [-5.5825e-02, -6.3877e-02, -6.0623e-02]],\n",
       "              \n",
       "                       [[-3.6692e-02, -3.2975e-02, -1.0278e-02],\n",
       "                        [-3.1835e-02, -5.3899e-03, -3.2266e-02],\n",
       "                        [-3.2715e-02, -3.3150e-02, -2.8616e-02]],\n",
       "              \n",
       "                       [[-8.2592e-03, -2.5406e-02, -1.7791e-02],\n",
       "                        [-1.1399e-02, -2.3384e-02, -1.1688e-02],\n",
       "                        [-2.2682e-02, -3.1900e-02, -7.9640e-03]]],\n",
       "              \n",
       "              \n",
       "                      [[[ 1.8231e-02, -2.9938e-02, -3.0918e-02],\n",
       "                        [ 6.0674e-04, -4.5208e-02, -4.3832e-03],\n",
       "                        [-6.0434e-03, -6.1846e-02, -1.4543e-02]],\n",
       "              \n",
       "                       [[ 1.3111e-02,  1.5979e-02, -5.2871e-03],\n",
       "                        [-1.2694e-02,  1.2434e-03,  2.2471e-02],\n",
       "                        [ 2.1080e-02, -6.7459e-03,  1.9622e-02]],\n",
       "              \n",
       "                       [[ 3.0896e-02, -3.0819e-02,  5.1553e-03],\n",
       "                        [-5.7161e-04, -1.2514e-02,  2.3466e-02],\n",
       "                        [ 2.5366e-02,  3.1077e-02,  2.8808e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[-2.0043e-02, -3.3691e-02, -3.3587e-02],\n",
       "                        [-3.1222e-02, -4.3832e-02, -2.9784e-02],\n",
       "                        [-5.5385e-02, -6.6676e-02, -5.1232e-02]],\n",
       "              \n",
       "                       [[-1.8431e-02, -1.0482e-02, -1.3102e-02],\n",
       "                        [ 5.6541e-03, -2.6609e-02,  2.4034e-03],\n",
       "                        [-2.3949e-02,  8.9346e-03,  3.3861e-03]],\n",
       "              \n",
       "                       [[ 5.7008e-03, -1.4837e-02,  1.3501e-03],\n",
       "                        [-5.6762e-04,  6.3385e-03,  5.1770e-03],\n",
       "                        [ 5.5514e-03, -1.4905e-03, -2.3526e-02]]],\n",
       "              \n",
       "              \n",
       "                      ...,\n",
       "              \n",
       "              \n",
       "                      [[[ 1.8694e-02, -5.8550e-03,  7.5898e-03],\n",
       "                        [ 5.4762e-03, -5.1852e-02,  5.3048e-03],\n",
       "                        [ 5.4274e-02, -3.1494e-03, -5.8523e-03]],\n",
       "              \n",
       "                       [[ 1.0686e-02, -1.6323e-02,  2.0675e-02],\n",
       "                        [ 1.9367e-03, -3.2638e-02,  9.5754e-04],\n",
       "                        [ 2.1652e-03,  2.7449e-02,  5.5060e-02]],\n",
       "              \n",
       "                       [[ 5.0755e-03, -2.6414e-02,  1.8585e-02],\n",
       "                        [-2.5866e-03,  1.2334e-03,  4.8944e-03],\n",
       "                        [ 5.3379e-04, -8.6257e-03, -2.3603e-02]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 2.3444e-02,  9.6432e-03,  1.7831e-02],\n",
       "                        [-1.8331e-02, -5.7185e-02, -2.1030e-02],\n",
       "                        [ 1.0187e-02, -3.0167e-02, -8.0193e-03]],\n",
       "              \n",
       "                       [[ 1.7629e-02,  2.5001e-02, -4.5713e-02],\n",
       "                        [ 2.5250e-02,  3.9404e-02,  3.0033e-02],\n",
       "                        [ 1.6265e-02,  5.3023e-02,  8.1163e-03]],\n",
       "              \n",
       "                       [[ 4.1792e-03,  8.5183e-03,  3.5242e-03],\n",
       "                        [ 2.7552e-02,  4.5622e-02,  3.8444e-03],\n",
       "                        [ 2.1285e-02,  2.0103e-02,  1.6193e-03]]],\n",
       "              \n",
       "              \n",
       "                      [[[-1.4135e-02, -3.3222e-02, -5.2931e-03],\n",
       "                        [-8.9475e-03, -3.3265e-02,  4.3306e-03],\n",
       "                        [ 1.7092e-02, -2.8524e-02, -2.5048e-02]],\n",
       "              \n",
       "                       [[-1.4275e-02, -2.7405e-02,  1.1635e-02],\n",
       "                        [-2.2799e-02, -3.2017e-02, -9.8221e-03],\n",
       "                        [-2.0387e-02, -3.1692e-02, -9.8188e-03]],\n",
       "              \n",
       "                       [[-2.0245e-02, -7.1923e-03,  5.0653e-03],\n",
       "                        [ 9.2916e-03, -1.4393e-02, -2.8599e-02],\n",
       "                        [ 1.5616e-03, -1.7386e-02, -7.2756e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 1.4874e-02, -9.5781e-03,  2.7291e-02],\n",
       "                        [ 1.7682e-03, -1.8440e-02,  7.5908e-03],\n",
       "                        [-6.8078e-03, -1.3905e-03,  3.8714e-03]],\n",
       "              \n",
       "                       [[-2.7931e-02, -3.4299e-02, -1.5767e-02],\n",
       "                        [-1.5908e-02, -4.4211e-02, -1.5330e-02],\n",
       "                        [-1.9257e-02, -2.3017e-02,  7.2739e-04]],\n",
       "              \n",
       "                       [[-3.6868e-02,  1.5530e-02, -7.3656e-04],\n",
       "                        [-3.2403e-03, -2.5565e-02, -6.2375e-04],\n",
       "                        [-7.5825e-03, -1.7276e-02, -1.2110e-02]]],\n",
       "              \n",
       "              \n",
       "                      [[[-3.0703e-03, -3.1071e-02,  4.9807e-04],\n",
       "                        [ 1.3618e-02,  6.6683e-03, -1.3059e-02],\n",
       "                        [-1.5806e-02,  9.5659e-03, -2.8366e-02]],\n",
       "              \n",
       "                       [[-1.5622e-02,  1.0870e-02, -4.3911e-02],\n",
       "                        [-6.0366e-03,  4.7041e-02, -6.7736e-03],\n",
       "                        [ 2.3879e-02,  1.5275e-02,  2.1209e-02]],\n",
       "              \n",
       "                       [[-3.4367e-02, -2.7406e-03, -4.0337e-02],\n",
       "                        [-2.1270e-02, -2.4037e-02, -2.0263e-02],\n",
       "                        [-7.9488e-03, -1.4826e-03, -6.8217e-03]],\n",
       "              \n",
       "                       ...,\n",
       "              \n",
       "                       [[ 1.0063e-02, -2.2878e-02, -5.5036e-03],\n",
       "                        [ 3.2441e-02, -1.3513e-02,  2.5942e-03],\n",
       "                        [-2.8327e-02, -3.6309e-02, -1.9826e-02]],\n",
       "              \n",
       "                       [[-9.0379e-03,  1.6569e-02,  5.9027e-04],\n",
       "                        [-7.4690e-04,  3.3369e-02,  2.2972e-02],\n",
       "                        [ 1.0572e-02,  8.1467e-03, -3.4821e-02]],\n",
       "              \n",
       "                       [[ 7.4795e-03,  1.9532e-03,  8.9045e-03],\n",
       "                        [-7.3107e-03, -2.9493e-02,  1.2008e-03],\n",
       "                        [ 8.8704e-03,  1.1838e-02,  7.4983e-04]]]], device='cuda:0')),\n",
       "             ('features.10.bias',\n",
       "              tensor([ 0.0380,  0.4363,  0.1992,  0.4787,  0.1945,  0.0461,  0.0896,\n",
       "                      -0.0909,  0.1207,  0.1895,  0.0450,  0.4843,  0.2463, -0.0786,\n",
       "                      -0.0410, -0.4561,  0.6745,  0.0437, -0.3859,  0.2315,  0.1211,\n",
       "                       0.3223,  1.1335,  0.2115,  0.4799,  0.2896,  0.0646,  0.0240,\n",
       "                       0.1504,  0.2882,  0.0672,  0.2694,  0.0101,  0.3425,  0.1637,\n",
       "                       0.2052,  0.0751,  0.1107,  0.2657,  0.1228, -0.0465,  0.5106,\n",
       "                       0.1446,  0.4451,  0.3364,  0.4072, -0.0210,  1.1052,  0.1231,\n",
       "                       0.3105,  0.2302,  0.3644,  0.3806,  0.1763,  0.6309, -0.0138,\n",
       "                       0.2625,  0.2036,  0.3424,  0.0191,  0.1561,  0.2294, -0.0874,\n",
       "                       0.0316,  0.0862,  0.2475, -0.0078,  0.2237,  0.2482,  0.0718,\n",
       "                      -0.0660, -0.0681, -0.0210,  0.4095,  0.4249,  0.3053, -0.1138,\n",
       "                       0.0248,  0.1145,  0.0074, -0.0462,  0.6349, -0.1142,  0.6199,\n",
       "                       0.0928, -0.0147,  0.4625,  0.6339, -0.0178,  0.1375,  0.0122,\n",
       "                       0.1934,  0.0191,  0.0814,  0.0171,  0.0449,  0.1913,  0.2072,\n",
       "                       0.0018,  0.4555,  0.3821,  0.3259, -0.1008,  0.5078,  0.0125,\n",
       "                       0.5967,  0.8065,  0.4625,  0.1434,  0.0098, -0.0385,  0.0317,\n",
       "                       0.1349,  0.5540,  0.1388,  0.1851,  0.0351, -0.1292,  0.1549,\n",
       "                       0.0666,  0.2270, -0.0807,  0.5249,  0.1212,  0.0257,  2.4021,\n",
       "                       0.5470, -0.1738,  0.1062, -0.0423,  0.0376,  0.2426, -0.0767,\n",
       "                       0.0219,  0.4895,  0.4147,  0.0483, -0.0359,  0.1444, -0.2099,\n",
       "                       0.2691,  0.0615,  0.0047,  0.5872,  0.2177,  0.8185,  0.4179,\n",
       "                       0.0415,  0.1757,  0.9283,  0.1397,  0.3681,  0.6795, -0.0466,\n",
       "                       0.6185,  0.0991,  0.3758,  0.1208,  0.7321,  0.1422, -0.0159,\n",
       "                       0.4913,  0.0695,  0.1672,  0.6833,  0.3698,  0.1170,  0.0046,\n",
       "                       0.4440,  0.1650,  0.1285,  0.2183,  0.7052,  0.0041,  0.1612,\n",
       "                       0.1117, -0.0008,  0.2619,  0.0731,  0.2620,  0.0153,  0.2089,\n",
       "                       0.3399,  0.0394,  0.0987,  0.2809,  0.1330, -0.0987,  0.1587,\n",
       "                      -0.1098, -0.1290,  0.1509,  0.1398,  0.0019,  0.1087, -0.1650,\n",
       "                       0.0561,  0.1740,  0.4182,  0.1161,  0.3530,  0.8590,  0.2884,\n",
       "                       0.1183, -0.0507,  0.0288,  0.1351,  0.1938, -0.0193,  0.5422,\n",
       "                       0.5368,  0.4404,  0.5066,  0.3581,  0.5300,  0.5118,  0.2010,\n",
       "                       0.3700,  0.6013,  0.0043,  0.0190, -0.1798,  0.0655,  0.6466,\n",
       "                       0.2411,  0.1317,  0.6079, -0.0045,  0.6037,  0.2969,  0.6362,\n",
       "                       0.3371,  0.1005,  0.4111,  0.5291,  0.1821,  0.2334,  0.0359,\n",
       "                       0.0938,  0.1678,  0.0977,  0.4354,  0.0347, -0.0722,  0.1484,\n",
       "                      -0.0934,  0.4198, -0.0049,  0.1896,  0.5445,  0.5332,  0.3611,\n",
       "                       0.3276,  0.5591,  0.0020, -0.0665], device='cuda:0')),\n",
       "             ('classifier.fc1.weight',\n",
       "              tensor([[ 2.8755e-03,  1.9165e-03, -4.9982e-03,  ..., -3.3482e-03,\n",
       "                       -1.0174e-03, -9.9940e-03],\n",
       "                      [-5.8229e-03, -1.2313e-02, -1.2323e-02,  ..., -8.5002e-03,\n",
       "                       -1.6102e-03, -1.4016e-02],\n",
       "                      [-1.3275e-03, -3.0392e-03,  7.1168e-03,  ...,  1.7376e-02,\n",
       "                       -1.5704e-02, -8.1268e-03],\n",
       "                      ...,\n",
       "                      [-8.1606e-03, -1.0667e-02, -1.3804e-02,  ...,  5.8743e-03,\n",
       "                        3.3002e-03,  1.6865e-03],\n",
       "                      [-5.0199e-03, -1.7502e-02, -7.8969e-03,  ..., -1.4685e-02,\n",
       "                       -1.3075e-02,  9.0908e-03],\n",
       "                      [ 5.1778e-03, -6.0779e-04, -9.3765e-03,  ...,  1.5881e-02,\n",
       "                        4.6715e-03, -3.9510e-03]], device='cuda:0')),\n",
       "             ('classifier.fc1.bias',\n",
       "              tensor([-8.5333e-03,  5.1695e-04,  4.5851e-04,  ...,  5.5101e-03,\n",
       "                      -8.8263e-03, -2.2823e-03], device='cuda:0')),\n",
       "             ('classifier.fc2.weight',\n",
       "              tensor([[-1.7875e-02,  6.0618e-04,  5.2963e-03,  ..., -2.6597e-03,\n",
       "                       -7.8833e-03, -1.3484e-02],\n",
       "                      [-1.3863e-02,  1.3207e-02,  8.9084e-03,  ..., -9.8230e-04,\n",
       "                       -7.0363e-03, -3.2897e-03],\n",
       "                      [ 1.4092e-03, -1.7286e-02,  9.0501e-03,  ...,  1.0675e-03,\n",
       "                       -8.4186e-03, -8.0845e-04],\n",
       "                      ...,\n",
       "                      [-1.6949e-02, -2.0179e-02, -7.1454e-03,  ..., -1.4188e-02,\n",
       "                       -7.8431e-03,  7.9822e-03],\n",
       "                      [-1.9544e-03,  3.3688e-03, -8.7392e-03,  ..., -1.8902e-02,\n",
       "                       -7.5087e-03,  8.4347e-03],\n",
       "                      [ 1.9029e-02, -1.2634e-02,  2.0870e-02,  ..., -4.5847e-03,\n",
       "                        3.9306e-03,  2.0887e-02]], device='cuda:0')),\n",
       "             ('classifier.fc2.bias', tensor(1.00000e-02 *\n",
       "                     [-3.8443, -0.3190, -1.0334,  ..., -0.8348, -3.4338, -4.9148], device='cuda:0')),\n",
       "             ('classifier.fc3.weight',\n",
       "              tensor([[-1.7286e-01, -3.3158e-02, -4.5259e-03,  ..., -2.6428e-02,\n",
       "                       -3.9165e-02, -8.8125e-02],\n",
       "                      [-4.1994e-02, -1.5830e-03,  7.8628e-03,  ..., -3.8653e-02,\n",
       "                       -5.0699e-02, -6.6769e-02],\n",
       "                      [-3.6007e-02, -2.3434e-02, -1.6646e-02,  ..., -1.2421e-02,\n",
       "                       -3.6897e-02, -4.0691e-02],\n",
       "                      ...,\n",
       "                      [-3.5170e-02,  8.5819e-03, -2.6119e-02,  ..., -9.5101e-03,\n",
       "                        2.8694e-02, -4.1022e-02],\n",
       "                      [-2.1008e-02,  1.9521e-02,  6.9397e-03,  ...,  2.2954e-02,\n",
       "                       -1.7436e-03, -1.9127e-02],\n",
       "                      [-6.0792e-02,  2.3039e-02,  4.1772e-04,  ..., -1.9228e-02,\n",
       "                        1.4898e-03, -2.3924e-02]], device='cuda:0')),\n",
       "             ('classifier.fc3.bias', tensor(1.00000e-02 *\n",
       "                     [ 3.5449, -1.6236,  1.5002,  2.4637, -3.1404,  0.6800,  0.8838,\n",
       "                       0.4986,  4.7436,  1.1300,  1.4450,  1.1107,  3.4198, -2.0046,\n",
       "                      -1.1806,  3.5722,  1.6258,  1.5547,  3.4981,  3.6150, -1.4638,\n",
       "                      -3.1728, -1.0314,  0.0564, -2.7240, -0.4942, -3.4800,  2.7927,\n",
       "                       1.4610,  2.5868, -0.3396, -4.8799, -0.0389, -0.1582, -0.7080,\n",
       "                       0.4573,  0.3127,  3.4236, -4.0885,  1.8282, -1.0011, -0.8378,\n",
       "                       1.0390, -0.8675, -3.4964, -0.3239, -2.9261,  0.7662, -2.8487,\n",
       "                      -2.0789,  3.8331, -1.6695,  0.1495,  5.2645, -0.7996,  1.5395,\n",
       "                      -2.6015,  0.8670, -2.3512,  0.6807, -1.4424,  2.8752, -3.6459,\n",
       "                       0.7825,  1.3564, -3.1977, -0.9096, -0.4753,  2.6270, -2.4435,\n",
       "                      -0.1116, -2.2853,  2.3912,  0.0902,  1.8309,  0.5660,  1.3115,\n",
       "                       1.5469, -0.0862, -1.0795, -1.7786,  2.0098, -3.4122,  0.3666,\n",
       "                       4.1249,  3.2478, -1.5350,  0.8140, -2.9364,  1.7097,  2.8358,\n",
       "                      -3.8108, -0.5812, -3.1471,  0.0556,  4.2851, -3.3067,  2.1568,\n",
       "                       0.2263,  1.0656,  2.2770, -0.9225], device='cuda:0'))])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.state_dict ()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save the checkpoint\n",
    "\n",
    "Now that your network is trained, save the model so you can load it later for making predictions. You probably want to save other things such as the mapping of classes to indices which you get from one of the image datasets: `image_datasets['train'].class_to_idx`. You can attach this to the model as an attribute which makes inference easier later on.\n",
    "\n",
    "```model.class_to_idx = image_datasets['train'].class_to_idx```\n",
    "\n",
    "Remember that you'll want to completely rebuild the model later so you can use it for inference. Make sure to include any information you need in the checkpoint. If you want to load the model and keep training, you'll want to save the number of epochs as well as the optimizer state, `optimizer.state_dict`. You'll likely want to use this trained model in the next part of the project, so best to save it now."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.to ('cpu') #no need to use cuda for saving/loading model.\n",
    "# TODO: Save the checkpoint \n",
    "model.class_to_idx = train_image_datasets.class_to_idx #saving mapping between predicted class and class name, \n",
    "#second variable is a class name in numeric \n",
    "\n",
    "#creating dictionary \n",
    "checkpoint = {'classifier': model.classifier,\n",
    "              'state_dict': model.state_dict (),\n",
    "              'mapping':    model.class_to_idx\n",
    "             }        \n",
    "\n",
    "torch.save (checkpoint, 'project_checkpoint.pth')\n",
    "#you should also store other hyper-parameters like the number of epochs, the learning_rate, arch param \n",
    "#along with the checkpoint. This parameters are required in case you need to continue training your model\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the checkpoint\n",
    "\n",
    "At this point it's good to write a function that can load a checkpoint and rebuild the model. That way you can come back to this project and keep working on it without having to retrain the network."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: Write a function that loads a checkpoint and rebuilds the model\n",
    "\n",
    "def loading_model (file_path):\n",
    "    checkpoint = torch.load (file_path) #loading checkpoint from a file\n",
    "    model = models.alexnet (pretrained = True) #function works solely for Alexnet\n",
    "    #you can use the arch from the checkpoint and choose the model architecture in a more generic way:\n",
    "    #model = getattr(models, checkpoint['arch']\n",
    "        \n",
    "    model.classifier = checkpoint ['classifier']\n",
    "    model.load_state_dict (checkpoint ['state_dict'])\n",
    "    model.class_to_idx = checkpoint ['mapping']\n",
    "    \n",
    "    for param in model.parameters(): \n",
    "        param.requires_grad = False #turning off tuning of the model\n",
    "    \n",
    "    return model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://download.pytorch.org/models/alexnet-owt-4df8aa71.pth\" to /root/.torch/models/alexnet-owt-4df8aa71.pth\n",
      "100%|██████████| 244418560/244418560 [00:10<00:00, 24009225.54it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "AlexNet(\n",
       "  (features): Sequential(\n",
       "    (0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))\n",
       "    (1): ReLU(inplace)\n",
       "    (2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (3): Conv2d(64, 192, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
       "    (4): ReLU(inplace)\n",
       "    (5): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (6): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (7): ReLU(inplace)\n",
       "    (8): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (9): ReLU(inplace)\n",
       "    (10): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (11): ReLU(inplace)\n",
       "    (12): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "  )\n",
       "  (classifier): Sequential(\n",
       "    (fc1): Linear(in_features=9216, out_features=4096, bias=True)\n",
       "    (relu1): ReLU()\n",
       "    (dropout1): Dropout(p=0.3)\n",
       "    (fc2): Linear(in_features=4096, out_features=2048, bias=True)\n",
       "    (relu2): ReLU()\n",
       "    (dropout2): Dropout(p=0.3)\n",
       "    (fc3): Linear(in_features=2048, out_features=102, bias=True)\n",
       "    (output): LogSoftmax()\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Now let's test if we did everything in correct way.\n",
    "\n",
    "model_verify = loading_model ('project_checkpoint.pth')\n",
    "model_verify"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Inference for classification\n",
    "\n",
    "Now you'll write a function to use a trained network for inference. That is, you'll pass an image into the network and predict the class of the flower in the image. Write a function called `predict` that takes an image and a model, then returns the top $K$ most likely classes along with the probabilities. It should look like \n",
    "\n",
    "```python\n",
    "probs, classes = predict(image_path, model)\n",
    "print(probs)\n",
    "print(classes)\n",
    "> [ 0.01558163  0.01541934  0.01452626  0.01443549  0.01407339]\n",
    "> ['70', '3', '45', '62', '55']\n",
    "```\n",
    "\n",
    "First you'll need to handle processing the input image such that it can be used in your network. \n",
    "\n",
    "## Image Preprocessing\n",
    "\n",
    "You'll want to use `PIL` to load the image ([documentation](https://pillow.readthedocs.io/en/latest/reference/Image.html)). It's best to write a function that preprocesses the image so it can be used as input for the model. This function should process the images in the same manner used for training. \n",
    "\n",
    "First, resize the images where the shortest side is 256 pixels, keeping the aspect ratio. This can be done with the [`thumbnail`](http://pillow.readthedocs.io/en/3.1.x/reference/Image.html#PIL.Image.Image.thumbnail) or [`resize`](http://pillow.readthedocs.io/en/3.1.x/reference/Image.html#PIL.Image.Image.thumbnail) methods. Then you'll need to crop out the center 224x224 portion of the image.\n",
    "\n",
    "Color channels of images are typically encoded as integers 0-255, but the model expected floats 0-1. You'll need to convert the values. It's easiest with a Numpy array, which you can get from a PIL image like so `np_image = np.array(pil_image)`.\n",
    "\n",
    "As before, the network expects the images to be normalized in a specific way. For the means, it's `[0.485, 0.456, 0.406]` and for the standard deviations `[0.229, 0.224, 0.225]`. You'll want to subtract the means from each color channel, then divide by the standard deviation. \n",
    "\n",
    "And finally, PyTorch expects the color channel to be the first dimension but it's the third dimension in the PIL image and Numpy array. You can reorder dimensions using [`ndarray.transpose`](https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.ndarray.transpose.html). The color channel needs to be first and retain the order of the other two dimensions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def process_image(image):\n",
    "    ''' Scales, crops, and normalizes a PIL image for a PyTorch model,\n",
    "        returns an Numpy array\n",
    "    '''\n",
    "    #size = 256, 256\n",
    "    im = Image.open (image) #loading image\n",
    "    width, height = im.size #original size\n",
    "    #proportion = width/ float (height) #to keep aspect ratio\n",
    "    \n",
    "    if width > height: \n",
    "        height = 256\n",
    "        im.thumbnail ((50000, height), Image.ANTIALIAS)\n",
    "    else: \n",
    "        width = 256\n",
    "        im.thumbnail ((width,50000), Image.ANTIALIAS)\n",
    "        \n",
    "    \n",
    "    width, height = im.size #new size of im\n",
    "    #crop 224x224 in the center\n",
    "    reduce = 224\n",
    "    left = (width - reduce)/2 \n",
    "    top = (height - reduce)/2\n",
    "    right = left + 224 \n",
    "    bottom = top + 224\n",
    "    im = im.crop ((left, top, right, bottom))\n",
    "    \n",
    "    #preparing numpy array\n",
    "    np_image = np.array (im)/255 #to make values from 0 to 1\n",
    "    np_image -= np.array ([0.485, 0.456, 0.406]) \n",
    "    np_image /= np.array ([0.229, 0.224, 0.225])\n",
    "    \n",
    "    np_image= np_image.transpose ((2,0,1))\n",
    "    #np_image.transpose (1,2,0)\n",
    "    return np_image\n",
    "    \n",
    "    # TODO: Process a PIL image for use in a PyTorch model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To check your work, the function below converts a PyTorch tensor and displays it in the notebook. If your `process_image` function works, running the output through this function should return the original image (except for the cropped out portions)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fe786596550>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe78558d7b8>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 251,
       "width": 259
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def imshow(image, ax=None, title=None):\n",
    "    if ax is None:\n",
    "        fig, ax = plt.subplots()\n",
    "    \n",
    "    # PyTorch tensors assume the color channel is the first dimension\n",
    "    # but matplotlib assumes is the third dimension\n",
    "    image = np.array (image)\n",
    "    image = image.transpose((1, 2, 0))\n",
    "    \n",
    "    # Undo preprocessing\n",
    "    mean = np.array([0.485, 0.456, 0.406])\n",
    "    std = np.array([0.229, 0.224, 0.225])\n",
    "    image = std * image + mean\n",
    "    #image = np.multiply (std, image) + mean\n",
    "    \n",
    "    # Image needs to be clipped between 0 and 1 or it looks like noise when displayed\n",
    "    image = np.clip(image, 0, 1)\n",
    "    \n",
    "    ax.imshow(image)\n",
    "    \n",
    "    return ax\n",
    "\n",
    "image_path = 'flowers/train/10/image_07086.jpg'\n",
    "img = process_image(image_path)\n",
    "#img.shape\n",
    "imshow(img)\n",
    "#(np.array([0.229, 0.224, 0.225])).shape\n",
    "#img = img.transpose ((1,2,0))\n",
    "#img"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Class Prediction\n",
    "\n",
    "Once you can get images in the correct format, it's time to write a function for making predictions with your model. A common practice is to predict the top 5 or so (usually called top-$K$) most probable classes. You'll want to calculate the class probabilities then find the $K$ largest values.\n",
    "\n",
    "To get the top $K$ largest values in a tensor use [`x.topk(k)`](http://pytorch.org/docs/master/torch.html#torch.topk). This method returns both the highest `k` probabilities and the indices of those probabilities corresponding to the classes. You need to convert from these indices to the actual class labels using `class_to_idx` which hopefully you added to the model or from an `ImageFolder` you used to load the data ([see here](#Save-the-checkpoint)). Make sure to invert the dictionary so you get a mapping from index to class as well.\n",
    "\n",
    "Again, this method should take a path to an image and a model checkpoint, then return the probabilities and classes.\n",
    "\n",
    "```python\n",
    "probs, classes = predict(image_path, model)\n",
    "print(probs)\n",
    "print(classes)\n",
    "> [ 0.01558163  0.01541934  0.01452626  0.01443549  0.01407339]\n",
    "> ['70', '3', '45', '62', '55']\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#mapping = train_image_datasets.class_to_idx\n",
    "\n",
    "#indeces = np.array ([1, 10, 100, 101, 102])\n",
    "#classes = pd.DataFrame ([mapping [item] for item in indeces]) #replacing indeces with classes\n",
    "#classes = np.array (classes) #converting to Numpy array \n",
    "\n",
    "def predict(image_path, model, topkl):\n",
    "    ''' Predict the class (or classes) of an image using a trained deep learning model.\n",
    "    '''\n",
    "    # TODO: Implement the code to predict the class from an image file\n",
    "    image = process_image (image_path) #loading image and processing it using above defined function\n",
    "    \n",
    "    #we cannot pass image to model.forward 'as is' as it is expecting tensor, not numpy array\n",
    "    #converting to tensor\n",
    "    im = torch.from_numpy (image).type (torch.FloatTensor)\n",
    "    \n",
    "    im = im.unsqueeze (dim = 0) #used to make size of torch as expected. as forward method is working with batches,\n",
    "    #doing that we will have batch size = 1 \n",
    "        \n",
    "    with torch.no_grad ():\n",
    "        output = model.forward (im)\n",
    "    output_prob = torch.exp (output) #converting into a probability\n",
    "    \n",
    "    probs, indeces = output_prob.topk (topkl)\n",
    "    probs = probs.numpy () #converting both to numpy array\n",
    "    indeces = indeces.numpy () \n",
    "    \n",
    "    probs = probs.tolist () [0] #converting both to list\n",
    "    indeces = indeces.tolist () [0]\n",
    "    \n",
    "    \n",
    "    mapping = {val: key for key, val in\n",
    "                model.class_to_idx.items()\n",
    "                }\n",
    "    \n",
    "    classes = [mapping [item] for item in indeces]\n",
    "    #classes = pd.DataFrame ([mapping [item] for item in indeces]) #replacing indeces with classes\n",
    "    classes = np.array (classes) #converting to Numpy array \n",
    "    \n",
    "    return probs, classes\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sanity Checking\n",
    "\n",
    "Now that you can use a trained model for predictions, check to make sure it makes sense. Even if the testing accuracy is high, it's always good to check that there aren't obvious bugs. Use `matplotlib` to plot the probabilities for the top 5 classes as a bar graph, along with the input image. It should look like this:\n",
    "\n",
    "<img src='assets/inference_example.png' width=300px>\n",
    "\n",
    "You can convert from the class integer encoding to actual flower names with the `cat_to_name.json` file (should have been loaded earlier in the notebook). To show a PyTorch tensor as an image, use the `imshow` function defined above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe76916dc88>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 251,
       "width": 259
      }
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3wAAAIvCAYAAAAvednPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3XncZ2VdP/7XW3AXWcMlUqw0SHOUcaNYBnFByzCzXEvkq7lUrlnWVwWz7Zcp6tdcC9HSNDVRU9REQSVxmREsxDVHQQQCBlBZFLl+f5xz14ebzz33fc98hvuea57Px+N+nLnPuc513ufcn1lec51znWqtBQAAgP7caKULAAAAYNsQ+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABAp3Ze6QJge1JV30xy6yQbV7gUAAD6tm+Sy1trd9qaTgQ+WJ5b3/zmN99j//3332OlCwEAoF9nn312rrzyyq3uR+CD5dm4//7777F+/fqVrgMAgI6tXbs2GzZs2Li1/XiGDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApL16HZTrz3DOz+7N3X+kyAADYxjYdt2mlS9hqRvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4ZqCqjqqqVlVHbWU/J4z97DuTwmaoqtaNtR27jH2OHfdZt41qOqWq2rboGwAAeiDwkSSpqn3HcHbCStcyZzUHYAAA2B4IfGxLr06yf5LPrnQhAACwI9p5pQugX621i5JctNJ1AADAjsoI3xQ1eGZVfamqrqqq71TVq6tq16raWFUbl9HX2qp6d1VdWFVXV9W3quo1VXW7zex2o6p6TlV9eTz+uVV1XFXdeoFj7DPW91/jMS6uqvdV1b2XWOOxSb45fvuE8TbKua+jprS/R1V9oKouraorqurUqvrFaf1Oe4avqg6uqveP53V1VZ1fVadX1TETbVqSJ4zffnOino1LPKcHV9UHq+qi8RjfqKqXVtVuS9kfAAB6YIRvur9N8rQk5yV5Q5IfJvnVJPdJcuMkP1pKJ1X1K0nenaSSvCvJt5KsHfs+sqp+qbW2ccquxyU5JMk/J3lvkgcneVaSg6vqoNbaVRPHOCDJR5LskeTDSf4lyV5JHp7kU1X1a621Dy5S6ilJdkvyzCRnJjlxYtsZ89reK8kfJvl0kr9Lcockv57k5Kq6R2vtK5s7UFUdkeQDSS5P8r4k3xlr3z/J05O8eGz64vEc1iR5ZZJLx/WXZhFV9aJx/0uS/GuSC5PcPckfJHloVR3YWrt8sX4AAGB7J/DNU1UHZwhkX01y39bapeP6P0ny0SS3zxDcFuvnVklOyHCN17XWPjmx7Y+S/FWGMPmgKbv/UpJ7tNa+Nbb/4yTvTPKIJM9L8pJx/c4ZQuGtkhzWWjt14hi3T/K5JH9fVfu21q5eqNbW2injyNkzk5zRWjt2M6f2y0me2Fo7YeJYT0nyunH/p29m3yR5coaR5XWttTMnN1TVXhM1HTtO1rImySsWCMbXU1WHZQh7n07y0Lmf37jtqCRvGrc/e5F+1i+wab+l1AEAAKuBWzqvb+42wj+fDAuttR8m+eNl9HNkkj2TvGMy7I1elmRjkgdW1R2m7PvKubA3HvvaDEHv2iRHT7T75SQ/k+T/TYa9cZ/zkvx1ktsmOXwZdS/mtMmwNzo+yTUZRkCX6sr5K8Zn/rbWM8blkyd/fmP/J2QYsXzcDI4DAACrnhG+67vnuPzUlG2nZwg2S3HAuPzY/A2ttWuq6hNJ9h2P9+15TU6dss9/VdU5Sfatqt3GMHPguPmOC7wf787jcv8ki93WuVSfn1Lbj6rqgiS7L2H/t2YYqfxMVb0jycczhMhzZ1TfgRluuf2NqvqNKdtvkuQnqmrP1trFC3XSWls7bf048nfAtG0AALDaCHzXt+u4vGD+htbaj6tqwZCwQD/fXWD73Pppk4hc79ij85Pccez70gwjiEkyLdhMutUi25djoWforkmy02I7t9b+ZXy28bkZRiufkvxPkPrj1tq/bWV9e2b4XB+zSLtbJVnqzxIAALZLbum8vrnJPG4zf0NV7ZT/DVmLuWxc3naB7beb127S9Y49r6/L5i2PbK3VZr5ePKWvFdNa+0Br7f4ZRgQPzzBJzV2T/GtV/fxWdn9Zkk2LXI+avGUWAAB6JfBd3xfG5UFTtt0vSx8Vnetn3fwN42Qrc/1vmLLvoVP2+ekkP5Vk48SzaaePy4OXWNPm/HhcLjpKNyuttR+01j7WWntOkr/IcLvlQ7ayptOT7F5Vd51RmQAAsN0S+K7vLePy/1bV3G2ZqaqbZAglS3VihtcCPKaq7jdv27OS/HSSj7bW5j+/lyTPrKo7Thz7RklemuHn9aaJdu9N8o0kv1tVD51WRFUdWFW3WEK9m5K0DK9Z2Gaq6vCquvmUTXOjmldMrJu75XI5NR03Lt84zlQ6//i3nPLzAACALnmGb57W2qlV9YYkv5PkrKp6d4ZJQB6W4XbB8zLMlrlYP9+vqqMzvE7h1Kp6Z4bJWdZmeBXD+RmfX5vitCRnjJOaXJbhPXxrkqzPMPPm3DF+VFWPyPD+vQ9U1b9nmIXyigyjgffOECxvl+sGqYXq/UyGd/29NcNrKX6c5H2ttS8udr7L8LIME8+ckmGm0h9muCb3z/C6i7dPtD05w+ykb6yqdyX5fpJLW2uv3sx5nFxVz0/yl0m+VlUfzPBS+VtleP7x0AwT8hwxw3MCAIBVSeCb7mlJvpwhkD01w0jTe5L8SZJzM4yqLaq19t6q+qVxvwdnmGzl/AzvrHvJ+OqEaZ6d5NcyvLNu3/H4r0zyosmXro/H+GJVrUnynCS/kuSJGQLpdzPcVnpMkqW+7uC3MoyQHZHkMRleGH9uklkGvr/IcG73SvKAsdZvj+tf0VrbNNewtfbhqnpuhuvw7Ay3fH4ryYKBb9zv/6uq0zK8ouGgDK/IuCzDS97fkORtMzwfAABYtaq1ttI1bDeq6s4ZRr7e3lp7zErXww2vqtbvtPdOB+zy2F1WuhQAALaxTcdtWrzRNrJ27dps2LBhw0KvC1sqz/BNUVW3HZ+bm1x3iySvGL99zw1fFQAAwPK4pXO6Z2WYbOWUDLdG3jbD6wP2SXJShufyAAAAVjWBb7p/yzBJyoOS7JHhpeJfTfKqDM+ZuQ8WAABY9QS+KVprJ2eYIRIAAGC75Rk+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAndp5pQuA7c2afdZk/XHrV7oMAABYlBE+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6tfNKFwDbmzPPPTO7P3v3lS4Dtsqm4zatdAkAwA3ACB8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwLfDqqqjq2qVlXrtqKPVlWnzK6qlTkGAAD0SuDrVFUdNYalo1a6FgAAYGUIfDuuVyfZP8lnV7oQAABg29h5pQtgZbTWLkpy0UrXAQAAbDtG+FaR8TbMd1fVf1XVlVV1eVWdVlWPn9L2lPGWzZtU1Yuq6itVdXVVnTA+8/amsembxnZzX/uO+y/4DF9V7VdVx1fVxrHPC6vqk1X1tAXq3quq3lBV3x3bn1VVT1yg7Y2q6qlV9bmq+n5V/WD89dOqasmfx6p6XlVdO16fPeZte0xVfbyqNlXVVVV1dlW9oKpuutT+AQCgB0b4VpfXJvlSkk8k+W6SPZM8NMk/VNXPtdZeOGWfdye5d5KTkpyY5MIkpyS5NMmRSd6b5IyJ9pduroCq+uUk70xy0yQfSvJPSXZLsibJH441TtotyWlJfpjkXUluluSRSY6vqmtba2+e1/4fkjw2yTlJ/i5JS/JrSV6T5KAkj1ukvhsleUWS30/yniSPba1dNbH975McneTcJP8ynu/9krwkyeFV9cDW2jWbOwYAAPRC4Ftd7tZa+8bkiqq6SYYw9/yqel1r7Tvz9rnjuN9F8/ZLhsB3YmvthKUcvKr2SvK2DJ+L+7fWTp23fZ8pu61J8vdJntJa+/HY7rgkX0zyR0nePLH/YzKEvS8kOaS19v1x/QuSnJrksVX1gdba2xao72ZJ3prkERmeQXxma+3aie1HZQh770nyuNbalRPbjk1yTJLfTfLKJVyL9Qts2m+xfQEAYLVwS+cqMj/sjet+mORvM4Sww6fs9sL5YW8rPCHJrZO8dn7YG2s5d8o+VyR5zlzYG9t9KcOo3/5VtctE26PH5fPnwt7Y/gcZwmGSPGlaYeNtmx/NMBr4/Nba70+GvdEzk1yT5OjJsDd6SZKLs8gIIgAA9MQI3ypSVXfIEHwOT3KHJDef1+Qnp+w2y1k27zcuT1rGPl9rrV0+Zf0543K3JN8bf31Akmsz3HI636lJfpzknlO23SZDgPzpJL/VWnvr/AZVdYsMo40XJXnWOMI539UZZiZdVGtt7bT148jfAUvpAwAAVprAt0pU1U9nCG+7J/lkko8kuSxDCNo3w+jbtElHzp9hGbuNy/m3jW7OQs8Ezj0nt9PEul2TXDKOWl5Ha+2aqrooyd5T+rpthpHHczNcm2l2T1JJfiLDrZsAALDDE/hWj+dkmKTlifOfuRuffXvCtJ1aa22GNcyFt59M8h8z7HfOZUn2qKobt9Z+NLmhqnZOsleSaaOFZ2aY4OWEJJ+oqvu31v5rSt9J8oXWmhE4AACIZ/hWk58dl++esu3QLehv7pm6nTbb6rpOH5cP2YLjLcUXMnzmDpmy7ZAMtW6YtmNr7R+TPDrJ7TOEvrvM2/79JGcluev81zQAAMCOSuBbPTaOy3WTK6vqwVlgIpNFXDwu77CMfd6cYYTtaVV1vVC2wCydy3H8uPzL8Zm7uX5vkeSvxm//fqGdW2vvSvLrGUYCT62qu85r8vIkN8nwSojd5u9fVbtXldE/AAB2GG7pXD1ek+SJSd5ZVe/O8Bzd3ZIckeSfkzxqmf19OsMMms8aR7wuGNf/v9baZdN2aK1dVFWPzfA+vY9X1UkZXq9w6yR3T/JTSe60zDom+39bVR2Z5DeTnFVVJ2Z4D9/Dx37/edqELPP6eP/Yx3uSnFJVD2itnTluO76q1iZ5epJvVNWHk3w7yR5j/4dkeCH9U7f0HAAAYHsi8K0SrbUvVtVhSf4sw8vWd87w7NojMjxbt6zA11rbVFW/nmECkycmueW46R/zv8+7TdvvA1V1r/zvbKEPSrIpyZeT/OVyaljAYzLMyHl0kqeM685O8rJc/6XuC9X44ap6aJL3ZwimD26tfW7c9rtjUH1qkgdkmIjmkgzB76UZzh8AAHYINds5P6BvVbV+p713OmCXx+6yeGNYxTYdt2mlSwAANmPt2rXZsGHDhoVeF7ZUnuEDAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAndp5pQuA7c2afdZk/XHrV7oMAABYlBE+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6tfNKFwDbmzPPPTO7P3v3bdb/puM2bbO+AQDYsRjhAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADp1gwe+qjqqqlpVHXVDH3tbGc/nlGW0Xzfuc+y2q2r5qurWVfWqqtpYVdeMNd5jtdYLAABs3s4rXQCryl8neUqSf03yD0l+nOT8JLutZFEAAMCWWYnA954kpyf57goce7X4bJL9k1y00oXM8ytJvtpae9jkyqrab4XqAQAAtsINHvhaa5clueyGPu5q0lq7IsmXV7qOKW6f5BMrXQQAADAbW/UMX1XtOz7bdUJV7VdVJ1bVJVX1g6r6VFU9aMo+U5/hG58b21hVt6iql1bVt6vq6qr6elX9UVXVAjXcp6reUVXfGdt/t6o+UlW/OdFms8+gzR173rqbVNUzqmpDVW2qqivGdu+tqgcs0M9eVfWGsYarq+qsqnrilHYL1lNVd66qt4zn88OqOm/8/s5T2h479rOuqh5ZVZ8d67ykqt5eVT85rc4p/ZxSVS1JJTl07HNJzyUutd6qesrY55PnrT96XH9FVd103rbPVtVVVXXzeevvW1Xvqqrzx2OeU1Wvr6rbL3Ru48/zRVX1lfFnc8JSrg0AAGzPZjXCd6ckn07yn0len+R2SR6V5KSqemxr7R1L7OfGST6SYaTppCTXJHl4kr9KcrMkL55sPIaH12Z41ux9Sb6WZO8k90ry9CT/vBXndEKSx4zn9JYkV451HZTkiCQfndd+tySnJflhkneN9T4yyfFVdW1r7c2LHbCq7j32u8t4Pl9Ksl+SxyU5sqoOb619fsquT0/yq+M+pya5b4brv6aq7tFau3oJ53pKkmOSfGv8Pkk2zrDek8fl4UneONHN/cflzZMcONaRqto1yQFJPtlau3LimE8c9796POY5Se6c5ElJHlZV92utfXtKue9Ocu8Mn6sTk1y4uXMDAIAezCrwHZLkb1prz5tbUVWvzhACX1dVJ7XWLl9CP7dPcmaSB879I7+qXpzkq0meXVV/0Vr70bj+55O8JsnlSQ5urZ012VFV7bOlJzOGjUcnWZ/kvq21H8/bvueU3dYk+fskT5lrX1XHJflikj9KstnAN45gviXJrZM8vrX21oltj0ry9iT/WFU/31q7dt7uRyS5d2vtPyb2eVuGwHpkFgm+rbUTxn2OSbKxtXbs5tpvSb2tta9X1beT3L+qqrXWxub3T/KxJOsyhMFTxvXrkuw0bpvr9y4Z/kNhY5JDW2vfmdh2/yT/luSVSX5tSsl3THK31tpqe24SAAC2mVm9luGyJH86uWIc2XlrhpGvaf8AX8gzJkd0WmsXJnlvkl2T/NxEu6dlCKwvmR/2xv3OXcYxr7d7htsbr04yP1yltXbxlH2uSPKcyXDYWvtShlG//atql0WO+YsZRsc+PRmexn7ekeRTGc7/oCn7vmoy7I3mRtHus8hxt9SW1PuxJD+R5BeS/wntt8swIrohQ+CbM/frkyfWPS3DKPAzJ8PeeMyPZRjxe9gC1/qFywl7VbV+2td4zgAAsF2YVeDb0Fr73pT1p4zLey6xn8taa1+fsv6ccbn7xLr7jcuTltj3ko2jke/PEGrOGJ/9OqyqbrGZ3b62wCjmXO2LvdrggHH5sQW2z62fdi2n3eY57ZrN0pbUO7duLszN3c558rjt3hNh7f5Jvp9hRtM5B47LQ8fnF6/zleF23p2S3GVKPZ+dsg4AALo2q1s6L1hg/fnjctcl9nPpAuuvGZc7TaybC1DfybbxqAy3Yj42//vs4FVV9a4kf9Bam3/Oy6l9mrlrtNDrKubWTwuO04691ONuqS2pd/I5vuPG5bmtta9W1clJ/jBDmPtckrsm+WBr7ZqJ/edupX1eNu9WU9adP2Xdglpra6etH0f5Dpi2DQAAVptZjfDdZoH1tx2X2+I1DHMhZykzUc7dlrlQwL1eIG2tXdlaO7a1dpckd0jy+Ay3KT4+wy2IszZ3jW67wPbbzWu30pZdb2vtvCRfyRDqbprhOb25EPipDBPePCD/OwI4f/Rwrq9dW2u1ma9T5xcz8cwgAADsMGYV+A5Y4LmpdePyCzM6zqTTx+VDltB207j8qfkbqupns8jtlq21c8bn1B6cYSbQgxaYuGVrzF2jdQtsn1u/YcbH3VJbWu/JGUbgnpbhup+c/M+7CU/PEPYmb/WcNPczP3hLCgYAgB3NrALfrkleNLmiqu6VYXr+y5K8Z0bHmfTaDLctvnCc/OM65s3S+eUMs3keWVV7T7S5eZJXTdn3J6rqvlOOecsMryC4JsNo1CydlmH066CqeuS8eh6ZYSbUr2YYCVsNtrTeuVG7P573/dyv75bhFRMXZ5ixddKrk/woyXHjjJ3XMb5rTxgEAIDRrJ7h+0SSJ40h6bT873v4bpThNQVLeSXDsrTWvlRVT0/yuiRfqKr3Zhh92zPDe/i+l+Swse2PquqVSV44tn1PhnN/YJLzxq9JP5nk9Ko6O8MI1TkZXj/wKxluYXzVApPUbM35tKp6QoZXC7xjPJ8vZ5jp8uHj+fz2lFcyrIitqPfjGW6x3TvJl+fNtnlykmMzzOT5rvm3YbbWvlxVRyc5PslZVfWhDKHyxhluuz04yX/HTJoAAJBkdoHvm0memuEF6U9NctMMQelPW2sfntExrqe19saq+s8kf5DhFsKHJ7kow7vv/m5e82MyvDrhyUl+J8MkHm/PEDC+NK/txrH9ugyhca8kl2QY0Xr+uN/MtdY+M77M/AUZnmV7WIbz+acMr5/4yrY47pbaknpba5dU1RkZJj6Z/4zeZ5L8IMNI6tTZP1tr/1hVZyZ5boafzYPGfc7L8GzlO2ZwagAA0IXamrksqmrfDGHvza21o2ZTEqxeVbV+p713OmCXxy72WsUtt+m4TYs3AgCga2vXrs2GDRs2LDR7/FLN6hk+AAAAVhmBDwAAoFMCHwAAQKe2atKW1trGJDWbUgAAAJglI3wAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0aueVLgC2N2v2WZP1x61f6TIAAGBRRvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAfLdOa5Z650CQAAsCQCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgY8lqaoTqqpV1b7L2GdjVW3cRvXsO9ZzwrboHwAAeiDwkSSpqmPHALVupWuZM9ZzykrXAQAA26udV7oAunb4ShcAAAA7MoGPbaa19o2VrgEAAHZkbumcp6p+tapOrqrvVtXVVXVeVZ1aVU+f1+6U8ZbDm1bVn1XVN8f236iqY6rqJgv0v9/4PNw5Y/sLquptVfVzU9repar+qqo+X1X/Pbb/VlW9oar2mdJ+3VjTsVV1n6r6QFVdstizd+NzdseM3358bN+qqi3Q/ilV9R9VddVY/xuqatdp/c5/hq+qblJVz6iqDVW1qaquGNu9t6oeMLY5auLYh07WU1XHLnQeE8e4RVX9cVWdUVU/qKrvV9Wnq+oxi+0LAAA9McI3oap+J8nrk5yf5P1JLkqyd5K7J3liktdM2e2fk9w7ybuS/CjJkUmOTXKvqvrV1tr/hKaqOiLJvyS58dj/15Psk+QRSX65qg5rrW2Y6PsRSZ6a5ONJ/j3JD5PcNcmTkjysqu7VWvvOlJoOTPLHST6V5Pgke437LuQVSR6e5NAkb06ycTNt/zrJg8f6P5LksCRPTvKzSe6/mf3mnJDkMUn+M8lbklyZ5PZJDkpyRJKPJjkjyYszhNBvjfsepNRsAAAYmklEQVTMOWVznVfVbkk+luSeSTZkOP8bjTW/raru2lp7wRLqBACA7Z7Ad11PyRCM1rTWLpzcUFV7LbDP/knu2lrbNLb7vxkC2q8keXySfxjX757kn5JckeSQ1tqXJvq+a5LPJPm7JAdM9P0PSY5rrV09r5YHJTkpyQuSPG1KTQ9K8tTW2uuXcM5prb1iDEqHJjmhtXbKZprfL8kvtNa+Pdayc4aAdVhV3ae19tmFdhxHAR+dZH2S+7bWfjxv+55jPWckOaOqjkmysbV27FLOY/SKDGHvj1prfz3R982SnJjkT6rqXeMxFlRV6xfYtN8yagEAgBXlls7ruybDSN11tNYuWqD9S+bC3tjuqgyja0ly9ES7306yW5JjJsPeuM9ZSd6Y5J5V9fMT678zP+yN6z+S5KwMo1bTnLHUsLcF/nQu7I21XJPkTeO391lk35akklyd5NrrbWzt4q0pbAyMj0/y+cmwN/Z9VZI/Go//2K05DgAAbC+M8F3XW5O8LMlZVfWOJKcmOa219t+b2efUKes+mSE43nNi3YHjcs0Cz6HdZVzun+RLSVJVleRxSY5KsibJ7kl2mthnods0Fxxlm4HPT1l3zrjcfXM7ttYur6r3J3lYhhG8d2e4Vp9prV0xg9runeH6LPSs343H5f6LddRaWztt/Tjyd8C0bQAAsNoIfBNaay+vqouSPD3JM5I8K0N4ODXJ81pr08LOBVP6+XFVXZzh+b85e47LJy9Sxq0mfv3ysYbvJvlwku9keOYtGULgHRfo4/xFjrE1Lp2y7ppxudOUbfM9KsNI22MzPKeXJFdV1buS/EFr7XrXcxnmrvG9x6+F3Goz2wAAoBsC3zyttbckecv4TNsvJvm1DLdmfriq9p//bF+S2yT59uSKqtopQ/i4fGL1ZeNyTWvti4vVUVV7Zwid/5nkF1tr35u3fXMzTk6dXXM1aK1dmWFSm2Or6qeSHJIhvD4+yb5JDt6K7ueu8XGttedsRT8AANAFz/AtoLV2aWvtg621J2eYJXKPTA8jh05Zd3CGMP2FiXWnT2xbip/O8PP5yJSwt8+4fZbmJlBZyijdTLTWzmmtvTXDs4hfS3LQ3MQto2uXWc9nx322JjQCAEA3BL4JVXXEOOvkfHO3Zk57zuyF4wycc33cLMlfjt++aaLdmzLcDnlMVV1vcpOqulFVrZtYtXFcHjSOGM61u1WGCV5mPTo7N2HKHWbc7/+oqp+oqvtO2XTLJLtkuDV08rnEi5P81FL7H0df35rhlRgvnPazrKqfqao7La9yAADYPrml87renuF5sk9lCFyVYbTo3hleJfDRKfucnWGSl8n38P1Mkg9kfCVDMsxAWVWPTPKeJKdX1ckZZtq8NkPIOjDDbaA3G9ufX1Vvz/AagzOq6iNJdk3ywCRXZXhX3T1meO4fH2v5y6q6W5JNYx1/NsNj/GSGcz87wzvyzkly6wyvsLhtklfNG808Ocmjx4le1mcIhJ9orX1iM8f4vSR3TvKnSX5r/FlekOFdf/tn+Fk+Jsk3Z3heAACwKgl81/X8DLcXHpDkoRmC1bcyTDLy2tba9V7XkOQ3k7www2yat88wscqxSf5q8qXrSdJaO7mq7p7kD8bjHJxhROu8DO+ye/e8vv9Pkv/KMNHJ7yb57yTvS/KiKW23Smvt7Kp6wljb0zMGzySzDHwbM7xMfV2GF7bvleSSJF/JcO3fPq/9MzM8j3h4hp/HjTJM9LJg4BtnAj00ye9kmBjm1zOcywUZbht9dpJ/m9H5AADAqlbzMglLVFWnJDm0tVYrXQs3nKpav9PeOx1wzQXXLN4YAAC20Nq1a7Nhw4YNC70ubKk8wwcAANApgQ8AAKBTAh8AAECnTNqyhVpr61a6BgAAgM0xwgcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD5YpjX7rFnpEgAAYEkEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwwTKdee6ZK10CAAAsicAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgWyWqat+qalV1wkrXshRVtW6s99hteIzt6poAAMBqI/ABAAB0SuADAADolMAHAADQKYFvFaqq/arqxKq6pKp+UFWfqqoHbab9Y6rq41W1qaquqqqzq+oFVXXTzfR/QlWdU1VXV9UFVfW2qvq5Laz3wKr6aFVdVlXfq6oPV9W9prS7fVW9qKpOq6rzq+qHVXXeeOz9l3G8E8Zn++5UVb9XVV8az3tjVf1JVdXY7jeq6rPjNbywql5dVTfbknMEAIDtkcC3+twpyaeT7Jnk9UnemWRtkpOq6lHzG1fV3yd5W5KfTfIvSf42ySVJXpLkQ1W187z2RyTZkORxST6X5JVJTk7yiCSfraoDllnvfZOckuTq8dgnJTk8ySer6uB5bQ9J8vwklyZ5d5Ljkpye5JFJPldVa5Z57L9J8uIkn03yuiTXJvnzJMdU1TOSvDnJ18dt5yf53fGYAACwQ9h58SbcwA5J8jettefNraiqV2cIga+rqpNaa5eP649KcnSS9yR5XGvtyol9jk1yTIaQ88px3e5J/inJFUkOaa19aaL9XZN8JsnfJVlO6Dsiye+31l490deRSU5McnxV/Vxr7dpx08eS3Ka19r3JDsagd1qSv0rykGUce22Su7fWvjP2c2yGgPe88RzXttbOHrfdNMkXkhxdVce01i7cXMdVtX6BTfstoz4AAFhRRvhWn8uS/Onkitba55O8NcluSX5tYtMzk1yT5OjJsDd6SZKLM4zkzfntsY9jJsPeeIyzkrwxyT2r6ueXUe/Xk7xmXl/vTXJqhlHHgyfWXzg/7I3rz8wQBg+rqhsv49gvmQt7Yz+XJnlfklskee1c2Bu3XZ3kHUlukmTJt48CAMD2zAjf6rNhWijKcNvkE5LcM8mbq+oWSdYkuSjJs8bH1ua7OtcNNweOyzULvD/vLuNy/yRfmrJ9mk9OjODNr/fQsd5T51ZW1S8neWqSeyXZK9f/DO6V5LtLPPbnp6w7b1xOG6GbC4f7LNZxa23ttPXjyN9yb3sFAIAVIfCtPhcssP78cbnruNw9SSX5iQy3bi7FnuPyyYu0u9US+0uWXm/G5+pemWRTkn9L8u0Mt162JA/PEGCnTjSzgMumrLtmCduWM4oIAADbLYFv9bnNAutvOy4vm7f8QmttqSNOc/usaa19cUuKm2JJ9Y6Tx7w4QxA8oLV2nVG8qjowAADATHmGb/U5oKp2mbJ+3bj8QpK01r6f5Kwkd62qPZbY9+njcv7smVvjoKqa9jlaNy6/MC73yvD84L9PCXu3itskAQBg5gS+1WfXJC+aXDG+0+5xGUbL3jOx6eUZJiE5vqp2m99RVe0+7zULb8rwSoRjquo+U9rfqKrWLbPeOyd5+rx+jszw/N7Xk3xyXH1hxpkzx4A31/bGGW7z3GuZxwUAABbhls7V5xNJnlRV983wqoLbJXlUhnD+lLlXMiRJa+34qlqbIXB9o6o+nOG5uD0yvM/vkAwh76lj+4ur6pEZQuPpVXVyhlHCa5PcIcOkLnsmWc7LyT+U5GVV9ZAkZ2aYmfMRSa5K8n/mJnRprV1bVa/K8B6+/6iq92YIq4eN9X58/DUAADAjRvhWn28m+cUME5s8NclvZnhR+kNba++Y37i19rtJHpbhPX0PSPKcJL+aYaTwpUleMa/9yUnunuFVCvuOx3hSkrtleDXCo5dZ72cy3L550yS/l+E9eh/L8J6/T8xr+8Ikz01yZZKnZAiGn09ynwxBFQAAmKFqra10DbDdqKr1O+290wHXXHDN4o0BAGALrV27Nhs2bNiw0OvClsoIHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ8AAKBTAh8AAECnBD4AAIBOCXwAAACdEvgAAAA6JfABAAB0SuADAADolMAHAADQKYEPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTABwAA0CmBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANApgQ+Wac0+a1a6BAAAWBKBDwAAoFMCHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHRK4AMAAOiUwAcAANCpaq2tdA2w3aiqi29+85vvsf/++690KQAAdOzss8/OlVdeeUlrbc+t6Ufgg2WoqquT7JTkzJWupRP7jcsvr2gVfXFNZ8v1nC3Xc7Zcz9lyPWfL9dx6+ya5vLV2p63pZOfZ1AI7jP9Mktba2pUupAdVtT5xPWfJNZ0t13O2XM/Zcj1ny/WcLddz9fAMHwAAQKcEPgAAgE4JfAAAAJ0S+AAAADol8AEAAHTKaxkAAAA6ZYQPAACgUwIfAABApwQ+AACATgl8AAAAnRL4AAAAOiXwAQAAdErgAwAA6JTAxw6vqvapquOr6ryqurqqNlbVK6pq92X2s8e438axn/PGfvfZVrWvRrO4nlX1wKp6WVWdXFWXVFWrqk9ty7pXq629nlV1y6p6XFW9raq+XFU/qKrvVdXnq+q5VXWTbX0Oq8mMPp/Pq6oPjvt+v6our6r/qKqX+/2+ZX9+zuvzkKr68fj7/s9mWe9qN6PP5ynjtVvo62bb8hxWm1l+RqvqF6rqLVV1ztjXhVV1alX99raofTWawd9J6xb5fM59/dS2PpcdiRevs0Orqp9J8u9J9k7y3iRfTnKfJIcl+UqSX2qtXbyEfvYc+7lLko8l+VyS/ZIcmeTCJAe21v5rW5zDajLD63lihmt3VZKvJ7lbktNaawdto9JXpVlcz6o6IslJSS5J8vEM13OPJA9Lctux/8Nba1dto9NYNWb4+fx6ku8nOTPJBUlunOSeSQ5NcnmSda21L2yLc1hNZnU95/W5S5IvJtkrya2S/Hlr7QWzrHu1muHn85QMn8UXL9Dkz1pr18yi5tVulp/Rqjoqyd8luSLJvybZmGS3DH8/nddae/SMy191ZvR30r5Jjlpg8y8keUSSs1prd5tJ0Qxaa7587bBfST6cpCX5/XnrXz6uf90S+3n92P7l89Y/Y1z/oZU+1+3seh6Y5K5Jdkqy77jvp1b6/LbH65nkHkkel+Qm89bvkmT92M9zV/pct5frOba/2QLrnzz288GVPtft6XrO2/f4DP858SdjH3+20ue5vV3PJKcM/7xb+XNa6a8ZXtP7JbkmyRlJbjtl+41X+ly3p+u5mf7/aeznGSt9rr19GeFjh1VVP53kGxn+l+5nWmvXTmzbJcl3k1SSvVtrP9hMP7dM8t9Jrk1yu9ba9ya23Wg8xr7jMbod5ZvV9ZzS775JvpkdbIRvW13Pecd4bJK3JvnX1trDtrroVewGup67Jrk0yddba3fe6qJXsW1xPavqyCQnJvmtJDsneVN2kBG+WV7PuRG+1lpts4K3AzO+pp9IcnCSX2it/ec2K3oV29Z/ho53Sn0nw7+lfrK1tmkWdTPwDB87svuPy49M/sGVJGNoOy3JLTL8z97mHJjk5hkCyfcmN4z9fmT89rCtrnh1m9X1ZHBDXM8fjcsd4fauG+J6zoXmL25FH9uLmV7Pqto7yRuTnNha+8dZFrqdmPnns6oeVVXPr6rnVNVDquqmsyt3uzCTazo+l3twks8nOauqDquqPxifgT58/I/dHcG2/jP0qCQ3TfJOYW/2dpQPKUzzc+Pyqwts/9q4vMsN1M/2znWYrRvieh49Lj+0FX1sL2Z+PavqSVV1bFX9TVV9OMmbk3wryfO3vMztxqyv5xsy/JvkqVtT1HZsW/x+f3uSv0zysiQfTPLtqnrklpW3XZrVNb33RPuPjV8vTfI3ST6a5Iyq+tmtqHN7sa3/TnrSuHz9Fu7PZgh87Mh2HZeXLbB9bv1uN1A/2zvXYba26fWsqt9LckSGZ1KO35I+tjPb4no+KckxSZ6b5EEZnol8QGvta5vdqw8zu55VdXSGSZqe3lq7YAa1bY9m+fl8b4bR5n0y3H2yX4bgt1uSd1TVQ7aizu3JrK7p3uPyN5Psn2FSkV2T/GySf8gw0cgHqv8Zj7fZ30lVdWiGz+lZrbV/34LaWITABwube/5hax90nVU/2zvXYba2+HpW1SOSvCLJ+Ul+vbX2o0V22REs+3q21u43Pie1V4bAlyTrx5lRd3RLup7jM7qvyHAb1z9v45q2Z0v+fLbWjmut/Wtr7Tuttataa19p7f9v715CrarCAI7/P4sg7aUZDUyDonIiRIMCLcNBQREkSLNAB1bQw0mzmhREgwZRRJOIiNskKioHERVlEphIET2gFALLgWApVpI9hNXg26d7OZ3ro7v3PXev8//BZuM952zX+ljn8e299vrKI+SJiUXAk102tEdON6ZnzdhvLaW8VUr5tZTyPbCZnOp5NbCpm2b2xly+4+9t9l7d64gJnybZ4GzUhbM8fsHQ87o+Tt8Zh3Z1Es+I2EhO9TpElg+odiGhIZ2Nz1LK4VLKB2TSdxyYiohzz7yJvdJWPF8iY3Z/G43qsfn4/HyRvF/32maRjdq1FdPB/WR/klNj/1Vy5cPtzT+vP9MG9kxX30nLyGT5OHnFVB0w4dMk29vsZ5tvPlhlb7b56m0fp++MQ7taj2dE3AW8TtaOu7mUsvcUL6lJ5+OzlHIU+BS4hCwrUrO24nkdOWXup5lFl8kVOgEebf729tyau+DNx/j8AxgsLLbk/x6nR9r+jv9teLGSxiAhrP0kT1djdDO5WMtrzWeoOnD2uBsgjdGOZn9rRCwascTwOvKM0+5THGd387x1EXH+iLIMg6leO0a9uCJtxVOp1Xg2JRimyGWvN0zQlb2B+RqfK5p97SufthXPKXJlv2FXAevJe0w/B2ovZN/5+IyIa4ClZNL38xza2hdtxfQrMl7LI+LSEfeZDgqE7597kxe0rsboPc3+hbk3UbPxCp8mVjP//n2yRt4DQw8/Tp4BnZpZTyYiVkfE6qHjHCOnISwBHhs6zoPN8d+r/Qd2W/FUajOeEbGZHKM/AutrH4ujtBXPiLi8qUf1HxFxH7mi3wHg6/Zav/C0+Pm5rZSydXhj+grfO83fnu+sMwtAi+PziohYMfR6ImI50zF9tZRS+wmJNsfoCabvLXtqZhmGiFhDlhM4AbzRchcWlC6+4yPiJnIhnG9crKVbFl7XRIuIK4Fd5JSi7cC3wA1kzbx9wNpSyuEZzy8AwwVtm4Khu8ipDh8Be8gPsTvJe6XWNh+WVWsxnjcyvUTzeeT8/kPAu4PnlFK2dNWPhaKNeEbEBnLp8EXk/VIHRvxXR0spz3TUjQWjpXhuBN5sjrOPnB57MVl7ag1wDLijlLJzHro0Vm2932c59hYmqPA6tDY+t5D36u0ki2QfAVYBt5P3Xn0G3DIpU+da/E5aDHxIvs+/AD4mp25vIqdyPlxKebrj7oxd2+/5iHgFuBvYVkp5rtvWT7hSipvbRG/ASvKHxUHgL7KO1rPAshHPLTT3aY94bFnzuh+a4xwkf2BfNu4+9i2e5BnTcrJt3P3sSzxPJ5bA/nH3s0fxXEXWNdtDJnt/k1PkviTrcq0cdx/7FM+THHcwbp8Ydx/7FE/ypMPL5BXmw834PAJ8AjwEnDPuPvYtpjMeW0zO4vmOXMDlF/Jk2m3j7mNP47mUnAL6O3DRuPtV++YVPkmSJEmqlPfwSZIkSVKlTPgkSZIkqVImfJIkSZJUKRM+SZIkSaqUCZ8kSZIkVcqET5IkSZIqZcInSZIkSZUy4ZMkSZKkSpnwSZIkSVKlTPgkSZIkqVImfJIkSZJUKRM+SZIkSaqUCZ8kSZIkVcqET5IkSZIqZcInSZIkSZUy4ZMkSZKkSpnwSZIkSVKl/gHPKOJhKq0SIAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe7690684e0>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 279,
       "width": 446
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# TODO: Display an image along with the top 5 classes\n",
    "model = model_verify #using the restored one \n",
    "file_path = 'flowers/test/10/image_07090.jpg' #an example from test set\n",
    "\n",
    "img = process_image (file_path)\n",
    "imshow (img)\n",
    "plt.show()\n",
    "probs, classes = predict (file_path, model, 5)\n",
    "\n",
    "#print (probs)\n",
    "#print (classes)\n",
    "\n",
    "#preparing class_names using mapping with cat_to_name\n",
    "\n",
    "class_names = [cat_to_name [item] for item in classes]\n",
    "\n",
    "#fig, (ax2) = plt.subplots(figsize=(6,9), ncols=2)\n",
    "plt.figure(figsize = (6,10))\n",
    "plt.subplot(2,1,2)\n",
    "#ax2.barh(class_names, probs)\n",
    "#ax2.set_aspect(0.1)\n",
    "#ax2.set_yticks(classes)\n",
    "#ax2.set_title('Flower Class Probability')\n",
    "#ax2.set_xlim(0, 1.1)\n",
    "\n",
    "sns.barplot(x=probs, y=class_names, color= 'green');\n",
    "\n",
    "#width = 1/5\n",
    "#plt.subplot(2,1,2)\n",
    "#plt.bar (classes, probs, width, color = 'blue')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}