assignment1/two_layer_net.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pYtLbKEm8hzG",
    "tags": [
     "pdf-ignore"
    ]
   },
   "source": [
    "# Fully-Connected Neural Nets\n",
    "In this exercise we will implement fully-connected networks using a modular approach. For each layer we will implement a `forward` and a `backward` function. The `forward` function will receive inputs, weights, and other parameters and will return both an output and a `cache` object storing data needed for the backward pass, like this:\n",
    "\n",
    "```python\n",
    "def layer_forward(x, w):\n",
    "  \"\"\" Receive inputs x and weights w \"\"\"\n",
    "  # Do some computations ...\n",
    "  z = # ... some intermediate value\n",
    "  # Do some more computations ...\n",
    "  out = # the output\n",
    "   \n",
    "  cache = (x, w, z, out) # Values we need to compute gradients\n",
    "   \n",
    "  return out, cache\n",
    "```\n",
    "\n",
    "The backward pass will receive upstream derivatives and the `cache` object, and will return gradients with respect to the inputs and weights, like this:\n",
    "\n",
    "```python\n",
    "def layer_backward(dout, cache):\n",
    "  \"\"\"\n",
    "  Receive dout (derivative of loss with respect to outputs) and cache,\n",
    "  and compute derivative with respect to inputs.\n",
    "  \"\"\"\n",
    "  # Unpack cache values\n",
    "  x, w, z, out = cache\n",
    "  \n",
    "  # Use values in cache to compute derivatives\n",
    "  dx = # Derivative of loss with respect to x\n",
    "  dw = # Derivative of loss with respect to w\n",
    "  \n",
    "  return dx, dw\n",
    "```\n",
    "\n",
    "After implementing a bunch of layers this way, we will be able to easily combine them to build classifiers with different architectures.\n",
    "  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "executionInfo": {
     "elapsed": 3547,
     "status": "ok",
     "timestamp": 1626360122363,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "2jbEJ1i08hzH",
    "tags": [
     "pdf-ignore"
    ]
   },
   "outputs": [],
   "source": [
    "# As usual, a bit of setup\n",
    "from __future__ import print_function\n",
    "import time\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from cs231n.classifiers.fc_net import *\n",
    "from cs231n.data_utils import get_CIFAR10_data\n",
    "from cs231n.gradient_check import eval_numerical_gradient, eval_numerical_gradient_array\n",
    "from cs231n.solver import Solver\n",
    "\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (10.0, 8.0) # set default size of plots\n",
    "plt.rcParams['image.interpolation'] = 'nearest'\n",
    "plt.rcParams['image.cmap'] = 'gray'\n",
    "\n",
    "# for auto-reloading external modules\n",
    "# see http://stackoverflow.com/questions/1907993/autoreload-of-modules-in-ipython\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "def rel_error(x, y):\n",
    "  \"\"\" returns relative error \"\"\"\n",
    "  return np.max(np.abs(x - y) / (np.maximum(1e-8, np.abs(x) + np.abs(y))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 7746,
     "status": "ok",
     "timestamp": 1626360130099,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "WIb0nUDx8hzH",
    "outputId": "23431ede-8407-4bf7-fc66-187e00095400",
    "tags": [
     "pdf-ignore"
    ]
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('X_train: ', (49000, 3, 32, 32))\n",
      "('y_train: ', (49000,))\n",
      "('X_val: ', (1000, 3, 32, 32))\n",
      "('y_val: ', (1000,))\n",
      "('X_test: ', (1000, 3, 32, 32))\n",
      "('y_test: ', (1000,))\n"
     ]
    }
   ],
   "source": [
    "# Load the (preprocessed) CIFAR10 data.\n",
    "\n",
    "data = get_CIFAR10_data()\n",
    "for k, v in list(data.items()):\n",
    "  print(('%s: ' % k, v.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bhrtSBcd8hzH"
   },
   "source": [
    "# Affine layer: forward\n",
    "Open the file `cs231n/layers.py` and implement the `affine_forward` function.\n",
    "\n",
    "Once you are done you can test your implementaion by running the following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 21,
     "status": "ok",
     "timestamp": 1626360130099,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "V7C0Ezmg8hzI",
    "outputId": "2949843e-4325-494b-fe11-82b72eccd859"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing affine_forward function:\n",
      "difference:  9.769849468192957e-10\n"
     ]
    }
   ],
   "source": [
    "# Test the affine_forward function\n",
    "\n",
    "num_inputs = 2\n",
    "input_shape = (4, 5, 6)\n",
    "output_dim = 3\n",
    "\n",
    "input_size = num_inputs * np.prod(input_shape)\n",
    "weight_size = output_dim * np.prod(input_shape)\n",
    "\n",
    "x = np.linspace(-0.1, 0.5, num=input_size).reshape(num_inputs, *input_shape)\n",
    "w = np.linspace(-0.2, 0.3, num=weight_size).reshape(np.prod(input_shape), output_dim)\n",
    "b = np.linspace(-0.3, 0.1, num=output_dim)\n",
    "\n",
    "out, _ = affine_forward(x, w, b)\n",
    "correct_out = np.array([[ 1.49834967,  1.70660132,  1.91485297],\n",
    "                        [ 3.25553199,  3.5141327,   3.77273342]])\n",
    "\n",
    "# Compare your output with ours. The error should be around e-9 or less.\n",
    "print('Testing affine_forward function:')\n",
    "print('difference: ', rel_error(out, correct_out))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RLcIbv3L8hzI"
   },
   "source": [
    "# Affine layer: backward\n",
    "Now implement the `affine_backward` function and test your implementation using numeric gradient checking."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 19,
     "status": "ok",
     "timestamp": 1626360130100,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "IFFUeBKW8hzI",
    "outputId": "a3ae8e30-7b2c-49d1-a7cf-7dd8cec932b9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing affine_backward function:\n",
      "dx error:  5.399100368651805e-11\n",
      "dw error:  9.904211865398145e-11\n",
      "db error:  2.4122867568119087e-11\n"
     ]
    }
   ],
   "source": [
    "# Test the affine_backward function\n",
    "np.random.seed(231)\n",
    "x = np.random.randn(10, 2, 3)\n",
    "w = np.random.randn(6, 5)\n",
    "b = np.random.randn(5)\n",
    "dout = np.random.randn(10, 5)\n",
    "\n",
    "dx_num = eval_numerical_gradient_array(lambda x: affine_forward(x, w, b)[0], x, dout)\n",
    "dw_num = eval_numerical_gradient_array(lambda w: affine_forward(x, w, b)[0], w, dout)\n",
    "db_num = eval_numerical_gradient_array(lambda b: affine_forward(x, w, b)[0], b, dout)\n",
    "\n",
    "_, cache = affine_forward(x, w, b)\n",
    "dx, dw, db = affine_backward(dout, cache)\n",
    "\n",
    "# The error should be around e-10 or less\n",
    "print('Testing affine_backward function:')\n",
    "print('dx error: ', rel_error(dx_num, dx))\n",
    "print('dw error: ', rel_error(dw_num, dw))\n",
    "print('db error: ', rel_error(db_num, db))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "BrjouC4w8hzJ"
   },
   "source": [
    "# ReLU activation: forward\n",
    "Implement the forward pass for the ReLU activation function in the `relu_forward` function and test your implementation using the following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 16,
     "status": "ok",
     "timestamp": 1626360130100,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "bU3Og7e88hzJ",
    "outputId": "bd121537-1c2a-4dac-f25f-04f99e4e4579"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing relu_forward function:\n",
      "difference:  4.999999798022158e-08\n"
     ]
    }
   ],
   "source": [
    "# Test the relu_forward function\n",
    "\n",
    "x = np.linspace(-0.5, 0.5, num=12).reshape(3, 4)\n",
    "\n",
    "out, _ = relu_forward(x)\n",
    "correct_out = np.array([[ 0.,          0.,          0.,          0.,        ],\n",
    "                        [ 0.,          0.,          0.04545455,  0.13636364,],\n",
    "                        [ 0.22727273,  0.31818182,  0.40909091,  0.5,       ]])\n",
    "\n",
    "# Compare your output with ours. The error should be on the order of e-8\n",
    "print('Testing relu_forward function:')\n",
    "print('difference: ', rel_error(out, correct_out))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NHMuARx28hzJ"
   },
   "source": [
    "# ReLU activation: backward\n",
    "Now implement the backward pass for the ReLU activation function in the `relu_backward` function and test your implementation using numeric gradient checking:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 14,
     "status": "ok",
     "timestamp": 1626360130101,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "8ZimuVUp8hzJ",
    "outputId": "12b0a7a2-4c3e-4761-d08d-ccf3a3b75fcd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing relu_backward function:\n",
      "dx error:  3.2756349136310288e-12\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(231)\n",
    "x = np.random.randn(10, 10)\n",
    "dout = np.random.randn(*x.shape)\n",
    "\n",
    "dx_num = eval_numerical_gradient_array(lambda x: relu_forward(x)[0], x, dout)\n",
    "\n",
    "_, cache = relu_forward(x)\n",
    "dx = relu_backward(dout, cache)\n",
    "\n",
    "# The error should be on the order of e-12\n",
    "print('Testing relu_backward function:')\n",
    "print('dx error: ', rel_error(dx_num, dx))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "SwHHJuRb8hzK"
   },
   "source": [
    "# \"Sandwich\" layers\n",
    "There are some common patterns of layers that are frequently used in neural nets. For example, affine layers are frequently followed by a ReLU nonlinearity. To make these common patterns easy, we define several convenience layers in the file `cs231n/layer_utils.py`.\n",
    "\n",
    "For now take a look at the `affine_relu_forward` and `affine_relu_backward` functions, and run the following to numerically gradient check the backward pass:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 11,
     "status": "ok",
     "timestamp": 1626360130101,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "Ns3HQkmB8hzK",
    "outputId": "68dd6acc-da8a-4692-ae24-f870490f5876"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing affine_relu_forward and affine_relu_backward:\n",
      "dx error:  2.299579177309368e-11\n",
      "dw error:  8.162011105764925e-11\n",
      "db error:  7.826724021458994e-12\n"
     ]
    }
   ],
   "source": [
    "from cs231n.layer_utils import affine_relu_forward, affine_relu_backward\n",
    "np.random.seed(231)\n",
    "x = np.random.randn(2, 3, 4)\n",
    "w = np.random.randn(12, 10)\n",
    "b = np.random.randn(10)\n",
    "dout = np.random.randn(2, 10)\n",
    "\n",
    "out, cache = affine_relu_forward(x, w, b)\n",
    "dx, dw, db = affine_relu_backward(dout, cache)\n",
    "\n",
    "dx_num = eval_numerical_gradient_array(lambda x: affine_relu_forward(x, w, b)[0], x, dout)\n",
    "dw_num = eval_numerical_gradient_array(lambda w: affine_relu_forward(x, w, b)[0], w, dout)\n",
    "db_num = eval_numerical_gradient_array(lambda b: affine_relu_forward(x, w, b)[0], b, dout)\n",
    "\n",
    "# Relative error should be around e-10 or less\n",
    "print('Testing affine_relu_forward and affine_relu_backward:')\n",
    "print('dx error: ', rel_error(dx_num, dx))\n",
    "print('dw error: ', rel_error(dw_num, dw))\n",
    "print('db error: ', rel_error(db_num, db))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1cucUzEe8hzK"
   },
   "source": [
    "# Loss layers: Softmax and SVM\n",
    "Now implement the loss and gradient for softmax and SVM in the `softmax_loss` and `svm_loss` function in `cs231n/layers.py`. These should be similar to what you implemented in `cs231n/classifiers/softmax.py` and `cs231n/classifiers/linear_svm.py`.\n",
    "\n",
    "You can make sure that the implementations are correct by running the following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 379,
     "status": "ok",
     "timestamp": 1626360130472,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "zszMcu_D8hzK",
    "outputId": "8728be95-d1ff-4813-a026-d62e548abfcf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing svm_loss:\n",
      "loss:  8.999602749096233\n",
      "dx error:  1.4021566006651672e-09\n",
      "\n",
      "Testing softmax_loss:\n",
      "loss:  2.302545844500738\n",
      "dx error:  7.457395323221382e-09\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(231)\n",
    "num_classes, num_inputs = 10, 50\n",
    "x = 0.001 * np.random.randn(num_inputs, num_classes)\n",
    "y = np.random.randint(num_classes, size=num_inputs)\n",
    "\n",
    "dx_num = eval_numerical_gradient(lambda x: svm_loss(x, y)[0], x, verbose=False)\n",
    "loss, dx = svm_loss(x, y)\n",
    "\n",
    "# Test svm_loss function. Loss should be around 9 and dx error should be around the order of e-9\n",
    "print('Testing svm_loss:')\n",
    "print('loss: ', loss)\n",
    "print('dx error: ', rel_error(dx_num, dx))\n",
    "\n",
    "dx_num = eval_numerical_gradient(lambda x: softmax_loss(x, y)[0], x, verbose=False)\n",
    "loss, dx = softmax_loss(x, y)\n",
    "\n",
    "# Test softmax_loss function. Loss should be close to 2.3 and dx error should be around e-8\n",
    "print('\\nTesting softmax_loss:')\n",
    "print('loss: ', loss)\n",
    "print('dx error: ', rel_error(dx_num, dx))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "im8TKx9T8hzL"
   },
   "source": [
    "# Two-layer network\n",
    "Open the file `cs231n/classifiers/fc_net.py` and complete the implementation of the `TwoLayerNet` class. Read through it to make sure you understand the API. You can run the cell below to test your implementation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 617,
     "status": "ok",
     "timestamp": 1626360131086,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "R6Kd6sJk8hzL",
    "outputId": "5c65bc4d-682b-41ad-e7a1-2d167a50e4d9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing initialization ... \n",
      "Testing test-time forward pass ... \n",
      "Testing training loss (no regularization)\n",
      "Running numeric gradient check with reg =  0.0\n",
      "W1 relative error: 1.22e-08\n",
      "W2 relative error: 3.31e-10\n",
      "b1 relative error: 6.55e-09\n",
      "b2 relative error: 8.99e-11\n",
      "Running numeric gradient check with reg =  0.7\n",
      "W1 relative error: 2.53e-07\n",
      "W2 relative error: 7.98e-08\n",
      "b1 relative error: 1.56e-08\n",
      "b2 relative error: 7.76e-10\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(231)\n",
    "N, D, H, C = 3, 5, 50, 7\n",
    "X = np.random.randn(N, D)\n",
    "y = np.random.randint(C, size=N)\n",
    "\n",
    "std = 1e-3\n",
    "model = TwoLayerNet(input_dim=D, hidden_dim=H, num_classes=C, weight_scale=std)\n",
    "\n",
    "print('Testing initialization ... ')\n",
    "W1_std = abs(model.params['W1'].std() - std)\n",
    "b1 = model.params['b1']\n",
    "W2_std = abs(model.params['W2'].std() - std)\n",
    "b2 = model.params['b2']\n",
    "assert W1_std < std / 10, 'First layer weights do not seem right'\n",
    "assert np.all(b1 == 0), 'First layer biases do not seem right'\n",
    "assert W2_std < std / 10, 'Second layer weights do not seem right'\n",
    "assert np.all(b2 == 0), 'Second layer biases do not seem right'\n",
    "\n",
    "print('Testing test-time forward pass ... ')\n",
    "model.params['W1'] = np.linspace(-0.7, 0.3, num=D*H).reshape(D, H)\n",
    "model.params['b1'] = np.linspace(-0.1, 0.9, num=H)\n",
    "model.params['W2'] = np.linspace(-0.3, 0.4, num=H*C).reshape(H, C)\n",
    "model.params['b2'] = np.linspace(-0.9, 0.1, num=C)\n",
    "X = np.linspace(-5.5, 4.5, num=N*D).reshape(D, N).T\n",
    "scores = model.loss(X)\n",
    "correct_scores = np.asarray(\n",
    "  [[11.53165108,  12.2917344,   13.05181771,  13.81190102,  14.57198434, 15.33206765,  16.09215096],\n",
    "   [12.05769098,  12.74614105,  13.43459113,  14.1230412,   14.81149128, 15.49994135,  16.18839143],\n",
    "   [12.58373087,  13.20054771,  13.81736455,  14.43418138,  15.05099822, 15.66781506,  16.2846319 ]])\n",
    "scores_diff = np.abs(scores - correct_scores).sum()\n",
    "assert scores_diff < 1e-6, 'Problem with test-time forward pass'\n",
    "\n",
    "print('Testing training loss (no regularization)')\n",
    "y = np.asarray([0, 5, 1])\n",
    "loss, grads = model.loss(X, y)\n",
    "correct_loss = 3.4702243556\n",
    "assert abs(loss - correct_loss) < 1e-10, 'Problem with training-time loss'\n",
    "\n",
    "model.reg = 1.0\n",
    "loss, grads = model.loss(X, y)\n",
    "correct_loss = 26.5948426952\n",
    "assert abs(loss - correct_loss) < 1e-10, 'Problem with regularization loss'\n",
    "\n",
    "# Errors should be around e-7 or less\n",
    "for reg in [0.0, 0.7]:\n",
    "  print('Running numeric gradient check with reg = ', reg)\n",
    "  model.reg = reg\n",
    "  loss, grads = model.loss(X, y)\n",
    "\n",
    "  for name in sorted(grads):\n",
    "    f = lambda _: model.loss(X, y)[0]\n",
    "    grad_num = eval_numerical_gradient(f, model.params[name], verbose=False)\n",
    "    print('%s relative error: %.2e' % (name, rel_error(grad_num, grads[name])))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pfFALj-n8hzL"
   },
   "source": [
    "# Solver\n",
    "Open the file `cs231n/solver.py` and read through it to familiarize yourself with the API. You also need to imeplement the `sgd` function in `cs231n/optim.py`. After doing so, use a `Solver` instance to train a `TwoLayerNet` that achieves about `36%` accuracy on the validation set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 48958,
     "status": "ok",
     "timestamp": 1626360180039,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "IOIwQAED8hzL",
    "outputId": "06beae73-52aa-4cd8-8401-eb62d60d09f0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(Iteration 1 / 4900) loss: 2.300089\n",
      "(Epoch 0 / 10) train acc: 0.171000; val_acc: 0.170000\n",
      "(Iteration 11 / 4900) loss: 2.258416\n",
      "(Iteration 21 / 4900) loss: 2.160948\n",
      "(Iteration 31 / 4900) loss: 2.016287\n",
      "(Iteration 41 / 4900) loss: 2.138642\n",
      "(Iteration 51 / 4900) loss: 2.059372\n",
      "(Iteration 61 / 4900) loss: 1.897230\n",
      "(Iteration 71 / 4900) loss: 1.801362\n",
      "(Iteration 81 / 4900) loss: 1.816051\n",
      "(Iteration 91 / 4900) loss: 1.767155\n",
      "(Iteration 101 / 4900) loss: 1.782419\n",
      "(Iteration 111 / 4900) loss: 1.861558\n",
      "(Iteration 121 / 4900) loss: 1.650334\n",
      "(Iteration 131 / 4900) loss: 1.698891\n",
      "(Iteration 141 / 4900) loss: 1.833884\n",
      "(Iteration 151 / 4900) loss: 1.763588\n",
      "(Iteration 161 / 4900) loss: 1.635848\n",
      "(Iteration 171 / 4900) loss: 1.962973\n",
      "(Iteration 181 / 4900) loss: 1.716530\n",
      "(Iteration 191 / 4900) loss: 1.652857\n",
      "(Iteration 201 / 4900) loss: 1.803466\n",
      "(Iteration 211 / 4900) loss: 1.646484\n",
      "(Iteration 221 / 4900) loss: 1.753302\n",
      "(Iteration 231 / 4900) loss: 1.665357\n",
      "(Iteration 241 / 4900) loss: 1.558509\n",
      "(Iteration 251 / 4900) loss: 1.553072\n",
      "(Iteration 261 / 4900) loss: 1.875181\n",
      "(Iteration 271 / 4900) loss: 1.571899\n",
      "(Iteration 281 / 4900) loss: 1.671443\n",
      "(Iteration 291 / 4900) loss: 1.563806\n",
      "(Iteration 301 / 4900) loss: 1.712676\n",
      "(Iteration 311 / 4900) loss: 1.687622\n",
      "(Iteration 321 / 4900) loss: 1.695280\n",
      "(Iteration 331 / 4900) loss: 1.704362\n",
      "(Iteration 341 / 4900) loss: 1.763766\n",
      "(Iteration 351 / 4900) loss: 1.485922\n",
      "(Iteration 361 / 4900) loss: 1.552275\n",
      "(Iteration 371 / 4900) loss: 1.833757\n",
      "(Iteration 381 / 4900) loss: 1.760161\n",
      "(Iteration 391 / 4900) loss: 1.494989\n",
      "(Iteration 401 / 4900) loss: 1.693946\n",
      "(Iteration 411 / 4900) loss: 1.547319\n",
      "(Iteration 421 / 4900) loss: 1.333266\n",
      "(Iteration 431 / 4900) loss: 1.526631\n",
      "(Iteration 441 / 4900) loss: 1.568662\n",
      "(Iteration 451 / 4900) loss: 1.543955\n",
      "(Iteration 461 / 4900) loss: 1.413793\n",
      "(Iteration 471 / 4900) loss: 1.519715\n",
      "(Iteration 481 / 4900) loss: 1.548025\n",
      "(Epoch 1 / 10) train acc: 0.399000; val_acc: 0.428000\n",
      "(Iteration 491 / 4900) loss: 1.662400\n",
      "(Iteration 501 / 4900) loss: 1.712717\n",
      "(Iteration 511 / 4900) loss: 1.563596\n",
      "(Iteration 521 / 4900) loss: 1.573986\n",
      "(Iteration 531 / 4900) loss: 1.608347\n",
      "(Iteration 541 / 4900) loss: 1.693400\n",
      "(Iteration 551 / 4900) loss: 1.612438\n",
      "(Iteration 561 / 4900) loss: 1.741232\n",
      "(Iteration 571 / 4900) loss: 1.707330\n",
      "(Iteration 581 / 4900) loss: 1.492098\n",
      "(Iteration 591 / 4900) loss: 1.617551\n",
      "(Iteration 601 / 4900) loss: 1.460141\n",
      "(Iteration 611 / 4900) loss: 1.392740\n",
      "(Iteration 621 / 4900) loss: 1.578618\n",
      "(Iteration 631 / 4900) loss: 1.568944\n",
      "(Iteration 641 / 4900) loss: 1.571395\n",
      "(Iteration 651 / 4900) loss: 1.506858\n",
      "(Iteration 661 / 4900) loss: 1.584271\n",
      "(Iteration 671 / 4900) loss: 1.525855\n",
      "(Iteration 681 / 4900) loss: 1.505960\n",
      "(Iteration 691 / 4900) loss: 1.517432\n",
      "(Iteration 701 / 4900) loss: 1.563845\n",
      "(Iteration 711 / 4900) loss: 1.443370\n",
      "(Iteration 721 / 4900) loss: 1.691380\n",
      "(Iteration 731 / 4900) loss: 1.569986\n",
      "(Iteration 741 / 4900) loss: 1.611344\n",
      "(Iteration 751 / 4900) loss: 1.426493\n",
      "(Iteration 761 / 4900) loss: 1.686239\n",
      "(Iteration 771 / 4900) loss: 1.582190\n",
      "(Iteration 781 / 4900) loss: 1.605619\n",
      "(Iteration 791 / 4900) loss: 1.489545\n",
      "(Iteration 801 / 4900) loss: 1.535489\n",
      "(Iteration 811 / 4900) loss: 1.618398\n",
      "(Iteration 821 / 4900) loss: 1.546814\n",
      "(Iteration 831 / 4900) loss: 1.524163\n",
      "(Iteration 841 / 4900) loss: 1.560121\n",
      "(Iteration 851 / 4900) loss: 1.728512\n",
      "(Iteration 861 / 4900) loss: 1.451588\n",
      "(Iteration 871 / 4900) loss: 1.512195\n",
      "(Iteration 881 / 4900) loss: 1.714012\n",
      "(Iteration 891 / 4900) loss: 1.733377\n",
      "(Iteration 901 / 4900) loss: 1.359039\n",
      "(Iteration 911 / 4900) loss: 1.581966\n",
      "(Iteration 921 / 4900) loss: 1.428448\n",
      "(Iteration 931 / 4900) loss: 1.561284\n",
      "(Iteration 941 / 4900) loss: 1.568419\n",
      "(Iteration 951 / 4900) loss: 1.518509\n",
      "(Iteration 961 / 4900) loss: 1.494198\n",
      "(Iteration 971 / 4900) loss: 1.328252\n",
      "(Epoch 2 / 10) train acc: 0.468000; val_acc: 0.444000\n",
      "(Iteration 981 / 4900) loss: 1.446261\n",
      "(Iteration 991 / 4900) loss: 1.393918\n",
      "(Iteration 1001 / 4900) loss: 1.339096\n",
      "(Iteration 1011 / 4900) loss: 1.344386\n",
      "(Iteration 1021 / 4900) loss: 1.256831\n",
      "(Iteration 1031 / 4900) loss: 1.422381\n",
      "(Iteration 1041 / 4900) loss: 1.738975\n",
      "(Iteration 1051 / 4900) loss: 1.553349\n",
      "(Iteration 1061 / 4900) loss: 1.603838\n",
      "(Iteration 1071 / 4900) loss: 1.486361\n",
      "(Iteration 1081 / 4900) loss: 1.417496\n",
      "(Iteration 1091 / 4900) loss: 1.429808\n",
      "(Iteration 1101 / 4900) loss: 1.397059\n",
      "(Iteration 1111 / 4900) loss: 1.653973\n",
      "(Iteration 1121 / 4900) loss: 1.450812\n",
      "(Iteration 1131 / 4900) loss: 1.437969\n",
      "(Iteration 1141 / 4900) loss: 1.448549\n",
      "(Iteration 1151 / 4900) loss: 1.273936\n",
      "(Iteration 1161 / 4900) loss: 1.430609\n",
      "(Iteration 1171 / 4900) loss: 1.559522\n",
      "(Iteration 1181 / 4900) loss: 1.324973\n",
      "(Iteration 1191 / 4900) loss: 1.679667\n",
      "(Iteration 1201 / 4900) loss: 1.499237\n",
      "(Iteration 1211 / 4900) loss: 1.509699\n",
      "(Iteration 1221 / 4900) loss: 1.402940\n",
      "(Iteration 1231 / 4900) loss: 1.277391\n",
      "(Iteration 1241 / 4900) loss: 1.450792\n",
      "(Iteration 1251 / 4900) loss: 1.351350\n",
      "(Iteration 1261 / 4900) loss: 1.600031\n",
      "(Iteration 1271 / 4900) loss: 1.534798\n",
      "(Iteration 1281 / 4900) loss: 1.448485\n",
      "(Iteration 1291 / 4900) loss: 1.116751\n",
      "(Iteration 1301 / 4900) loss: 1.349809\n",
      "(Iteration 1311 / 4900) loss: 1.348709\n",
      "(Iteration 1321 / 4900) loss: 1.377703\n",
      "(Iteration 1331 / 4900) loss: 1.702516\n",
      "(Iteration 1341 / 4900) loss: 1.418050\n",
      "(Iteration 1351 / 4900) loss: 1.537872\n",
      "(Iteration 1361 / 4900) loss: 1.391911\n",
      "(Iteration 1371 / 4900) loss: 1.354068\n",
      "(Iteration 1381 / 4900) loss: 1.705837\n",
      "(Iteration 1391 / 4900) loss: 1.600954\n",
      "(Iteration 1401 / 4900) loss: 1.449001\n",
      "(Iteration 1411 / 4900) loss: 1.453994\n",
      "(Iteration 1421 / 4900) loss: 1.489336\n",
      "(Iteration 1431 / 4900) loss: 1.495201\n",
      "(Iteration 1441 / 4900) loss: 1.226016\n",
      "(Iteration 1451 / 4900) loss: 1.263878\n",
      "(Iteration 1461 / 4900) loss: 1.477084\n",
      "(Epoch 3 / 10) train acc: 0.466000; val_acc: 0.429000\n",
      "(Iteration 1471 / 4900) loss: 1.400271\n",
      "(Iteration 1481 / 4900) loss: 1.309975\n",
      "(Iteration 1491 / 4900) loss: 1.410661\n",
      "(Iteration 1501 / 4900) loss: 1.319916\n",
      "(Iteration 1511 / 4900) loss: 1.393971\n",
      "(Iteration 1521 / 4900) loss: 1.445102\n",
      "(Iteration 1531 / 4900) loss: 1.489275\n",
      "(Iteration 1541 / 4900) loss: 1.500323\n",
      "(Iteration 1551 / 4900) loss: 1.613116\n",
      "(Iteration 1561 / 4900) loss: 1.215678\n",
      "(Iteration 1571 / 4900) loss: 1.375205\n",
      "(Iteration 1581 / 4900) loss: 1.250286\n",
      "(Iteration 1591 / 4900) loss: 1.442918\n",
      "(Iteration 1601 / 4900) loss: 1.441210\n",
      "(Iteration 1611 / 4900) loss: 1.584907\n",
      "(Iteration 1621 / 4900) loss: 1.555882\n",
      "(Iteration 1631 / 4900) loss: 1.538311\n",
      "(Iteration 1641 / 4900) loss: 1.350846\n",
      "(Iteration 1651 / 4900) loss: 1.586443\n",
      "(Iteration 1661 / 4900) loss: 1.244578\n",
      "(Iteration 1671 / 4900) loss: 1.529737\n",
      "(Iteration 1681 / 4900) loss: 1.490461\n",
      "(Iteration 1691 / 4900) loss: 1.429346\n",
      "(Iteration 1701 / 4900) loss: 1.351760\n",
      "(Iteration 1711 / 4900) loss: 1.439255\n",
      "(Iteration 1721 / 4900) loss: 1.404750\n",
      "(Iteration 1731 / 4900) loss: 1.478821\n",
      "(Iteration 1741 / 4900) loss: 1.645068\n",
      "(Iteration 1751 / 4900) loss: 1.452901\n",
      "(Iteration 1761 / 4900) loss: 1.591717\n",
      "(Iteration 1771 / 4900) loss: 1.594355\n",
      "(Iteration 1781 / 4900) loss: 1.531285\n",
      "(Iteration 1791 / 4900) loss: 1.314133\n",
      "(Iteration 1801 / 4900) loss: 1.430905\n",
      "(Iteration 1811 / 4900) loss: 1.280109\n",
      "(Iteration 1821 / 4900) loss: 1.330116\n",
      "(Iteration 1831 / 4900) loss: 1.325624\n",
      "(Iteration 1841 / 4900) loss: 1.394343\n",
      "(Iteration 1851 / 4900) loss: 1.371149\n",
      "(Iteration 1861 / 4900) loss: 1.345813\n",
      "(Iteration 1871 / 4900) loss: 1.425325\n",
      "(Iteration 1881 / 4900) loss: 1.410831\n",
      "(Iteration 1891 / 4900) loss: 1.484850\n",
      "(Iteration 1901 / 4900) loss: 1.347602\n",
      "(Iteration 1911 / 4900) loss: 1.382680\n",
      "(Iteration 1921 / 4900) loss: 1.733093\n",
      "(Iteration 1931 / 4900) loss: 1.418113\n",
      "(Iteration 1941 / 4900) loss: 1.313570\n",
      "(Iteration 1951 / 4900) loss: 1.435940\n",
      "(Epoch 4 / 10) train acc: 0.504000; val_acc: 0.451000\n",
      "(Iteration 1961 / 4900) loss: 1.232119\n",
      "(Iteration 1971 / 4900) loss: 1.611118\n",
      "(Iteration 1981 / 4900) loss: 1.624994\n",
      "(Iteration 1991 / 4900) loss: 1.439298\n",
      "(Iteration 2001 / 4900) loss: 1.452116\n",
      "(Iteration 2011 / 4900) loss: 1.377280\n",
      "(Iteration 2021 / 4900) loss: 1.345182\n",
      "(Iteration 2031 / 4900) loss: 1.723098\n",
      "(Iteration 2041 / 4900) loss: 1.337191\n",
      "(Iteration 2051 / 4900) loss: 1.367573\n",
      "(Iteration 2061 / 4900) loss: 1.363604\n",
      "(Iteration 2071 / 4900) loss: 1.373193\n",
      "(Iteration 2081 / 4900) loss: 1.527777\n",
      "(Iteration 2091 / 4900) loss: 1.294007\n",
      "(Iteration 2101 / 4900) loss: 1.369872\n",
      "(Iteration 2111 / 4900) loss: 1.252515\n",
      "(Iteration 2121 / 4900) loss: 1.632426\n",
      "(Iteration 2131 / 4900) loss: 1.221843\n",
      "(Iteration 2141 / 4900) loss: 1.588294\n",
      "(Iteration 2151 / 4900) loss: 1.382462\n",
      "(Iteration 2161 / 4900) loss: 1.451325\n",
      "(Iteration 2171 / 4900) loss: 1.232785\n",
      "(Iteration 2181 / 4900) loss: 1.428744\n",
      "(Iteration 2191 / 4900) loss: 1.528158\n",
      "(Iteration 2201 / 4900) loss: 1.561029\n",
      "(Iteration 2211 / 4900) loss: 1.545789\n",
      "(Iteration 2221 / 4900) loss: 1.216724\n",
      "(Iteration 2231 / 4900) loss: 1.368028\n",
      "(Iteration 2241 / 4900) loss: 1.419180\n",
      "(Iteration 2251 / 4900) loss: 1.338375\n",
      "(Iteration 2261 / 4900) loss: 1.297813\n",
      "(Iteration 2271 / 4900) loss: 1.293287\n",
      "(Iteration 2281 / 4900) loss: 1.397591\n",
      "(Iteration 2291 / 4900) loss: 1.448465\n",
      "(Iteration 2301 / 4900) loss: 1.242521\n",
      "(Iteration 2311 / 4900) loss: 1.520294\n",
      "(Iteration 2321 / 4900) loss: 1.354138\n",
      "(Iteration 2331 / 4900) loss: 1.309486\n",
      "(Iteration 2341 / 4900) loss: 1.309089\n",
      "(Iteration 2351 / 4900) loss: 1.420478\n",
      "(Iteration 2361 / 4900) loss: 1.299870\n",
      "(Iteration 2371 / 4900) loss: 1.374915\n",
      "(Iteration 2381 / 4900) loss: 1.128089\n",
      "(Iteration 2391 / 4900) loss: 1.195701\n",
      "(Iteration 2401 / 4900) loss: 1.433959\n",
      "(Iteration 2411 / 4900) loss: 1.271562\n",
      "(Iteration 2421 / 4900) loss: 1.342458\n",
      "(Iteration 2431 / 4900) loss: 1.389719\n",
      "(Iteration 2441 / 4900) loss: 1.508970\n",
      "(Epoch 5 / 10) train acc: 0.493000; val_acc: 0.465000\n",
      "(Iteration 2451 / 4900) loss: 1.403120\n",
      "(Iteration 2461 / 4900) loss: 1.362264\n",
      "(Iteration 2471 / 4900) loss: 1.612289\n",
      "(Iteration 2481 / 4900) loss: 1.489801\n",
      "(Iteration 2491 / 4900) loss: 1.429105\n",
      "(Iteration 2501 / 4900) loss: 1.229392\n",
      "(Iteration 2511 / 4900) loss: 1.320781\n",
      "(Iteration 2521 / 4900) loss: 1.348747\n",
      "(Iteration 2531 / 4900) loss: 1.265773\n",
      "(Iteration 2541 / 4900) loss: 1.437888\n",
      "(Iteration 2551 / 4900) loss: 1.245283\n",
      "(Iteration 2561 / 4900) loss: 1.283544\n",
      "(Iteration 2571 / 4900) loss: 1.332414\n",
      "(Iteration 2581 / 4900) loss: 1.376941\n",
      "(Iteration 2591 / 4900) loss: 1.242511\n",
      "(Iteration 2601 / 4900) loss: 1.527411\n",
      "(Iteration 2611 / 4900) loss: 1.485586\n",
      "(Iteration 2621 / 4900) loss: 1.393928\n",
      "(Iteration 2631 / 4900) loss: 1.332372\n",
      "(Iteration 2641 / 4900) loss: 1.302352\n",
      "(Iteration 2651 / 4900) loss: 1.707333\n",
      "(Iteration 2661 / 4900) loss: 1.350690\n",
      "(Iteration 2671 / 4900) loss: 1.335264\n",
      "(Iteration 2681 / 4900) loss: 1.427223\n",
      "(Iteration 2691 / 4900) loss: 1.449200\n",
      "(Iteration 2701 / 4900) loss: 1.262844\n",
      "(Iteration 2711 / 4900) loss: 1.164357\n",
      "(Iteration 2721 / 4900) loss: 1.364723\n",
      "(Iteration 2731 / 4900) loss: 1.279289\n",
      "(Iteration 2741 / 4900) loss: 1.248773\n",
      "(Iteration 2751 / 4900) loss: 1.449679\n",
      "(Iteration 2761 / 4900) loss: 1.137617\n",
      "(Iteration 2771 / 4900) loss: 1.293644\n",
      "(Iteration 2781 / 4900) loss: 1.296194\n",
      "(Iteration 2791 / 4900) loss: 1.434576\n",
      "(Iteration 2801 / 4900) loss: 1.540281\n",
      "(Iteration 2811 / 4900) loss: 1.651625\n",
      "(Iteration 2821 / 4900) loss: 1.333709\n",
      "(Iteration 2831 / 4900) loss: 1.406024\n",
      "(Iteration 2841 / 4900) loss: 1.375746\n",
      "(Iteration 2851 / 4900) loss: 1.317436\n",
      "(Iteration 2861 / 4900) loss: 1.410073\n",
      "(Iteration 2871 / 4900) loss: 1.326769\n",
      "(Iteration 2881 / 4900) loss: 1.210212\n",
      "(Iteration 2891 / 4900) loss: 1.361641\n",
      "(Iteration 2901 / 4900) loss: 1.061335\n",
      "(Iteration 2911 / 4900) loss: 1.490168\n",
      "(Iteration 2921 / 4900) loss: 1.531250\n",
      "(Iteration 2931 / 4900) loss: 1.734393\n",
      "(Epoch 6 / 10) train acc: 0.537000; val_acc: 0.488000\n",
      "(Iteration 2941 / 4900) loss: 1.343652\n",
      "(Iteration 2951 / 4900) loss: 1.309466\n",
      "(Iteration 2961 / 4900) loss: 1.137099\n",
      "(Iteration 2971 / 4900) loss: 1.404897\n",
      "(Iteration 2981 / 4900) loss: 1.499112\n",
      "(Iteration 2991 / 4900) loss: 1.182964\n",
      "(Iteration 3001 / 4900) loss: 1.375969\n",
      "(Iteration 3011 / 4900) loss: 1.475385\n",
      "(Iteration 3021 / 4900) loss: 1.297671\n",
      "(Iteration 3031 / 4900) loss: 1.455192\n",
      "(Iteration 3041 / 4900) loss: 1.288831\n",
      "(Iteration 3051 / 4900) loss: 1.252507\n",
      "(Iteration 3061 / 4900) loss: 1.273133\n",
      "(Iteration 3071 / 4900) loss: 1.245781\n",
      "(Iteration 3081 / 4900) loss: 1.526020\n",
      "(Iteration 3091 / 4900) loss: 1.244705\n",
      "(Iteration 3101 / 4900) loss: 1.168029\n",
      "(Iteration 3111 / 4900) loss: 1.199980\n",
      "(Iteration 3121 / 4900) loss: 1.616235\n",
      "(Iteration 3131 / 4900) loss: 1.228325\n",
      "(Iteration 3141 / 4900) loss: 1.372394\n",
      "(Iteration 3151 / 4900) loss: 1.261772\n",
      "(Iteration 3161 / 4900) loss: 1.472383\n",
      "(Iteration 3171 / 4900) loss: 1.129794\n",
      "(Iteration 3181 / 4900) loss: 1.396840\n",
      "(Iteration 3191 / 4900) loss: 1.405426\n",
      "(Iteration 3201 / 4900) loss: 1.270886\n",
      "(Iteration 3211 / 4900) loss: 1.421649\n",
      "(Iteration 3221 / 4900) loss: 1.503245\n",
      "(Iteration 3231 / 4900) loss: 1.568326\n",
      "(Iteration 3241 / 4900) loss: 1.393653\n",
      "(Iteration 3251 / 4900) loss: 1.318088\n",
      "(Iteration 3261 / 4900) loss: 1.332783\n",
      "(Iteration 3271 / 4900) loss: 1.606893\n",
      "(Iteration 3281 / 4900) loss: 1.276238\n",
      "(Iteration 3291 / 4900) loss: 1.216147\n",
      "(Iteration 3301 / 4900) loss: 1.360948\n",
      "(Iteration 3311 / 4900) loss: 1.328076\n",
      "(Iteration 3321 / 4900) loss: 1.329054\n",
      "(Iteration 3331 / 4900) loss: 1.150582\n",
      "(Iteration 3341 / 4900) loss: 1.499849\n",
      "(Iteration 3351 / 4900) loss: 1.232690\n",
      "(Iteration 3361 / 4900) loss: 1.357872\n",
      "(Iteration 3371 / 4900) loss: 1.577824\n",
      "(Iteration 3381 / 4900) loss: 1.394309\n",
      "(Iteration 3391 / 4900) loss: 1.458092\n",
      "(Iteration 3401 / 4900) loss: 1.218180\n",
      "(Iteration 3411 / 4900) loss: 1.264785\n",
      "(Iteration 3421 / 4900) loss: 1.199487\n",
      "(Epoch 7 / 10) train acc: 0.531000; val_acc: 0.468000\n",
      "(Iteration 3431 / 4900) loss: 1.311152\n",
      "(Iteration 3441 / 4900) loss: 1.364160\n",
      "(Iteration 3451 / 4900) loss: 1.306556\n",
      "(Iteration 3461 / 4900) loss: 1.399136\n",
      "(Iteration 3471 / 4900) loss: 1.317526\n",
      "(Iteration 3481 / 4900) loss: 1.274651\n",
      "(Iteration 3491 / 4900) loss: 1.581665\n",
      "(Iteration 3501 / 4900) loss: 1.492338\n",
      "(Iteration 3511 / 4900) loss: 1.366844\n",
      "(Iteration 3521 / 4900) loss: 1.406317\n",
      "(Iteration 3531 / 4900) loss: 1.242057\n",
      "(Iteration 3541 / 4900) loss: 1.591405\n",
      "(Iteration 3551 / 4900) loss: 1.339095\n",
      "(Iteration 3561 / 4900) loss: 1.088366\n",
      "(Iteration 3571 / 4900) loss: 1.433389\n",
      "(Iteration 3581 / 4900) loss: 1.272246\n",
      "(Iteration 3591 / 4900) loss: 1.236100\n",
      "(Iteration 3601 / 4900) loss: 1.193582\n",
      "(Iteration 3611 / 4900) loss: 1.481540\n",
      "(Iteration 3621 / 4900) loss: 1.297923\n",
      "(Iteration 3631 / 4900) loss: 1.281803\n",
      "(Iteration 3641 / 4900) loss: 1.311063\n",
      "(Iteration 3651 / 4900) loss: 1.380828\n",
      "(Iteration 3661 / 4900) loss: 1.321568\n",
      "(Iteration 3671 / 4900) loss: 1.343057\n",
      "(Iteration 3681 / 4900) loss: 1.389960\n",
      "(Iteration 3691 / 4900) loss: 1.223307\n",
      "(Iteration 3701 / 4900) loss: 1.478843\n",
      "(Iteration 3711 / 4900) loss: 1.594951\n",
      "(Iteration 3721 / 4900) loss: 1.250862\n",
      "(Iteration 3731 / 4900) loss: 1.162776\n",
      "(Iteration 3741 / 4900) loss: 1.340488\n",
      "(Iteration 3751 / 4900) loss: 1.367186\n",
      "(Iteration 3761 / 4900) loss: 1.366398\n",
      "(Iteration 3771 / 4900) loss: 1.417237\n",
      "(Iteration 3781 / 4900) loss: 1.344297\n",
      "(Iteration 3791 / 4900) loss: 1.396264\n",
      "(Iteration 3801 / 4900) loss: 1.238349\n",
      "(Iteration 3811 / 4900) loss: 1.344179\n",
      "(Iteration 3821 / 4900) loss: 1.219159\n",
      "(Iteration 3831 / 4900) loss: 1.361244\n",
      "(Iteration 3841 / 4900) loss: 1.443448\n",
      "(Iteration 3851 / 4900) loss: 1.334387\n",
      "(Iteration 3861 / 4900) loss: 1.360704\n",
      "(Iteration 3871 / 4900) loss: 1.257395\n",
      "(Iteration 3881 / 4900) loss: 1.345110\n",
      "(Iteration 3891 / 4900) loss: 1.215053\n",
      "(Iteration 3901 / 4900) loss: 1.245863\n",
      "(Iteration 3911 / 4900) loss: 1.357334\n",
      "(Epoch 8 / 10) train acc: 0.541000; val_acc: 0.463000\n",
      "(Iteration 3921 / 4900) loss: 1.150966\n",
      "(Iteration 3931 / 4900) loss: 1.319383\n",
      "(Iteration 3941 / 4900) loss: 1.069093\n",
      "(Iteration 3951 / 4900) loss: 1.462634\n",
      "(Iteration 3961 / 4900) loss: 1.284100\n",
      "(Iteration 3971 / 4900) loss: 1.159020\n",
      "(Iteration 3981 / 4900) loss: 1.127782\n",
      "(Iteration 3991 / 4900) loss: 1.110383\n",
      "(Iteration 4001 / 4900) loss: 1.189466\n",
      "(Iteration 4011 / 4900) loss: 1.271651\n",
      "(Iteration 4021 / 4900) loss: 1.461067\n",
      "(Iteration 4031 / 4900) loss: 1.365532\n",
      "(Iteration 4041 / 4900) loss: 1.070091\n",
      "(Iteration 4051 / 4900) loss: 1.431787\n",
      "(Iteration 4061 / 4900) loss: 1.414741\n",
      "(Iteration 4071 / 4900) loss: 1.361378\n",
      "(Iteration 4081 / 4900) loss: 1.522182\n",
      "(Iteration 4091 / 4900) loss: 1.287178\n",
      "(Iteration 4101 / 4900) loss: 1.308431\n",
      "(Iteration 4111 / 4900) loss: 1.589505\n",
      "(Iteration 4121 / 4900) loss: 1.423470\n",
      "(Iteration 4131 / 4900) loss: 1.724517\n",
      "(Iteration 4141 / 4900) loss: 1.183873\n",
      "(Iteration 4151 / 4900) loss: 1.029758\n",
      "(Iteration 4161 / 4900) loss: 1.384384\n",
      "(Iteration 4171 / 4900) loss: 1.469837\n",
      "(Iteration 4181 / 4900) loss: 1.529209\n",
      "(Iteration 4191 / 4900) loss: 1.190913\n",
      "(Iteration 4201 / 4900) loss: 1.214756\n",
      "(Iteration 4211 / 4900) loss: 1.133497\n",
      "(Iteration 4221 / 4900) loss: 1.354617\n",
      "(Iteration 4231 / 4900) loss: 1.273540\n",
      "(Iteration 4241 / 4900) loss: 1.477614\n",
      "(Iteration 4251 / 4900) loss: 1.584060\n",
      "(Iteration 4261 / 4900) loss: 1.378007\n",
      "(Iteration 4271 / 4900) loss: 1.200474\n",
      "(Iteration 4281 / 4900) loss: 1.268054\n",
      "(Iteration 4291 / 4900) loss: 1.154626\n",
      "(Iteration 4301 / 4900) loss: 1.206561\n",
      "(Iteration 4311 / 4900) loss: 1.067481\n",
      "(Iteration 4321 / 4900) loss: 1.508419\n",
      "(Iteration 4331 / 4900) loss: 1.348757\n",
      "(Iteration 4341 / 4900) loss: 1.383121\n",
      "(Iteration 4351 / 4900) loss: 1.300875\n",
      "(Iteration 4361 / 4900) loss: 1.233478\n",
      "(Iteration 4371 / 4900) loss: 1.221144\n",
      "(Iteration 4381 / 4900) loss: 1.239890\n",
      "(Iteration 4391 / 4900) loss: 1.323033\n",
      "(Iteration 4401 / 4900) loss: 1.226766\n",
      "(Epoch 9 / 10) train acc: 0.580000; val_acc: 0.469000\n",
      "(Iteration 4411 / 4900) loss: 1.534061\n",
      "(Iteration 4421 / 4900) loss: 1.022927\n",
      "(Iteration 4431 / 4900) loss: 1.294994\n",
      "(Iteration 4441 / 4900) loss: 1.360441\n",
      "(Iteration 4451 / 4900) loss: 1.380112\n",
      "(Iteration 4461 / 4900) loss: 1.143059\n",
      "(Iteration 4471 / 4900) loss: 1.245423\n",
      "(Iteration 4481 / 4900) loss: 1.408525\n",
      "(Iteration 4491 / 4900) loss: 1.210972\n",
      "(Iteration 4501 / 4900) loss: 1.394010\n",
      "(Iteration 4511 / 4900) loss: 1.203786\n",
      "(Iteration 4521 / 4900) loss: 1.060802\n",
      "(Iteration 4531 / 4900) loss: 1.464206\n",
      "(Iteration 4541 / 4900) loss: 1.260809\n",
      "(Iteration 4551 / 4900) loss: 1.332313\n",
      "(Iteration 4561 / 4900) loss: 1.226907\n",
      "(Iteration 4571 / 4900) loss: 1.556329\n",
      "(Iteration 4581 / 4900) loss: 1.326948\n",
      "(Iteration 4591 / 4900) loss: 1.211093\n",
      "(Iteration 4601 / 4900) loss: 1.254027\n",
      "(Iteration 4611 / 4900) loss: 1.474277\n",
      "(Iteration 4621 / 4900) loss: 1.288402\n",
      "(Iteration 4631 / 4900) loss: 1.466134\n",
      "(Iteration 4641 / 4900) loss: 1.088726\n",
      "(Iteration 4651 / 4900) loss: 1.397102\n",
      "(Iteration 4661 / 4900) loss: 1.441696\n",
      "(Iteration 4671 / 4900) loss: 1.584400\n",
      "(Iteration 4681 / 4900) loss: 1.289660\n",
      "(Iteration 4691 / 4900) loss: 1.137633\n",
      "(Iteration 4701 / 4900) loss: 1.267596\n",
      "(Iteration 4711 / 4900) loss: 1.196905\n",
      "(Iteration 4721 / 4900) loss: 1.246403\n",
      "(Iteration 4731 / 4900) loss: 1.285876\n",
      "(Iteration 4741 / 4900) loss: 1.326565\n",
      "(Iteration 4751 / 4900) loss: 1.170252\n",
      "(Iteration 4761 / 4900) loss: 1.246038\n",
      "(Iteration 4771 / 4900) loss: 1.379887\n",
      "(Iteration 4781 / 4900) loss: 1.176485\n",
      "(Iteration 4791 / 4900) loss: 1.128133\n",
      "(Iteration 4801 / 4900) loss: 1.317326\n",
      "(Iteration 4811 / 4900) loss: 1.425641\n",
      "(Iteration 4821 / 4900) loss: 1.359672\n",
      "(Iteration 4831 / 4900) loss: 1.331216\n",
      "(Iteration 4841 / 4900) loss: 1.210385\n",
      "(Iteration 4851 / 4900) loss: 1.258650\n",
      "(Iteration 4861 / 4900) loss: 1.214546\n",
      "(Iteration 4871 / 4900) loss: 1.285254\n",
      "(Iteration 4881 / 4900) loss: 1.383855\n",
      "(Iteration 4891 / 4900) loss: 1.421827\n",
      "(Epoch 10 / 10) train acc: 0.524000; val_acc: 0.456000\n"
     ]
    }
   ],
   "source": [
    "input_size = 32 * 32 * 3\n",
    "hidden_size = 50\n",
    "num_classes = 10\n",
    "model = TwoLayerNet(input_size, hidden_size, num_classes)\n",
    "solver = None\n",
    "\n",
    "##############################################################################\n",
    "# TODO: Use a Solver instance to train a TwoLayerNet that achieves about 36% #\n",
    "# accuracy on the validation set.                                            #\n",
    "##############################################################################\n",
    "# *****START OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****\n",
    "from cs231n.optim import sgd\n",
    "solver = Solver(model, data, update_rule='sgd', optim_config={'learning_rate': 1e-3})\n",
    "solver.train()\n",
    "\n",
    "# *****END OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****\n",
    "##############################################################################\n",
    "#                             END OF YOUR CODE                               #\n",
    "##############################################################################"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jH0jq7nS8hzM"
   },
   "source": [
    "# Debug the training\n",
    "With the default parameters we provided above, you should get a validation accuracy of about 0.36 on the validation set. This isn't very good.\n",
    "\n",
    "One strategy for getting insight into what's wrong is to plot the loss function and the accuracies on the training and validation sets during optimization.\n",
    "\n",
    "Another strategy is to visualize the weights that were learned in the first layer of the network. In most neural networks trained on visual data, the first layer weights typically show some visible structure when visualized."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 730
    },
    "executionInfo": {
     "elapsed": 1025,
     "status": "ok",
     "timestamp": 1626360181061,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "flY6vet28hzM",
    "outputId": "647f7b03-04d8-4f58-83fd-b79dbe5d6fcf"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x864 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Run this cell to visualize training loss and train / val accuracy\n",
    "\n",
    "plt.subplot(2, 1, 1)\n",
    "plt.title('Training loss')\n",
    "plt.plot(solver.loss_history, 'o')\n",
    "plt.xlabel('Iteration')\n",
    "\n",
    "plt.subplot(2, 1, 2)\n",
    "plt.title('Accuracy')\n",
    "plt.plot(solver.train_acc_history, '-o', label='train')\n",
    "plt.plot(solver.val_acc_history, '-o', label='val')\n",
    "plt.plot([0.5] * len(solver.val_acc_history), 'k--')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(loc='lower right')\n",
    "plt.gcf().set_size_inches(15, 12)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 466
    },
    "executionInfo": {
     "elapsed": 1021,
     "status": "ok",
     "timestamp": 1626360182065,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "QoNGSZ9B8hzM",
    "outputId": "2cc620fa-223c-4056-f454-dc69bbcbadf3"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from cs231n.vis_utils import visualize_grid\n",
    "\n",
    "# Visualize the weights of the network\n",
    "\n",
    "def show_net_weights(net):\n",
    "    W1 = net.params['W1']\n",
    "    #W1 = W1.reshape(32, 32, 3, -1).transpose(3, 0, 1, 2)\n",
    "    W1 = W1.reshape(3, 32, 32, -1).transpose(3, 1, 2, 0)\n",
    "    plt.imshow(visualize_grid(W1, padding=3).astype('uint8'))\n",
    "    plt.gca().axis('off')\n",
    "    plt.show()\n",
    "\n",
    "show_net_weights(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZMld7bk98hzM"
   },
   "source": [
    "# Tune your hyperparameters\n",
    "\n",
    "**What's wrong?**. Looking at the visualizations above, we see that the loss is decreasing more or less linearly, which seems to suggest that the learning rate may be too low. Moreover, there is no gap between the training and validation accuracy, suggesting that the model we used has low capacity, and that we should increase its size. On the other hand, with a very large model we would expect to see more overfitting, which would manifest itself as a very large gap between the training and validation accuracy.\n",
    "\n",
    "**Tuning**. Tuning the hyperparameters and developing intuition for how they affect the final performance is a large part of using Neural Networks, so we want you to get a lot of practice. Below, you should experiment with different values of the various hyperparameters, including hidden layer size, learning rate, numer of training epochs, and regularization strength. You might also consider tuning the learning rate decay, but you should be able to get good performance using the default value.\n",
    "\n",
    "**Approximate results**. You should be aim to achieve a classification accuracy of greater than 48% on the validation set. Our best network gets over 52% on the validation set.\n",
    "\n",
    "**Experiment**: You goal in this exercise is to get as good of a result on CIFAR-10 as you can (52% could serve as a reference), with a fully-connected Neural Network. Feel free implement your own techniques (e.g. PCA to reduce dimensionality, or adding dropout, or adding features to the solver, etc.)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 47205,
     "status": "ok",
     "timestamp": 1626360229256,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "-f65xLErjsSi",
    "outputId": "51b20158-e5ed-4eaf-980e-fb20147f56ff"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(Iteration 1 / 4900) loss: 2.303135\n",
      "(Epoch 0 / 10) train acc: 0.134000; val_acc: 0.119000\n",
      "(Iteration 11 / 4900) loss: 2.273055\n",
      "(Iteration 21 / 4900) loss: 2.151980\n",
      "(Iteration 31 / 4900) loss: 2.037189\n",
      "(Iteration 41 / 4900) loss: 2.025704\n",
      "(Iteration 51 / 4900) loss: 1.951077\n",
      "(Iteration 61 / 4900) loss: 1.913436\n",
      "(Iteration 71 / 4900) loss: 1.982525\n",
      "(Iteration 81 / 4900) loss: 1.754864\n",
      "(Iteration 91 / 4900) loss: 1.980825\n",
      "(Iteration 101 / 4900) loss: 1.762772\n",
      "(Iteration 111 / 4900) loss: 1.793151\n",
      "(Iteration 121 / 4900) loss: 1.733250\n",
      "(Iteration 131 / 4900) loss: 1.880929\n",
      "(Iteration 141 / 4900) loss: 1.814176\n",
      "(Iteration 151 / 4900) loss: 1.638323\n",
      "(Iteration 161 / 4900) loss: 1.846015\n",
      "(Iteration 171 / 4900) loss: 1.606403\n",
      "(Iteration 181 / 4900) loss: 1.801575\n",
      "(Iteration 191 / 4900) loss: 1.582364\n",
      "(Iteration 201 / 4900) loss: 1.714300\n",
      "(Iteration 211 / 4900) loss: 1.802526\n",
      "(Iteration 221 / 4900) loss: 1.888274\n",
      "(Iteration 231 / 4900) loss: 1.752473\n",
      "(Iteration 241 / 4900) loss: 1.773418\n",
      "(Iteration 251 / 4900) loss: 1.554089\n",
      "(Iteration 261 / 4900) loss: 1.661432\n",
      "(Iteration 271 / 4900) loss: 1.680302\n",
      "(Iteration 281 / 4900) loss: 1.745812\n",
      "(Iteration 291 / 4900) loss: 1.765550\n",
      "(Iteration 301 / 4900) loss: 1.629307\n",
      "(Iteration 311 / 4900) loss: 1.621465\n",
      "(Iteration 321 / 4900) loss: 1.563124\n",
      "(Iteration 331 / 4900) loss: 1.661411\n",
      "(Iteration 341 / 4900) loss: 1.573371\n",
      "(Iteration 351 / 4900) loss: 1.389815\n",
      "(Iteration 361 / 4900) loss: 1.464124\n",
      "(Iteration 371 / 4900) loss: 1.568624\n",
      "(Iteration 381 / 4900) loss: 1.773950\n",
      "(Iteration 391 / 4900) loss: 1.383119\n",
      "(Iteration 401 / 4900) loss: 1.580161\n",
      "(Iteration 411 / 4900) loss: 1.595416\n",
      "(Iteration 421 / 4900) loss: 1.517748\n",
      "(Iteration 431 / 4900) loss: 1.813143\n",
      "(Iteration 441 / 4900) loss: 1.454937\n",
      "(Iteration 451 / 4900) loss: 1.619568\n",
      "(Iteration 461 / 4900) loss: 1.538195\n",
      "(Iteration 471 / 4900) loss: 1.722874\n",
      "(Iteration 481 / 4900) loss: 1.597439\n",
      "(Epoch 1 / 10) train acc: 0.461000; val_acc: 0.432000\n",
      "(Iteration 491 / 4900) loss: 1.604157\n",
      "(Iteration 501 / 4900) loss: 1.462797\n",
      "(Iteration 511 / 4900) loss: 1.463962\n",
      "(Iteration 521 / 4900) loss: 1.421876\n",
      "(Iteration 531 / 4900) loss: 1.529825\n",
      "(Iteration 541 / 4900) loss: 1.639849\n",
      "(Iteration 551 / 4900) loss: 1.681699\n",
      "(Iteration 561 / 4900) loss: 1.779913\n",
      "(Iteration 571 / 4900) loss: 1.470497\n",
      "(Iteration 581 / 4900) loss: 1.563881\n",
      "(Iteration 591 / 4900) loss: 1.552871\n",
      "(Iteration 601 / 4900) loss: 1.629276\n",
      "(Iteration 611 / 4900) loss: 1.678338\n",
      "(Iteration 621 / 4900) loss: 1.589351\n",
      "(Iteration 631 / 4900) loss: 1.622317\n",
      "(Iteration 641 / 4900) loss: 1.636637\n",
      "(Iteration 651 / 4900) loss: 1.666684\n",
      "(Iteration 661 / 4900) loss: 1.654210\n",
      "(Iteration 671 / 4900) loss: 1.606297\n",
      "(Iteration 681 / 4900) loss: 1.532440\n",
      "(Iteration 691 / 4900) loss: 1.658671\n",
      "(Iteration 701 / 4900) loss: 1.558206\n",
      "(Iteration 711 / 4900) loss: 1.584947\n",
      "(Iteration 721 / 4900) loss: 1.374618\n",
      "(Iteration 731 / 4900) loss: 1.570100\n",
      "(Iteration 741 / 4900) loss: 1.513613\n",
      "(Iteration 751 / 4900) loss: 1.519132\n",
      "(Iteration 761 / 4900) loss: 1.533646\n",
      "(Iteration 771 / 4900) loss: 1.477033\n",
      "(Iteration 781 / 4900) loss: 1.567298\n",
      "(Iteration 791 / 4900) loss: 1.536872\n",
      "(Iteration 801 / 4900) loss: 1.386903\n",
      "(Iteration 811 / 4900) loss: 1.599250\n",
      "(Iteration 821 / 4900) loss: 1.534907\n",
      "(Iteration 831 / 4900) loss: 1.415091\n",
      "(Iteration 841 / 4900) loss: 1.521825\n",
      "(Iteration 851 / 4900) loss: 1.648699\n",
      "(Iteration 861 / 4900) loss: 1.643029\n",
      "(Iteration 871 / 4900) loss: 1.598316\n",
      "(Iteration 881 / 4900) loss: 1.472998\n",
      "(Iteration 891 / 4900) loss: 1.431323\n",
      "(Iteration 901 / 4900) loss: 1.549559\n",
      "(Iteration 911 / 4900) loss: 1.608492\n",
      "(Iteration 921 / 4900) loss: 1.371177\n",
      "(Iteration 931 / 4900) loss: 1.441437\n",
      "(Iteration 941 / 4900) loss: 1.536256\n",
      "(Iteration 951 / 4900) loss: 1.530842\n",
      "(Iteration 961 / 4900) loss: 1.360509\n",
      "(Iteration 971 / 4900) loss: 1.435590\n",
      "(Epoch 2 / 10) train acc: 0.494000; val_acc: 0.494000\n",
      "(Iteration 981 / 4900) loss: 1.542649\n",
      "(Iteration 991 / 4900) loss: 1.228115\n",
      "(Iteration 1001 / 4900) loss: 1.441777\n",
      "(Iteration 1011 / 4900) loss: 1.499143\n",
      "(Iteration 1021 / 4900) loss: 1.313547\n",
      "(Iteration 1031 / 4900) loss: 1.743127\n",
      "(Iteration 1041 / 4900) loss: 1.454153\n",
      "(Iteration 1051 / 4900) loss: 1.513233\n",
      "(Iteration 1061 / 4900) loss: 1.316207\n",
      "(Iteration 1071 / 4900) loss: 1.485452\n",
      "(Iteration 1081 / 4900) loss: 1.643911\n",
      "(Iteration 1091 / 4900) loss: 1.576459\n",
      "(Iteration 1101 / 4900) loss: 1.497573\n",
      "(Iteration 1111 / 4900) loss: 1.585340\n",
      "(Iteration 1121 / 4900) loss: 1.405618\n",
      "(Iteration 1131 / 4900) loss: 1.508545\n",
      "(Iteration 1141 / 4900) loss: 1.616868\n",
      "(Iteration 1151 / 4900) loss: 1.350805\n",
      "(Iteration 1161 / 4900) loss: 1.613982\n",
      "(Iteration 1171 / 4900) loss: 1.361131\n",
      "(Iteration 1181 / 4900) loss: 1.313979\n",
      "(Iteration 1191 / 4900) loss: 1.749133\n",
      "(Iteration 1201 / 4900) loss: 1.402233\n",
      "(Iteration 1211 / 4900) loss: 1.436248\n",
      "(Iteration 1221 / 4900) loss: 1.691761\n",
      "(Iteration 1231 / 4900) loss: 1.769987\n",
      "(Iteration 1241 / 4900) loss: 1.476786\n",
      "(Iteration 1251 / 4900) loss: 1.432007\n",
      "(Iteration 1261 / 4900) loss: 1.609349\n",
      "(Iteration 1271 / 4900) loss: 1.547582\n",
      "(Iteration 1281 / 4900) loss: 1.410155\n",
      "(Iteration 1291 / 4900) loss: 1.574010\n",
      "(Iteration 1301 / 4900) loss: 1.551780\n",
      "(Iteration 1311 / 4900) loss: 1.588495\n",
      "(Iteration 1321 / 4900) loss: 1.413478\n",
      "(Iteration 1331 / 4900) loss: 1.414720\n",
      "(Iteration 1341 / 4900) loss: 1.531052\n",
      "(Iteration 1351 / 4900) loss: 1.652981\n",
      "(Iteration 1361 / 4900) loss: 1.398163\n",
      "(Iteration 1371 / 4900) loss: 1.385532\n",
      "(Iteration 1381 / 4900) loss: 1.486365\n",
      "(Iteration 1391 / 4900) loss: 1.410783\n",
      "(Iteration 1401 / 4900) loss: 1.401632\n",
      "(Iteration 1411 / 4900) loss: 1.213304\n",
      "(Iteration 1421 / 4900) loss: 1.468859\n",
      "(Iteration 1431 / 4900) loss: 1.289342\n",
      "(Iteration 1441 / 4900) loss: 1.425800\n",
      "(Iteration 1451 / 4900) loss: 1.380220\n",
      "(Iteration 1461 / 4900) loss: 1.468510\n",
      "(Epoch 3 / 10) train acc: 0.507000; val_acc: 0.486000\n",
      "(Iteration 1471 / 4900) loss: 1.407960\n",
      "(Iteration 1481 / 4900) loss: 1.578560\n",
      "(Iteration 1491 / 4900) loss: 1.512675\n",
      "(Iteration 1501 / 4900) loss: 1.345975\n",
      "(Iteration 1511 / 4900) loss: 1.437019\n",
      "(Iteration 1521 / 4900) loss: 1.333207\n",
      "(Iteration 1531 / 4900) loss: 1.640925\n",
      "(Iteration 1541 / 4900) loss: 1.414993\n",
      "(Iteration 1551 / 4900) loss: 1.329616\n",
      "(Iteration 1561 / 4900) loss: 1.492991\n",
      "(Iteration 1571 / 4900) loss: 1.395845\n",
      "(Iteration 1581 / 4900) loss: 1.474963\n",
      "(Iteration 1591 / 4900) loss: 1.481767\n",
      "(Iteration 1601 / 4900) loss: 1.501716\n",
      "(Iteration 1611 / 4900) loss: 1.356404\n",
      "(Iteration 1621 / 4900) loss: 1.375487\n",
      "(Iteration 1631 / 4900) loss: 1.501533\n",
      "(Iteration 1641 / 4900) loss: 1.405608\n",
      "(Iteration 1651 / 4900) loss: 1.448722\n",
      "(Iteration 1661 / 4900) loss: 1.481932\n",
      "(Iteration 1671 / 4900) loss: 1.493373\n",
      "(Iteration 1681 / 4900) loss: 1.619640\n",
      "(Iteration 1691 / 4900) loss: 1.488721\n",
      "(Iteration 1701 / 4900) loss: 1.397466\n",
      "(Iteration 1711 / 4900) loss: 1.437717\n",
      "(Iteration 1721 / 4900) loss: 1.425500\n",
      "(Iteration 1731 / 4900) loss: 1.318667\n",
      "(Iteration 1741 / 4900) loss: 1.453693\n",
      "(Iteration 1751 / 4900) loss: 1.264408\n",
      "(Iteration 1761 / 4900) loss: 1.296150\n",
      "(Iteration 1771 / 4900) loss: 1.466363\n",
      "(Iteration 1781 / 4900) loss: 1.344513\n",
      "(Iteration 1791 / 4900) loss: 1.538730\n",
      "(Iteration 1801 / 4900) loss: 1.287073\n",
      "(Iteration 1811 / 4900) loss: 1.333683\n",
      "(Iteration 1821 / 4900) loss: 1.287518\n",
      "(Iteration 1831 / 4900) loss: 1.500036\n",
      "(Iteration 1841 / 4900) loss: 1.465890\n",
      "(Iteration 1851 / 4900) loss: 1.226041\n",
      "(Iteration 1861 / 4900) loss: 1.433105\n",
      "(Iteration 1871 / 4900) loss: 1.337296\n",
      "(Iteration 1881 / 4900) loss: 1.484059\n",
      "(Iteration 1891 / 4900) loss: 1.425265\n",
      "(Iteration 1901 / 4900) loss: 1.309830\n",
      "(Iteration 1911 / 4900) loss: 1.360629\n",
      "(Iteration 1921 / 4900) loss: 1.354950\n",
      "(Iteration 1931 / 4900) loss: 1.428912\n",
      "(Iteration 1941 / 4900) loss: 1.508167\n",
      "(Iteration 1951 / 4900) loss: 1.325889\n",
      "(Epoch 4 / 10) train acc: 0.522000; val_acc: 0.480000\n",
      "(Iteration 1961 / 4900) loss: 1.471540\n",
      "(Iteration 1971 / 4900) loss: 1.652782\n",
      "(Iteration 1981 / 4900) loss: 1.330952\n",
      "(Iteration 1991 / 4900) loss: 1.358119\n",
      "(Iteration 2001 / 4900) loss: 1.328040\n",
      "(Iteration 2011 / 4900) loss: 1.297900\n",
      "(Iteration 2021 / 4900) loss: 1.453255\n",
      "(Iteration 2031 / 4900) loss: 1.345309\n",
      "(Iteration 2041 / 4900) loss: 1.564728\n",
      "(Iteration 2051 / 4900) loss: 1.434479\n",
      "(Iteration 2061 / 4900) loss: 1.239752\n",
      "(Iteration 2071 / 4900) loss: 1.352898\n",
      "(Iteration 2081 / 4900) loss: 1.253595\n",
      "(Iteration 2091 / 4900) loss: 1.711069\n",
      "(Iteration 2101 / 4900) loss: 1.403191\n",
      "(Iteration 2111 / 4900) loss: 1.326502\n",
      "(Iteration 2121 / 4900) loss: 1.502625\n",
      "(Iteration 2131 / 4900) loss: 1.389530\n",
      "(Iteration 2141 / 4900) loss: 1.304314\n",
      "(Iteration 2151 / 4900) loss: 1.452928\n",
      "(Iteration 2161 / 4900) loss: 1.553347\n",
      "(Iteration 2171 / 4900) loss: 1.342083\n",
      "(Iteration 2181 / 4900) loss: 1.402166\n",
      "(Iteration 2191 / 4900) loss: 1.431434\n",
      "(Iteration 2201 / 4900) loss: 1.542245\n",
      "(Iteration 2211 / 4900) loss: 1.482945\n",
      "(Iteration 2221 / 4900) loss: 1.297490\n",
      "(Iteration 2231 / 4900) loss: 1.556315\n",
      "(Iteration 2241 / 4900) loss: 1.440932\n",
      "(Iteration 2251 / 4900) loss: 1.511809\n",
      "(Iteration 2261 / 4900) loss: 1.386688\n",
      "(Iteration 2271 / 4900) loss: 1.327102\n",
      "(Iteration 2281 / 4900) loss: 1.403692\n",
      "(Iteration 2291 / 4900) loss: 1.257225\n",
      "(Iteration 2301 / 4900) loss: 1.612087\n",
      "(Iteration 2311 / 4900) loss: 1.438910\n",
      "(Iteration 2321 / 4900) loss: 1.414824\n",
      "(Iteration 2331 / 4900) loss: 1.591107\n",
      "(Iteration 2341 / 4900) loss: 1.275700\n",
      "(Iteration 2351 / 4900) loss: 1.467277\n",
      "(Iteration 2361 / 4900) loss: 1.481485\n",
      "(Iteration 2371 / 4900) loss: 1.254640\n",
      "(Iteration 2381 / 4900) loss: 1.434616\n",
      "(Iteration 2391 / 4900) loss: 1.506719\n",
      "(Iteration 2401 / 4900) loss: 1.386933\n",
      "(Iteration 2411 / 4900) loss: 1.486555\n",
      "(Iteration 2421 / 4900) loss: 1.468220\n",
      "(Iteration 2431 / 4900) loss: 1.074896\n",
      "(Iteration 2441 / 4900) loss: 1.206837\n",
      "(Epoch 5 / 10) train acc: 0.510000; val_acc: 0.482000\n",
      "(Iteration 2451 / 4900) loss: 1.429948\n",
      "(Iteration 2461 / 4900) loss: 1.384690\n",
      "(Iteration 2471 / 4900) loss: 1.376129\n",
      "(Iteration 2481 / 4900) loss: 1.333767\n",
      "(Iteration 2491 / 4900) loss: 1.527799\n",
      "(Iteration 2501 / 4900) loss: 1.397218\n",
      "(Iteration 2511 / 4900) loss: 1.337624\n",
      "(Iteration 2521 / 4900) loss: 1.642286\n",
      "(Iteration 2531 / 4900) loss: 1.444448\n",
      "(Iteration 2541 / 4900) loss: 1.356476\n",
      "(Iteration 2551 / 4900) loss: 1.344169\n",
      "(Iteration 2561 / 4900) loss: 1.262164\n",
      "(Iteration 2571 / 4900) loss: 1.311652\n",
      "(Iteration 2581 / 4900) loss: 1.474845\n",
      "(Iteration 2591 / 4900) loss: 1.348677\n",
      "(Iteration 2601 / 4900) loss: 1.337110\n",
      "(Iteration 2611 / 4900) loss: 1.396681\n",
      "(Iteration 2621 / 4900) loss: 1.625528\n",
      "(Iteration 2631 / 4900) loss: 1.318971\n",
      "(Iteration 2641 / 4900) loss: 1.269142\n",
      "(Iteration 2651 / 4900) loss: 1.425045\n",
      "(Iteration 2661 / 4900) loss: 1.429053\n",
      "(Iteration 2671 / 4900) loss: 1.432592\n",
      "(Iteration 2681 / 4900) loss: 1.444134\n",
      "(Iteration 2691 / 4900) loss: 1.367325\n",
      "(Iteration 2701 / 4900) loss: 1.366615\n",
      "(Iteration 2711 / 4900) loss: 1.128349\n",
      "(Iteration 2721 / 4900) loss: 1.472440\n",
      "(Iteration 2731 / 4900) loss: 1.247875\n",
      "(Iteration 2741 / 4900) loss: 1.176453\n",
      "(Iteration 2751 / 4900) loss: 1.511109\n",
      "(Iteration 2761 / 4900) loss: 1.342345\n",
      "(Iteration 2771 / 4900) loss: 1.208522\n",
      "(Iteration 2781 / 4900) loss: 1.268814\n",
      "(Iteration 2791 / 4900) loss: 1.379475\n",
      "(Iteration 2801 / 4900) loss: 1.218327\n",
      "(Iteration 2811 / 4900) loss: 1.463972\n",
      "(Iteration 2821 / 4900) loss: 1.258116\n",
      "(Iteration 2831 / 4900) loss: 1.071176\n",
      "(Iteration 2841 / 4900) loss: 1.368130\n",
      "(Iteration 2851 / 4900) loss: 1.240065\n",
      "(Iteration 2861 / 4900) loss: 1.370225\n",
      "(Iteration 2871 / 4900) loss: 1.223054\n",
      "(Iteration 2881 / 4900) loss: 1.395475\n",
      "(Iteration 2891 / 4900) loss: 1.586112\n",
      "(Iteration 2901 / 4900) loss: 1.417836\n",
      "(Iteration 2911 / 4900) loss: 1.409256\n",
      "(Iteration 2921 / 4900) loss: 1.249377\n",
      "(Iteration 2931 / 4900) loss: 1.502717\n",
      "(Epoch 6 / 10) train acc: 0.503000; val_acc: 0.479000\n",
      "(Iteration 2941 / 4900) loss: 1.299867\n",
      "(Iteration 2951 / 4900) loss: 1.408961\n",
      "(Iteration 2961 / 4900) loss: 1.470504\n",
      "(Iteration 2971 / 4900) loss: 1.233400\n",
      "(Iteration 2981 / 4900) loss: 1.344499\n",
      "(Iteration 2991 / 4900) loss: 1.149203\n",
      "(Iteration 3001 / 4900) loss: 1.247066\n",
      "(Iteration 3011 / 4900) loss: 1.324201\n",
      "(Iteration 3021 / 4900) loss: 1.562802\n",
      "(Iteration 3031 / 4900) loss: 1.416145\n",
      "(Iteration 3041 / 4900) loss: 1.412975\n",
      "(Iteration 3051 / 4900) loss: 1.238884\n",
      "(Iteration 3061 / 4900) loss: 1.311654\n",
      "(Iteration 3071 / 4900) loss: 1.041467\n",
      "(Iteration 3081 / 4900) loss: 1.124308\n",
      "(Iteration 3091 / 4900) loss: 1.354627\n",
      "(Iteration 3101 / 4900) loss: 1.217266\n",
      "(Iteration 3111 / 4900) loss: 1.257042\n",
      "(Iteration 3121 / 4900) loss: 1.335529\n",
      "(Iteration 3131 / 4900) loss: 1.376822\n",
      "(Iteration 3141 / 4900) loss: 1.441919\n",
      "(Iteration 3151 / 4900) loss: 1.177463\n",
      "(Iteration 3161 / 4900) loss: 1.322404\n",
      "(Iteration 3171 / 4900) loss: 1.273115\n",
      "(Iteration 3181 / 4900) loss: 1.314732\n",
      "(Iteration 3191 / 4900) loss: 1.254292\n",
      "(Iteration 3201 / 4900) loss: 1.452738\n",
      "(Iteration 3211 / 4900) loss: 1.327701\n",
      "(Iteration 3221 / 4900) loss: 1.236431\n",
      "(Iteration 3231 / 4900) loss: 1.357195\n",
      "(Iteration 3241 / 4900) loss: 1.220247\n",
      "(Iteration 3251 / 4900) loss: 1.242103\n",
      "(Iteration 3261 / 4900) loss: 1.310295\n",
      "(Iteration 3271 / 4900) loss: 1.367992\n",
      "(Iteration 3281 / 4900) loss: 1.373840\n",
      "(Iteration 3291 / 4900) loss: 1.221510\n",
      "(Iteration 3301 / 4900) loss: 1.355792\n",
      "(Iteration 3311 / 4900) loss: 1.493888\n",
      "(Iteration 3321 / 4900) loss: 1.251904\n",
      "(Iteration 3331 / 4900) loss: 1.350602\n",
      "(Iteration 3341 / 4900) loss: 1.289824\n",
      "(Iteration 3351 / 4900) loss: 1.485660\n",
      "(Iteration 3361 / 4900) loss: 1.336867\n",
      "(Iteration 3371 / 4900) loss: 1.456877\n",
      "(Iteration 3381 / 4900) loss: 1.333900\n",
      "(Iteration 3391 / 4900) loss: 1.380549\n",
      "(Iteration 3401 / 4900) loss: 1.361404\n",
      "(Iteration 3411 / 4900) loss: 1.247970\n",
      "(Iteration 3421 / 4900) loss: 1.195305\n",
      "(Epoch 7 / 10) train acc: 0.522000; val_acc: 0.461000\n",
      "(Iteration 3431 / 4900) loss: 1.253163\n",
      "(Iteration 3441 / 4900) loss: 1.220391\n",
      "(Iteration 3451 / 4900) loss: 1.231082\n",
      "(Iteration 3461 / 4900) loss: 1.286781\n",
      "(Iteration 3471 / 4900) loss: 1.174692\n",
      "(Iteration 3481 / 4900) loss: 1.342814\n",
      "(Iteration 3491 / 4900) loss: 1.540751\n",
      "(Iteration 3501 / 4900) loss: 1.145661\n",
      "(Iteration 3511 / 4900) loss: 1.239822\n",
      "(Iteration 3521 / 4900) loss: 1.386300\n",
      "(Iteration 3531 / 4900) loss: 1.289263\n",
      "(Iteration 3541 / 4900) loss: 1.540068\n",
      "(Iteration 3551 / 4900) loss: 1.251051\n",
      "(Iteration 3561 / 4900) loss: 1.370374\n",
      "(Iteration 3571 / 4900) loss: 1.131538\n",
      "(Iteration 3581 / 4900) loss: 1.257873\n",
      "(Iteration 3591 / 4900) loss: 1.234221\n",
      "(Iteration 3601 / 4900) loss: 1.138189\n",
      "(Iteration 3611 / 4900) loss: 1.355347\n",
      "(Iteration 3621 / 4900) loss: 1.484198\n",
      "(Iteration 3631 / 4900) loss: 1.373449\n",
      "(Iteration 3641 / 4900) loss: 1.442051\n",
      "(Iteration 3651 / 4900) loss: 1.518721\n",
      "(Iteration 3661 / 4900) loss: 1.186392\n",
      "(Iteration 3671 / 4900) loss: 1.227780\n",
      "(Iteration 3681 / 4900) loss: 1.416594\n",
      "(Iteration 3691 / 4900) loss: 1.362029\n",
      "(Iteration 3701 / 4900) loss: 1.307071\n",
      "(Iteration 3711 / 4900) loss: 1.381370\n",
      "(Iteration 3721 / 4900) loss: 1.419332\n",
      "(Iteration 3731 / 4900) loss: 1.341585\n",
      "(Iteration 3741 / 4900) loss: 1.474639\n",
      "(Iteration 3751 / 4900) loss: 1.236077\n",
      "(Iteration 3761 / 4900) loss: 1.253424\n",
      "(Iteration 3771 / 4900) loss: 1.329195\n",
      "(Iteration 3781 / 4900) loss: 1.346301\n",
      "(Iteration 3791 / 4900) loss: 1.417309\n",
      "(Iteration 3801 / 4900) loss: 1.246915\n",
      "(Iteration 3811 / 4900) loss: 1.275841\n",
      "(Iteration 3821 / 4900) loss: 1.201909\n",
      "(Iteration 3831 / 4900) loss: 1.267490\n",
      "(Iteration 3841 / 4900) loss: 1.462690\n",
      "(Iteration 3851 / 4900) loss: 1.284182\n",
      "(Iteration 3861 / 4900) loss: 1.280765\n",
      "(Iteration 3871 / 4900) loss: 1.427871\n",
      "(Iteration 3881 / 4900) loss: 1.385371\n",
      "(Iteration 3891 / 4900) loss: 1.349876\n",
      "(Iteration 3901 / 4900) loss: 1.361986\n",
      "(Iteration 3911 / 4900) loss: 1.160661\n",
      "(Epoch 8 / 10) train acc: 0.520000; val_acc: 0.510000\n",
      "(Iteration 3921 / 4900) loss: 1.445247\n",
      "(Iteration 3931 / 4900) loss: 1.462449\n",
      "(Iteration 3941 / 4900) loss: 1.322963\n",
      "(Iteration 3951 / 4900) loss: 1.320245\n",
      "(Iteration 3961 / 4900) loss: 1.192441\n",
      "(Iteration 3971 / 4900) loss: 1.167780\n",
      "(Iteration 3981 / 4900) loss: 1.652476\n",
      "(Iteration 3991 / 4900) loss: 1.425691\n",
      "(Iteration 4001 / 4900) loss: 1.453555\n",
      "(Iteration 4011 / 4900) loss: 1.498506\n",
      "(Iteration 4021 / 4900) loss: 1.301759\n",
      "(Iteration 4031 / 4900) loss: 1.253802\n",
      "(Iteration 4041 / 4900) loss: 1.307667\n",
      "(Iteration 4051 / 4900) loss: 1.201523\n",
      "(Iteration 4061 / 4900) loss: 1.545734\n",
      "(Iteration 4071 / 4900) loss: 1.109353\n",
      "(Iteration 4081 / 4900) loss: 1.288681\n",
      "(Iteration 4091 / 4900) loss: 1.487455\n",
      "(Iteration 4101 / 4900) loss: 1.318335\n",
      "(Iteration 4111 / 4900) loss: 1.482385\n",
      "(Iteration 4121 / 4900) loss: 1.288189\n",
      "(Iteration 4131 / 4900) loss: 1.162212\n",
      "(Iteration 4141 / 4900) loss: 1.252696\n",
      "(Iteration 4151 / 4900) loss: 1.366217\n",
      "(Iteration 4161 / 4900) loss: 1.271151\n",
      "(Iteration 4171 / 4900) loss: 1.346902\n",
      "(Iteration 4181 / 4900) loss: 1.428850\n",
      "(Iteration 4191 / 4900) loss: 1.460410\n",
      "(Iteration 4201 / 4900) loss: 1.177940\n",
      "(Iteration 4211 / 4900) loss: 1.340138\n",
      "(Iteration 4221 / 4900) loss: 1.381828\n",
      "(Iteration 4231 / 4900) loss: 1.223938\n",
      "(Iteration 4241 / 4900) loss: 1.277379\n",
      "(Iteration 4251 / 4900) loss: 1.263687\n",
      "(Iteration 4261 / 4900) loss: 1.283112\n",
      "(Iteration 4271 / 4900) loss: 1.403412\n",
      "(Iteration 4281 / 4900) loss: 1.414768\n",
      "(Iteration 4291 / 4900) loss: 1.093311\n",
      "(Iteration 4301 / 4900) loss: 1.303047\n",
      "(Iteration 4311 / 4900) loss: 1.356442\n",
      "(Iteration 4321 / 4900) loss: 1.283251\n",
      "(Iteration 4331 / 4900) loss: 1.362231\n",
      "(Iteration 4341 / 4900) loss: 1.276411\n",
      "(Iteration 4351 / 4900) loss: 0.961972\n",
      "(Iteration 4361 / 4900) loss: 1.248219\n",
      "(Iteration 4371 / 4900) loss: 1.349502\n",
      "(Iteration 4381 / 4900) loss: 1.261375\n",
      "(Iteration 4391 / 4900) loss: 1.430441\n",
      "(Iteration 4401 / 4900) loss: 1.201776\n",
      "(Epoch 9 / 10) train acc: 0.530000; val_acc: 0.499000\n",
      "(Iteration 4411 / 4900) loss: 1.558850\n",
      "(Iteration 4421 / 4900) loss: 1.368284\n",
      "(Iteration 4431 / 4900) loss: 1.185474\n",
      "(Iteration 4441 / 4900) loss: 1.280525\n",
      "(Iteration 4451 / 4900) loss: 1.286667\n",
      "(Iteration 4461 / 4900) loss: 1.344505\n",
      "(Iteration 4471 / 4900) loss: 1.117470\n",
      "(Iteration 4481 / 4900) loss: 1.258071\n",
      "(Iteration 4491 / 4900) loss: 1.246874\n",
      "(Iteration 4501 / 4900) loss: 1.150258\n",
      "(Iteration 4511 / 4900) loss: 1.324550\n",
      "(Iteration 4521 / 4900) loss: 1.000908\n",
      "(Iteration 4531 / 4900) loss: 1.247108\n",
      "(Iteration 4541 / 4900) loss: 1.001888\n",
      "(Iteration 4551 / 4900) loss: 1.419370\n",
      "(Iteration 4561 / 4900) loss: 1.388162\n",
      "(Iteration 4571 / 4900) loss: 1.376967\n",
      "(Iteration 4581 / 4900) loss: 1.137955\n",
      "(Iteration 4591 / 4900) loss: 1.240578\n",
      "(Iteration 4601 / 4900) loss: 1.516793\n",
      "(Iteration 4611 / 4900) loss: 1.261544\n",
      "(Iteration 4621 / 4900) loss: 1.379268\n",
      "(Iteration 4631 / 4900) loss: 1.264970\n",
      "(Iteration 4641 / 4900) loss: 1.326048\n",
      "(Iteration 4651 / 4900) loss: 1.275987\n",
      "(Iteration 4661 / 4900) loss: 1.189246\n",
      "(Iteration 4671 / 4900) loss: 1.332017\n",
      "(Iteration 4681 / 4900) loss: 1.279221\n",
      "(Iteration 4691 / 4900) loss: 1.285809\n",
      "(Iteration 4701 / 4900) loss: 1.218577\n",
      "(Iteration 4711 / 4900) loss: 1.308963\n",
      "(Iteration 4721 / 4900) loss: 1.225454\n",
      "(Iteration 4731 / 4900) loss: 1.191841\n",
      "(Iteration 4741 / 4900) loss: 1.291613\n",
      "(Iteration 4751 / 4900) loss: 1.373882\n",
      "(Iteration 4761 / 4900) loss: 1.194951\n",
      "(Iteration 4771 / 4900) loss: 1.195520\n",
      "(Iteration 4781 / 4900) loss: 1.405788\n",
      "(Iteration 4791 / 4900) loss: 1.376422\n",
      "(Iteration 4801 / 4900) loss: 1.288391\n",
      "(Iteration 4811 / 4900) loss: 1.440267\n",
      "(Iteration 4821 / 4900) loss: 1.445503\n",
      "(Iteration 4831 / 4900) loss: 1.241860\n",
      "(Iteration 4841 / 4900) loss: 1.223117\n",
      "(Iteration 4851 / 4900) loss: 1.279437\n",
      "(Iteration 4861 / 4900) loss: 1.192344\n",
      "(Iteration 4871 / 4900) loss: 1.405768\n",
      "(Iteration 4881 / 4900) loss: 1.448201\n",
      "(Iteration 4891 / 4900) loss: 1.119873\n",
      "(Epoch 10 / 10) train acc: 0.548000; val_acc: 0.520000\n"
     ]
    }
   ],
   "source": [
    "nput_size = 32 * 32 * 3\n",
    "hidden_size = 50\n",
    "num_classes = 10\n",
    "model = TwoLayerNet(input_size, hidden_size, num_classes)\n",
    "solver = None\n",
    "\n",
    "from cs231n.optim import sgd\n",
    "solver = Solver(model, data, update_rule='sgd', \n",
    "                optim_config={'learning_rate': 1e-3}\n",
    "                )\n",
    "solver.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 64400,
     "status": "ok",
     "timestamp": 1626360547531,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "-Jpobz0Q8hzM",
    "outputId": "8858d080-66a2-4d08-b262-de5fe12cb578",
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(Iteration 1 / 4900) loss: 2.303642\n",
      "(Epoch 0 / 10) train acc: 0.211000; val_acc: 0.197000\n",
      "(Iteration 201 / 4900) loss: 1.536567\n",
      "(Iteration 401 / 4900) loss: 1.625281\n",
      "(Epoch 1 / 10) train acc: 0.461000; val_acc: 0.461000\n",
      "(Iteration 601 / 4900) loss: 1.363805\n",
      "(Iteration 801 / 4900) loss: 1.458114\n",
      "(Epoch 2 / 10) train acc: 0.500000; val_acc: 0.473000\n",
      "(Iteration 1001 / 4900) loss: 1.473016\n",
      "(Iteration 1201 / 4900) loss: 1.404041\n",
      "(Iteration 1401 / 4900) loss: 1.242607\n",
      "(Epoch 3 / 10) train acc: 0.514000; val_acc: 0.481000\n",
      "(Iteration 1601 / 4900) loss: 1.408209\n",
      "(Iteration 1801 / 4900) loss: 1.468805\n",
      "(Epoch 4 / 10) train acc: 0.521000; val_acc: 0.475000\n",
      "(Iteration 2001 / 4900) loss: 1.330417\n",
      "(Iteration 2201 / 4900) loss: 1.212162\n",
      "(Iteration 2401 / 4900) loss: 1.344331\n",
      "(Epoch 5 / 10) train acc: 0.527000; val_acc: 0.493000\n",
      "(Iteration 2601 / 4900) loss: 1.440301\n",
      "(Iteration 2801 / 4900) loss: 1.387977\n",
      "(Epoch 6 / 10) train acc: 0.554000; val_acc: 0.488000\n",
      "(Iteration 3001 / 4900) loss: 1.287681\n",
      "(Iteration 3201 / 4900) loss: 1.070210\n",
      "(Iteration 3401 / 4900) loss: 1.153304\n",
      "(Epoch 7 / 10) train acc: 0.576000; val_acc: 0.496000\n",
      "(Iteration 3601 / 4900) loss: 1.158098\n",
      "(Iteration 3801 / 4900) loss: 1.146596\n",
      "(Epoch 8 / 10) train acc: 0.550000; val_acc: 0.487000\n",
      "(Iteration 4001 / 4900) loss: 1.271845\n",
      "(Iteration 4201 / 4900) loss: 1.228656\n",
      "(Iteration 4401 / 4900) loss: 1.141818\n",
      "(Epoch 9 / 10) train acc: 0.544000; val_acc: 0.505000\n",
      "(Iteration 4601 / 4900) loss: 1.109155\n",
      "(Iteration 4801 / 4900) loss: 1.163082\n",
      "(Epoch 10 / 10) train acc: 0.590000; val_acc: 0.498000\n"
     ]
    }
   ],
   "source": [
    "best_model = None\n",
    "\n",
    "\n",
    "#################################################################################\n",
    "# TODO: Tune hyperparameters using the validation set. Store your best trained  #\n",
    "# model in best_model.                                                          #\n",
    "#                                                                               #\n",
    "# To help debug your network, it may help to use visualizations similar to the  #\n",
    "# ones we used above; these visualizations will have significant qualitative    #\n",
    "# differences from the ones we saw above for the poorly tuned network.          #\n",
    "#                                                                               #\n",
    "# Tweaking hyperparameters by hand can be fun, but you might find it useful to  #\n",
    "# write code to sweep through possible combinations of hyperparameters          #\n",
    "# automatically like we did on thexs previous exercises.                          #\n",
    "#################################################################################\n",
    "# *****START OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****\n",
    "input_size = 32 * 32 * 3\n",
    "hidden_dims = [70]\n",
    "num_classes = 10\n",
    "\n",
    "from cs231n.optim import adam\n",
    "new_model = TwoLayerNet(input_size, hidden_size, num_classes)\n",
    "solver = Solver(new_model, data,\n",
    "                    update_rule='adam',\n",
    "                    optim_config={\n",
    "                      'learning_rate': 1e-4\n",
    "                    },\n",
    "                    num_epochs=10,\n",
    "                    print_every=200\n",
    "                    )\n",
    "solver.train()\n",
    "\n",
    "best_model = new_model\n",
    "\n",
    "# *****END OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****\n",
    "################################################################################\n",
    "#                              END OF YOUR CODE                                #\n",
    "################################################################################"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "EtepCtNP8hzM"
   },
   "source": [
    "# Test your model!\n",
    "Run your best model on the validation and test sets. You should achieve above 48% accuracy on the validation set and the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 263,
     "status": "ok",
     "timestamp": 1626360552063,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "val_accuracy",
    "outputId": "36a794f7-2f67-4ba6-9944-165f38f84c65"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Validation set accuracy:  0.505\n"
     ]
    }
   ],
   "source": [
    "y_val_pred = np.argmax(best_model.loss(data['X_val']), axis=1)\n",
    "print('Validation set accuracy: ', (y_val_pred == data['y_val']).mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 254,
     "status": "ok",
     "timestamp": 1626360555967,
     "user": {
      "displayName": "이재민",
      "photoUrl": "",
      "userId": "14376317564121567982"
     },
     "user_tz": -540
    },
    "id": "test_accuracy",
    "outputId": "7dcbc80d-ee10-486f-e747-b1d439b8d302"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set accuracy:  0.5\n"
     ]
    }
   ],
   "source": [
    "y_test_pred = np.argmax(best_model.loss(data['X_test']), axis=1)\n",
    "print('Test set accuracy: ', (y_test_pred == data['y_test']).mean())"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "two_layer_net.ipynb",
   "provenance": []
  },
  "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}