Deep Learning with Pytorch - A complete Beginner Tutorial.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Created by Berkay Alan**\n",
    "\n",
    "**Deep Learning with Pytorch - A complete Tutorial**\n",
    "\n",
    "**2 June 2021**\n",
    "\n",
    "**For more Tutorial:** https://github.com/berkayalan"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Content\n",
    "\n",
    "- What is Pytorch?\n",
    "- Importing Libraries\n",
    "- Basics of Pytorch\n",
    "- Tensors\n",
    "- Math Operations\n",
    "- Common Funtions\n",
    "- Variables - Autograd\n",
    "- Datasets & DataLoaders\n",
    "- Common Modules: Optim - nn\n",
    "- Extra - Useful Resources\n",
    "\n",
    "**Soon**\n",
    "\n",
    "- Linear Regression with Pytorch\n",
    "- Logistic Regression with Pytorch\n",
    "- Artificial Neural Networks(ANN) with Pytorch\n",
    "- Recurrent Neural Networks(RNN) with Pytorch\n",
    "- Long Short Term Memory(LSTM) with Pytorch"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Resources**\n",
    "\n",
    "- [Pytorch Wikipedia](https://en.wikipedia.org/wiki/PyTorch)\n",
    "- [Pytorch Official Tutorial Page](https://pytorch.org/tutorials/beginner/basics/intro.html)\n",
    "- [A Beginner-Friendly Guide to PyTorch and How it Works from Scratch by Analytics Vidhya](https://www.analyticsvidhya.com/blog/2019/09/introduction-to-pytorch-from-scratch/)\n",
    "- [PyTorch Tutorial: How to Develop Deep Learning Models with Python by Machine Learning Mastery](https://machinelearningmastery.com/pytorch-tutorial-develop-deep-learning-models/)\n",
    "- [PyTorch - Variables, functionals and Autograd](https://jhui.github.io/2018/02/09/PyTorch-Variables-functionals-and-Autograd/)\n",
    "- [A Gentle Introduction To Torch.Autograd](https://pytorch.org/tutorials/beginner/blitz/autograd_tutorial.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What is Pytorch?"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pytorch is an open source machine learning framework that accelerates the path from research prototyping to production deployment.\n",
    "\n",
    "It is used for applications such as computer vision and natural language processing, primarily developed by Facebook's AI Research lab (FAIR).It is free and open-source software released under the Modified BSD license. Although the Python interface is more polished and the primary focus of development, PyTorch also has a C++ interface.\n",
    "\n",
    "A number of pieces of deep learning software are built on top of PyTorch, including Tesla Autopilot, Uber's Pyro, HuggingFace's Transformers, PyTorch Lightning,and Catalyst.\n",
    "\n",
    "**PyTorch provides two high-level features**:\n",
    "\n",
    "- Tensor computing (like NumPy) with strong acceleration via graphics processing units (GPU)\n",
    "\n",
    "- Deep neural networks built on a type-based automatic differentiation system\n",
    "\n",
    "Official Website: https://pytorch.org/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Importing Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torchvision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch.autograd import Variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch.utils.data import Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision.transforms import ToTensor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision import datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch.utils.data import DataLoader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch import optim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tensors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " *Numpy* provides an n-dimensional array object, and many functions for manipulating these arrays. Numpy is a generic framework for scientific computing; it does not know anything about computation graphs, or deep learning, or gradients. Let's look at the classical numpy array:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "array = [[4,7,9],[11,2,3]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "list"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = np.array(array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 4,  7,  9],\n",
       "       [11,  2,  3]])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 3)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "numpy.ndarray"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(a)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In Pytorch, matrices(arrays) are called as **Tensors**. Tensors are multidimensional arrays. These tensors can be used on a GPU as well (this is not the case with NumPy arrays). This is a major advantage of using tensors.\n",
    "\n",
    "PyTorch supports multiple types of tensors, including:\n",
    "\n",
    "- FloatTensor: 32-bit float\n",
    " \n",
    "- DoubleTensor: 64-bit float\n",
    " \n",
    "- HalfTensor: 16-bit float\n",
    " \n",
    "- IntTensor: 32-bit int\n",
    " \n",
    "- LongTensor: 64-bit int"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)\n",
    "\n",
    "                                            This image is taken by KDNuggets."
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)\n",
    "\n",
    "                                            This image is taken by Dev Community."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tensors can be created from Python lists with the **torch.tensor()** function."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's create a tensor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "l = [[11, 32, 43], [44, 51, 96]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "tns = torch.tensor(l)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[11, 32, 43],\n",
       "        [44, 51, 96]])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Tensor"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(tns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([2, 3])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([2, 3])"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns.size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor(6)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We can also initialize with scalars\n",
    "torch.tensor(6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to get a tensor filled with the scalar value 1, we can use **torch.ones()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[1., 1., 1.],\n",
       "        [1., 1., 1.]])"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.ones(2, 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([1., 1., 1., 1., 1.])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.ones(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to get tensor filled with the scalar value 0, we can use **torch.zeros()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0., 0., 0.],\n",
       "        [0., 0., 0.]])"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.zeros(2, 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([0., 0., 0., 0., 0., 0.])"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.zeros(6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to get a tensor filled with random numbers from a uniform distribution on the interval [0, 1), we can use **torch.rand()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0.9887, 0.1369, 0.4692],\n",
       "        [0.9278, 0.1875, 0.5616]])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.rand(2,3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([0.4784, 0.8791, 0.7710, 0.8271, 0.6942, 0.6638, 0.8837])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.rand(7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to create a Tensor from a numpy.ndarray we can use **torch.from_numpy**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = np.array([6,7,9])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "t = torch.from_numpy(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([6, 7, 9])"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Tensor"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(t)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Converting a torch Tensor to a numpy array and vice versa is a breeze. We can use **numpy()** for that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = torch.ones(12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "b = a.numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.], dtype=float32)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "numpy.ndarray"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(b)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Math Operations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Addition, subtraction, multiplication and division are common operations that we do while working with tensors. They are so similar to numpy and intuitive."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "x = torch.tensor([1, 2, 3])\n",
    "y = torch.tensor([4, 5, 6])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([1, 2, 3])\n",
      "tensor([4, 5, 6])\n"
     ]
    }
   ],
   "source": [
    "print(x)\n",
    "print(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Addition: tensor([5, 7, 9])\n",
      "Subtraction: tensor([-3, -3, -3])\n",
      "Multipication: tensor([ 4, 10, 18])\n",
      "Division: tensor([0.2500, 0.4000, 0.5000])\n"
     ]
    }
   ],
   "source": [
    "print(\"Addition:\", x+y)\n",
    "print(\"Subtraction:\", x-y)\n",
    "print(\"Multipication:\", x*y)\n",
    "print(\"Division:\", x/y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can also use their own functions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Addition: tensor([5, 7, 9])\n",
      "Subtraction: tensor([-3, -3, -3])\n",
      "Multipication: tensor([ 4, 10, 18])\n",
      "Division: tensor([0.2500, 0.4000, 0.5000])\n"
     ]
    }
   ],
   "source": [
    "print(\"Addition:\",torch.add(x,y))\n",
    "print(\"Subtraction:\", torch.sub(x,y))\n",
    "print(\"Multipication:\", torch.mul(x,y))\n",
    "print(\"Division:\", torch.div(x,y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can get transpoze of a matrix with **t()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "tns = torch.tensor([[2,5,7,9],[91,21,34,56]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 2,  5,  7,  9],\n",
       "        [91, 21, 34, 56]])"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 2, 91],\n",
       "        [ 5, 21],\n",
       "        [ 7, 34],\n",
       "        [ 9, 56]])"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns.t()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can also get statistical information from tensors."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In order to get mean of a tensor, we can use **mean()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "tns = torch.Tensor([91,21,34,56])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([91., 21., 34., 56.])"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor(50.5000)"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns.mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In order to get median of a tensor, we can use **median()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor(34.)"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns.median()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In order to get mode of a tensor, we can use **mode()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.return_types.mode(\n",
       "values=tensor(21.),\n",
       "indices=tensor(1))"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns.mode()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In order to get standart deviation of a tensor, we can use **std()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor(30.6213)"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tns.std()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Common Funtions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Reshaping**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "PyTorch allows a tensor to be a **View** of an existing tensor. View tensor shares the same underlying data with its base tensor. Supporting View avoids explicit data copy, thus allows us to do fast and memory efficient reshaping, slicing and element-wise operations.**view()** can be used for that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "t = torch.rand(8, 4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0.4233, 0.4872, 0.5597, 0.3469],\n",
       "        [0.8915, 0.6923, 0.6690, 0.3504],\n",
       "        [0.1325, 0.9195, 0.6741, 0.4909],\n",
       "        [0.6860, 0.9977, 0.6453, 0.2175],\n",
       "        [0.6688, 0.9905, 0.8088, 0.3846],\n",
       "        [0.9670, 0.2966, 0.4768, 0.5501],\n",
       "        [0.1592, 0.5706, 0.7089, 0.2012],\n",
       "        [0.0827, 0.1797, 0.7783, 0.8054]])"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0.4233, 0.4872, 0.5597, 0.3469, 0.8915, 0.6923, 0.6690, 0.3504, 0.1325,\n",
       "         0.9195, 0.6741, 0.4909, 0.6860, 0.9977, 0.6453, 0.2175],\n",
       "        [0.6688, 0.9905, 0.8088, 0.3846, 0.9670, 0.2966, 0.4768, 0.5501, 0.1592,\n",
       "         0.5706, 0.7089, 0.2012, 0.0827, 0.1797, 0.7783, 0.8054]])"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.view(2,16)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Concatinating**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    " Concatenates the given sequence of seq tensors in the given dimension. All tensors must either have the same shape (except in the concatenating dimension) or be empty.**torch.cat()** can be used to cancatinate different tensors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "x = torch.randn(2, 3)\n",
    "y = torch.randn(2, 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 1.5393, -0.7032,  0.3112],\n",
       "        [-2.6722, -0.4254, -0.4386]])"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[-0.2921,  1.2708,  1.6733],\n",
       "        [ 0.4800,  0.2355, -0.7317]])"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "If we want to concat vertically, we need to use additional brackets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 1.5393, -0.7032,  0.3112],\n",
       "        [-2.6722, -0.4254, -0.4386],\n",
       "        [-0.2921,  1.2708,  1.6733],\n",
       "        [ 0.4800,  0.2355, -0.7317]])"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.cat((x,y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can concatenate the tensors horizontally as well by setting the *dim* parameter to 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 1.5393, -0.7032,  0.3112, -0.2921,  1.2708,  1.6733],\n",
       "        [-2.6722, -0.4254, -0.4386,  0.4800,  0.2355, -0.7317]])"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.cat((x,y),dim=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Squeezing**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "It returns a tensor with all the dimensions of input of size 1 removed.\n",
    "\n",
    "For example, if input is of shape: (A x 1 x B x C x 1 x D) then the out tensor will be of shape: (A x B x C x D)\n",
    "\n",
    "When dim is given, a squeeze operation is done only in the given dimension. If input is of shape: (A×1×B) , squeeze(input, 0) leaves the tensor unchanged, but squeeze(input, 1) will squeeze the tensor to the shape (A×B) .\n",
    "\n",
    "**Note:** The returned tensor shares the storage with the input tensor, so changing the contents of one will change the contents of the other.\n",
    "\n",
    "In order to do that, we will use **squeeze()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "x = torch.ones(3,1,3,1,3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[[[[1., 1., 1.]],\n",
       "\n",
       "          [[1., 1., 1.]],\n",
       "\n",
       "          [[1., 1., 1.]]]],\n",
       "\n",
       "\n",
       "\n",
       "        [[[[1., 1., 1.]],\n",
       "\n",
       "          [[1., 1., 1.]],\n",
       "\n",
       "          [[1., 1., 1.]]]],\n",
       "\n",
       "\n",
       "\n",
       "        [[[[1., 1., 1.]],\n",
       "\n",
       "          [[1., 1., 1.]],\n",
       "\n",
       "          [[1., 1., 1.]]]]])"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([3, 1, 3, 1, 3])"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "y = torch.squeeze(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[[1., 1., 1.],\n",
       "         [1., 1., 1.],\n",
       "         [1., 1., 1.]],\n",
       "\n",
       "        [[1., 1., 1.],\n",
       "         [1., 1., 1.],\n",
       "         [1., 1., 1.]],\n",
       "\n",
       "        [[1., 1., 1.],\n",
       "         [1., 1., 1.],\n",
       "         [1., 1., 1.]]])"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([3, 3, 3])"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Variables - Autograd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "When training neural networks, the most frequently used algorithm is back propagation. In this algorithm, parameters (model weights) are adjusted according to the gradient of the loss function with respect to the given parameter. **torch.autograd** is PyTorch’s automatic differentiation engine that powers neural network training. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Background**\n",
    "\n",
    "Neural networks (NNs) are a collection of nested functions that are executed on some input data. These functions are defined by parameters (consisting of weights and biases), which in PyTorch are stored in tensors.\n",
    "\n",
    "Training a NN happens in two steps:\n",
    "\n",
    "*Forward Propagation*: In forward prop, the NN makes its best guess about the correct output. It runs the input data through each of its functions to make this guess.\n",
    "\n",
    "*Backward Propagation*: In backprop, the NN adjusts its parameters proportionate to the error in its guess. It does this by traversing backwards from the output, collecting the derivatives of the error with respect to the parameters of the functions (gradients), and optimizing the parameters using gradient descent. \n",
    "\n",
    "For a more detailed walkthrough of backprop, check out [this video](https://www.youtube.com/watch?v=IN2XmBhILt4&ab_channel=StatQuestwithJoshStarmer) from Statquest."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Difference between variables and tensors is variable accumulates gradients. In order to make back prob, we need variables.\n",
    "\n",
    "We can make math operations with variables too."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can define a variable with **variable()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([1., 1., 1., 1., 1.], requires_grad=True)"
      ]
     },
     "execution_count": 196,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x= Variable(torch.ones(5), requires_grad = True)\n",
    "x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Specifying requires_grad as True will make sure that the gradients are stored for this particular tensor whenever we perform some operation on it. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Tensor"
      ]
     },
     "execution_count": 197,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Let's make an example. \n",
    "\n",
    "Assume we have an equation y = x^3 and we define x = [6,12] variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "x = Variable(torch.Tensor([6,12]), requires_grad = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([ 6., 12.], requires_grad=True)"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "After calculation, we find that y = [216,1728] (y = x^3.)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor(216., grad_fn=<PowBackward0>)"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x[0]**3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor(1728., grad_fn=<PowBackward0>)"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x[1]**3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Recap o equation is that o = (1/2)sum(y) = (1/2)sum(x^3) and deriavative of o = 3x2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "y = x**3\n",
    "\n",
    "o = (1/2)*sum(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor(972., grad_fn=<MulBackward0>)"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "o"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "So gradients are [54,216]."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gradients:  tensor([ 54., 216.])\n"
     ]
    }
   ],
   "source": [
    "o.backward() # calculates gradients\n",
    "\n",
    "print(\"gradients: \",x.grad)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Datasets & DataLoaders"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    " Code for processing data samples can get messy and hard to maintain; we ideally want our dataset code to be decoupled from our model training code for better readability and modularity. PyTorch provides two data primitives: **torch.utils.data.DataLoader** and **torch.utils.data.Dataset** that allow us to use pre-loaded datasets as well as our own data. Dataset stores the samples and their corresponding labels, and DataLoader wraps an iterable around the Dataset to enable easy access to the samples."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    " PyTorch domain libraries provide a number of pre-loaded datasets (such as MNIST) that subclass torch.utils.data.Dataset and implement functions specific to the particular data. They can be used to prototype and benchmark our model. We can find them here: [Image Datasets](https://pytorch.org/vision/stable/datasets.html), [Text Datasets](https://pytorch.org/text/stable/datasets.html) and [Audio Datasets](https://pytorch.org/tutorials/beginner/basics/data_tutorial.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Loading a Dataset**\n",
    "\n",
    "Here is an example of how to load the Fashion-MNIST dataset from TorchVision. Fashion-MNIST is a dataset of Zalando’s article images consisting of of 60,000 training examples and 10,000 test examples. Each example comprises a 28×28 grayscale image and an associated label from one of 10 classes.\n",
    "\n",
    "We load the FashionMNIST Dataset with the following parameters:\n",
    "\n",
    "- root is the path where the train/test data is stored,\n",
    "\n",
    "- train specifies training or test dataset,\n",
    "\n",
    "- download=True downloads the data from the internet if it’s not available at root.\n",
    "\n",
    "- transform and target_transform specify the feature and label transformations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz to data/FashionMNIST/raw/train-images-idx3-ubyte.gz\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0814eee83e6644e0a20b361ec360445d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data/FashionMNIST/raw/train-images-idx3-ubyte.gz\n",
      "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz to data/FashionMNIST/raw/train-labels-idx1-ubyte.gz\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e8ded4a31b3420b939bb78f07eddddc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data/FashionMNIST/raw/train-labels-idx1-ubyte.gz\n",
      "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz to data/FashionMNIST/raw/t10k-images-idx3-ubyte.gz\n",
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a5c92aec3ce547e5a0e2a768e5556435",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data/FashionMNIST/raw/t10k-images-idx3-ubyte.gz\n",
      "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz to data/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0928f2f21b604111a50e7829299b8d62",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz\n",
      "Processing...\n",
      "Done!\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.8/site-packages/torchvision/datasets/mnist.py:335: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at  /Users/distiller/project/conda/conda-bld/pytorch_1616554845587/work/torch/csrc/utils/tensor_numpy.cpp:143.)\n",
      "  return torch.from_numpy(parsed).view(length, num_rows, num_cols)\n"
     ]
    }
   ],
   "source": [
    "training_data = datasets.FashionMNIST(\n",
    "    root=\"data\",\n",
    "    train=True,\n",
    "    download=True,\n",
    "    transform=ToTensor()\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "test_data = datasets.FashionMNIST(\n",
    "    root=\"data\",\n",
    "    train=False,\n",
    "    download=True,\n",
    "    transform=ToTensor())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dataset FashionMNIST\n",
       "    Number of datapoints: 60000\n",
       "    Split: train\n",
       "    Root Location: data\n",
       "    Transforms (if any): ToTensor()\n",
       "    Target Transforms (if any): None"
      ]
     },
     "execution_count": 213,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dataset FashionMNIST\n",
       "    Number of datapoints: 10000\n",
       "    Split: test\n",
       "    Root Location: data\n",
       "    Transforms (if any): ToTensor()\n",
       "    Target Transforms (if any): None"
      ]
     },
     "execution_count": 214,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can index Datasets manually like a list: *training_data[index]*. We use matplotlib to visualize some samples in our training data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "labels_map = {\n",
    "    0: \"T-Shirt\",\n",
    "    1: \"Trouser\",\n",
    "    2: \"Pullover\",\n",
    "    3: \"Dress\",\n",
    "    4: \"Coat\",\n",
    "    5: \"Sandal\",\n",
    "    6: \"Shirt\",\n",
    "    7: \"Sneaker\",\n",
    "    8: \"Bag\",\n",
    "    9: \"Ankle Boot\",\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure = plt.figure(figsize=(10, 10))\n",
    "cols, rows = 4, 4\n",
    "\n",
    "for i in range(1, cols * rows + 1):\n",
    "    sample_idx = torch.randint(len(training_data), size=(1,)).item()\n",
    "    img, label = training_data[sample_idx]\n",
    "    figure.add_subplot(rows, cols, i)\n",
    "    plt.title(labels_map[label])\n",
    "    plt.axis(\"off\")\n",
    "    plt.imshow(img.squeeze(), cmap=\"gray\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Creating a Custom Dataset** \n",
    " \n",
    " A custom Dataset class must implement three functions: __init__, __len__, and __getitem__. Take a look at this implementation; the FashionMNIST images are stored in a directory img_dir, and their labels are stored separately in a CSV file annotations_file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "class CustomImageDataset(Dataset):\n",
    "    def __init__(self, annotations_file, img_dir, transform=None, target_transform=None):\n",
    "        self.img_labels = pd.read_csv(annotations_file)\n",
    "        self.img_dir = img_dir\n",
    "        self.transform = transform\n",
    "        self.target_transform = target_transform\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.img_labels)\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        img_path = os.path.join(self.img_dir, self.img_labels.iloc[idx, 0])\n",
    "        image = read_image(img_path)\n",
    "        label = self.img_labels.iloc[idx, 1]\n",
    "        if self.transform:\n",
    "            image = self.transform(image)\n",
    "        if self.target_transform:\n",
    "            label = self.target_transform(label)\n",
    "        return image, label"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "- __init__(): The __init__ function is run once when instantiating the Dataset object. We initialize the directory containing the images, the annotations file, and both transforms\n",
    "\n",
    "- __len__(): The __len__ function returns the number of samples in our dataset.\n",
    "\n",
    "- __getitem__(): The __getitem__ function loads and returns a sample from the dataset at the given index idx. Based on the index, it identifies the image’s location on disk, converts that to a tensor using read_image, retrieves the corresponding label from the csv data in self.img_labels, calls the transform functions on them (if applicable), and returns the tensor image and corresponding label in a tuple."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Preparing Our data for training with DataLoaders**\n",
    "\n",
    "The Dataset retrieves our dataset’s features and labels one sample at a time. While training a model, we typically want to pass samples in “minibatches”, reshuffle the data at every epoch to reduce model overfitting, and use Python’s multiprocessing to speed up data retrieval.\n",
    "\n",
    "DataLoader is an iterable that abstracts this complexity for us in an easy API."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "train_dataloader = DataLoader(training_data, batch_size=64, shuffle=True)\n",
    "test_dataloader = DataLoader(test_data, batch_size=64, shuffle=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    " We have loaded that dataset into the Dataloader and can iterate through the dataset as needed. Each iteration below returns a batch of train_features and train_labels (containing batch_size=64 features and labels respectively). Because we specified shuffle=True, after we iterate over all batches the data is shuffled."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature batch shape: torch.Size([64, 1, 28, 28])\n",
      "Labels batch shape: torch.Size([64])\n"
     ]
    }
   ],
   "source": [
    "train_features, train_labels = next(iter(train_dataloader))\n",
    "print(f\"Feature batch shape: {train_features.size()}\")\n",
    "print(f\"Labels batch shape: {train_labels.size()}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Label: 4\n"
     ]
    }
   ],
   "source": [
    "img = train_features[0].squeeze()\n",
    "label = train_labels[0]\n",
    "plt.imshow(img, cmap=\"gray\")\n",
    "plt.show()\n",
    "print(f\"Label: {label}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Common Modules: Optim - nn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Optim Module"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "The Optim module in PyTorch has pre-written codes for most of the optimizers that are used while building a neural network. We just have to import them and then they can be used to build models."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "To construct an Optimizer we have to give it an iterable containing the parameters (all should be Variable s) to optimize. Then, we can specify optimizer-specific options such as the learning rate, weight decay, etc."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**SGD Optimizer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Adam Optimizer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "optimizer = optim.Adam([var1, var2], lr=0.0001)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Above are the examples to get the ADAM and SGD optimizers. Most of the commonly used optimizers are supported in PyTorch and hence we do not have to write them from scratch. Some of them are:\n",
    "\n",
    "- SGD\n",
    "\n",
    "- Adam\n",
    "\n",
    "- Adadelta\n",
    "\n",
    "- Adagrad\n",
    "\n",
    "- AdamW\n",
    "\n",
    "- SparseAdam\n",
    "\n",
    "- Adamax\n",
    "\n",
    "- ASGD (Averaged Stochastic Gradient Descent)\n",
    "\n",
    "- RMSprop\n",
    "\n",
    "- Rprop (resilient backpropagation)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NN(Neural Network) Module"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "NN Module defines a set of functions, similar to the layers of a neural network, which takes the input from the previous state and produces an output."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's make an example with Optim."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size, input_dimension, hidden_dimension, output_dimension = 128, 1500, 150, 20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create random Tensors to hold inputs and outputs\n",
    "x = torch.randn(batch_size, input_dimension)\n",
    "y = torch.randn(batch_size, output_dimension)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use the nn package to define our model and loss function.\n",
    "model = torch.nn.Sequential(\n",
    "    torch.nn.Linear(input_dimension, hidden_dimension),\n",
    "    torch.nn.Sigmoid(),\n",
    "    torch.nn.Linear(hidden_dimension, output_dimension),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss_fn = torch.nn.MSELoss(reduction='sum')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use the optim package to define an Optimizer that will update the weights of the model for us. \n",
    "# Here we will use SGD. The first argument to the SGD constructor tells the optimizer which Tensors it should update.\n",
    "optimizer = torch.optim.SGD(model.parameters(), lr=1e-4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99 346.8717041015625\n",
      "199 23.542585372924805\n",
      "299 2.4880869388580322\n",
      "399 0.4464450180530548\n",
      "499 0.10838571190834045\n"
     ]
    }
   ],
   "source": [
    "for t in range(500):\n",
    "    # Forward pass: compute predicted y by passing x to the model.\n",
    "    y_pred = model(x)\n",
    "\n",
    "    # Compute and print loss.\n",
    "    loss = loss_fn(y_pred, y)\n",
    "    if t % 100 == 99:\n",
    "        print(t, loss.item())\n",
    "\n",
    "    # Before the backward pass, use the optimizer object to zero all of the gradients for the variables \n",
    "    # it will update (which are the learnable weights of the model). This is because by default, gradients are\n",
    "    # accumulated in buffers( i.e, not overwritten) whenever .backward() is called. \n",
    "    optimizer.zero_grad()\n",
    "\n",
    "    # Backward pass: compute gradient of the loss with respect to model parameters\n",
    "    loss.backward()\n",
    "\n",
    "    # Calling the step function on an Optimizer makes an update to its parameters\n",
    "    optimizer.step()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extra - Useful Resources"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- [Scalars, Vectors, Matrices and Tensors with Tensorflow 2.0](https://dev.to/mmithrakumar/scalars-vectors-matrices-and-tensors-with-tensorflow-2-0-1f66)\n",
    "\n",
    "- [Introduction to PyTorch for Deep Learning](https://www.kdnuggets.com/2018/11/introduction-pytorch-deep-learning.html)\n",
    "\n",
    "- [Common Torch Functions](https://pytorch.org/docs/stable/torch.html)\n",
    "\n",
    "- [Introduction to Pytorch (a very gentle start)](https://www.kaggle.com/frtgnn/introduction-to-pytorch-a-very-gentle-start)\n",
    "\n",
    "- [Neural Networks from Statquest](https://www.youtube.com/watch?v=CqOfi41LfDw&ab_channel=StatQuestwithJoshStarmerStatQuestwithJoshStarmerDo%C4%9Fruland%C4%B1)\n",
    "\n",
    "- [Fast AI Pytorch Tutorial](https://docs.fast.ai/)\n",
    "\n",
    "- [Intro to Deep Learning with Pytorch by Udacity](https://www.udacity.com/course/deep-learning-pytorch--ud188?cjevent=becd1b75759d11ea83f301a10a24060d)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}