{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# MNIST-Neural Network-Batch Normalization and Weight Decay" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": false }, "outputs": [], "source": [ "# coding: utf-8\n", "import sys, os\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import math\n", "\n", "sys.path.append(os.pardir)\n", "from deeplink.mnist import *\n", "from deeplink.networks import *" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Multilayer Neural Network Model (Two Hidden Layers) and Learing/Validation" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Initializers " ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# coding: utf-8\n", "import numpy as np\n", "\n", "class Initializer:\n", " def __init__(self, params, params_size_list, use_batch_normalization=False):\n", " self.params = params\n", " self.params_size_list = params_size_list\n", " self.use_batch_normalization = use_batch_normalization\n", "\n", " def initialize_params(self):\n", " pass\n", "\n", " def get_params(self):\n", " return self.params\n", "\n", "\n", "class Zero_Initializer(Initializer):\n", " def initialize_params(self, use_batch_normalization):\n", " for idx in range(1, len(self.params_size_list)):\n", " self.params['W' + str(idx)] = np.zeros(self.params_size_list[idx - 1], self.params_size_list[idx])\n", " self.params['b' + str(idx)] = np.zeros(self.params_size_list[idx])\n", " if self.use_batch_normalization and idx < len(self.params_size_list) - 1:\n", " self.params['gamma' + str(idx)] = np.zeros(self.params_size_list[idx])\n", " self.params['beta' + str(idx)] = np.zeros(self.params_size_list[idx])\n", "\n", "class N1_Initializer(Initializer):\n", " def initialize_params(self):\n", " for idx in range(1, len(self.params_size_list)):\n", " self.params['W' + str(idx)] = np.random.randn(self.params_size_list[idx - 1], self.params_size_list[idx])\n", " self.params['b' + str(idx)] = np.random.randn(self.params_size_list[idx])\n", " if self.use_batch_normalization and idx < len(self.params_size_list) - 1:\n", " self.params['gamma' + str(idx)] = np.random.randn(self.params_size_list[idx])\n", " self.params['beta' + str(idx)] = np.random.randn(self.params_size_list[idx])\n", "\n", "class N2_Initializer(Initializer):\n", " def initialize_params(self):\n", " for idx in range(1, len(self.params_size_list)):\n", " self.params['W' + str(idx)] = np.random.randn(self.params_size_list[idx - 1], self.params_size_list[idx]) * 0.01\n", " self.params['b' + str(idx)] = np.random.randn(self.params_size_list[idx]) * 0.01\n", " if self.use_batch_normalization and idx < len(self.params_size_list) - 1:\n", " self.params['gamma' + str(idx)] = np.random.randn(self.params_size_list[idx]) * 0.01\n", " self.params['beta' + str(idx)] = np.random.randn(self.params_size_list[idx]) * 0.01\n", "\n", "class Xavier_Initializer(Initializer):\n", " def initialize_params(self):\n", " for idx in range(1, len(self.params_size_list)):\n", " self.params['W' + str(idx)] = np.random.randn(self.params_size_list[idx - 1], self.params_size_list[idx]) / np.sqrt(self.params_size_list[idx - 1])\n", " self.params['b' + str(idx)] = np.random.randn(self.params_size_list[idx]) / np.sqrt(self.params_size_list[idx - 1])\n", " if self.use_batch_normalization and idx < len(self.params_size_list) - 1:\n", " self.params['gamma' + str(idx)] = np.random.randn(self.params_size_list[idx]) / np.sqrt(self.params_size_list[idx - 1])\n", " self.params['beta' + str(idx)] = np.random.randn(self.params_size_list[idx]) / np.sqrt(self.params_size_list[idx - 1])\n", "\n", "\n", "class He_Initializer(Initializer):\n", " def initialize_params(self):\n", " for idx in range(1, len(self.params_size_list)):\n", " self.params['W' + str(idx)] = np.random.randn(self.params_size_list[idx - 1], self.params_size_list[idx]) * np.sqrt(2) / np.sqrt(self.params_size_list[idx - 1])\n", " self.params['b' + str(idx)] = np.random.randn(self.params_size_list[idx]) * np.sqrt(2) / np.sqrt(self.params_size_list[idx - 1])\n", " if self.use_batch_normalization and idx < len(self.params_size_list) - 1:\n", " self.params['gamma' + str(idx)] = np.random.randn(self.params_size_list[idx]) * np.sqrt(2) / np.sqrt(self.params_size_list[idx - 1])\n", " self.params['beta' + str(idx)] = np.random.randn(self.params_size_list[idx]) * np.sqrt(2) / np.sqrt(self.params_size_list[idx - 1])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### New Layer - Batch Normalization" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "collapsed": true }, "outputs": [], "source": [ "class BatchNormalization:\n", " def __init__(self, gamma, beta, momentum=0.9, running_mean=None, running_var=None):\n", " self.gamma = gamma\n", " self.beta = beta\n", " self.momentum = momentum\n", " self.input_shape = None\n", "\n", " self.running_mean = running_mean\n", " self.running_var = running_var \n", " \n", " self.batch_size = None\n", " self.xc = None\n", " self.std = None\n", " self.dgamma = None\n", " self.dbeta = None\n", "\n", " def forward(self, x, is_train=True):\n", " self.input_shape = x.shape\n", " if x.ndim != 2:\n", " N, C, H, W = x.shape\n", " x = x.reshape(N, -1)\n", "\n", " out = self.__forward(x, is_train)\n", " \n", " return out.reshape(*self.input_shape)\n", " \n", " def __forward(self, x, is_train):\n", " if self.running_mean is None:\n", " N, D = x.shape\n", " self.running_mean = np.zeros(D)\n", " self.running_var = np.zeros(D)\n", " \n", " if is_train:\n", " mu = x.mean(axis=0)\n", " xc = x - mu\n", " var = np.mean(xc**2, axis=0)\n", " std = np.sqrt(var + 10e-7)\n", " xn = xc / std\n", " \n", " self.batch_size = x.shape[0]\n", " self.xc = xc\n", " self.xn = xn\n", " self.std = std\n", " self.running_mean = self.momentum * self.running_mean + (1-self.momentum) * mu\n", " self.running_var = self.momentum * self.running_var + (1-self.momentum) * var \n", " else:\n", " xc = x - self.running_mean\n", " xn = xc / ((np.sqrt(self.running_var + 10e-7)))\n", " \n", " out = self.gamma * xn + self.beta \n", " return out\n", "\n", " def backward(self, dout):\n", " if dout.ndim != 2:\n", " N, C, H, W = dout.shape\n", " dout = dout.reshape(N, -1)\n", "\n", " dx = self.__backward(dout)\n", "\n", " dx = dx.reshape(*self.input_shape)\n", " return dx\n", "\n", " def __backward(self, dout):\n", " dbeta = dout.sum(axis=0)\n", " dgamma = np.sum(self.xn * dout, axis=0)\n", " dxn = self.gamma * dout\n", " dxc = dxn / self.std\n", " dstd = -np.sum((dxn * self.xc) / (self.std * self.std), axis=0)\n", " dvar = 0.5 * dstd / self.std\n", " dxc += (2.0 / self.batch_size) * self.xc * dvar\n", " dmu = np.sum(dxc, axis=0)\n", " dx = dxc - dmu / self.batch_size\n", " \n", " self.dgamma = dgamma\n", " self.dbeta = dbeta\n", " \n", " return dx" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "collapsed": false }, "outputs": [], "source": [ "activation_layers = {\n", " 'Sigmoid': Sigmoid,\n", " 'ReLU': ReLU\n", "}\n", "\n", "optimizers = {\n", " \"SGD\": SGD,\n", " \"Momentum\": Momentum,\n", " \"Nesterov\": Nesterov,\n", " \"AdaGrad\": AdaGrad,\n", " \"RMSprop\": RMSprop,\n", " \"Adam\": Adam\n", "}\n", "\n", "initializers = {\n", " 'Zero': Zero_Initializer,\n", " 'N1': N1_Initializer,\n", " 'N2': N2_Initializer, # We will use this as a new initializer for supporting Batch Normalization\n", " 'Xavier': Xavier_Initializer,\n", " 'He': He_Initializer\n", "}" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Multi Layer Model Class" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": false }, "outputs": [], "source": [ "class MultiLayerNetExtended(MultiLayerNet):\n", " def __init__(self, input_size, hidden_size_list, output_size, activation='ReLU', initializer='N2', \n", " optimizer='AdaGrad', learning_rate=0.01, \n", " use_batch_normalization=False, \n", " use_weight_decay=False, weight_decay_lambda=0.0):\n", " self.input_size = input_size\n", " self.output_size = output_size\n", " self.hidden_size_list = hidden_size_list\n", " self.hidden_layer_num = len(hidden_size_list)\n", " \n", " self.use_batch_normalization = use_batch_normalization\n", "\n", " self.use_weight_decay = use_weight_decay\n", " self.weight_decay_lambda = weight_decay_lambda\n", " \n", " # Weight Initialization\n", " self.params = {}\n", " self.weight_initialization(initializer)\n", " \n", " # Layering\n", " self.layers = OrderedDict()\n", " self.last_layer = None\n", " self.layering(activation)\n", "\n", " # Optimization Method\n", " self.optimizer = optimizers[optimizer](lr=learning_rate)\n", " \n", " def weight_initialization(self, initializer):\n", " params_size_list = [self.input_size] + self.hidden_size_list + [self.output_size]\n", " initializer_obj = initializers[initializer](self.params, \n", " params_size_list, \n", " self.use_batch_normalization)\n", " initializer_obj.initialize_params();\n", " \n", " def layering(self, activation):\n", " for idx in range(1, self.hidden_layer_num + 1):\n", " self.layers['Affine' + str(idx)] = Affine(self.params['W' + str(idx)], self.params['b' + str(idx)])\n", " if self.use_batch_normalization:\n", " self.layers['Batch_Normalization' + str(idx)] = BatchNormalization(self.params['gamma' + str(idx)], \n", " self.params['beta' + str(idx)])\n", " self.layers['Activation' + str(idx)] = activation_layers[activation]()\n", "\n", " idx = self.hidden_layer_num + 1\n", " self.layers['Affine' + str(idx)] = Affine(self.params['W' + str(idx)], self.params['b' + str(idx)])\n", "\n", " self.last_layer = SoftmaxWithCrossEntropyLoss() \n", "\n", " def predict(self, x, is_train=False):\n", " for key, layer in self.layers.items():\n", " if \"BatchNorm\" in key:\n", " x = layer.forward(x, is_train)\n", " else:\n", " x = layer.forward(x)\n", " return x\n", "\n", " def loss(self, x, t, is_train=False):\n", " y = self.predict(x, is_train)\n", "\n", " if self.use_weight_decay:\n", " weight_decay = 0.0\n", " for idx in range(1, self.hidden_layer_num + 2):\n", " W = self.params['W' + str(idx)]\n", " weight_decay += 0.5 * self.weight_decay_lambda * np.sum(W**2)\n", " return self.last_layer.forward(y, t) + weight_decay\n", " else:\n", " return self.last_layer.forward(y, t)\n", "\n", " def accuracy(self, x, t):\n", " y = self.predict(x, is_train=False)\n", " y = np.argmax(y, axis=1)\n", " if t.ndim != 1 : t = np.argmax(t, axis=1)\n", "\n", " accuracy = np.sum(y == t) / float(x.shape[0])\n", " return accuracy \n", "\n", " def backpropagation_gradient(self, x, t):\n", " # forward\n", " self.loss(x, t, is_train=True)\n", "\n", " # backward\n", " din = 1\n", " din = self.last_layer.backward(din)\n", "\n", " layers = list(self.layers.values())\n", " layers.reverse()\n", " for layer in layers:\n", " din = layer.backward(din)\n", "\n", " grads = {}\n", " for idx in range(1, self.hidden_layer_num + 2):\n", " if self.use_weight_decay:\n", " grads['W' + str(idx)] = self.layers['Affine' + str(idx)].dW + self.weight_decay_lambda * self.params['W' + str(idx)]\n", " else:\n", " grads['W' + str(idx)] = self.layers['Affine' + str(idx)].dW\n", " grads['b' + str(idx)] = self.layers['Affine' + str(idx)].db\n", "\n", " if self.use_batch_normalization and idx <= self.hidden_layer_num:\n", " grads['gamma' + str(idx)] = self.layers['Batch_Normalization' + str(idx)].dgamma\n", " grads['beta' + str(idx)] = self.layers['Batch_Normalization' + str(idx)].dbeta\n", " \n", " return grads\n", "\n", " def learning(self, x_batch, t_batch):\n", " grads = self.backpropagation_gradient(x_batch, t_batch)\n", " self.optimizer.update(self.params, grads)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Training and Evaluation" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": false }, "outputs": [], "source": [ "data = mnist_data(\"/Users/yhhan/git/aiclass/0.Professor/data/MNIST_data/.\")\n", "(img_train, label_train), (img_validation, label_validation), (img_test, label_test) = data.load_mnist(flatten=True, normalize=True, one_hot_label=True)\n", "\n", "input_size=784\n", "hidden_layer1_size=128\n", "hidden_layer2_size=128\n", "output_size=10\n", "\n", "num_epochs = 50\n", "train_size = img_train.shape[0]\n", "batch_size = 1000\n", "learning_rate = 0.1\n", "\n", "markers = {\"N2, AdaGrad, No_Batch_Norm, lambda=0.0\": \"+\", \n", " \"N2, AdaGrad, Batch_Norm, lambda=0.0\": \"*\", \n", " \"N2, AdaGrad, Batch_Norm, lambda=0.1\": \"o\",\n", " \"He, AdaGrad, No_Batch_Norm, lambda=0.0\": \"x\", \n", " \"He, AdaGrad, Batch_Norm, lambda=0.0\": \"h\", \n", " \"He, AdaGrad, Batch_Norm, lambda=0.1\": \"H\"}\n", "\n", "networks = {}\n", "train_errors = {}\n", "validation_errors = {}\n", "test_accuracy_values = {}\n", "max_test_accuracy_epoch = {}\n", "max_test_accuracy_value = {}\n", "\n", "for key in markers.keys():\n", " if key == \"N2, AdaGrad, No_Batch_Norm, lambda=0.0\":\n", " networks[key] = MultiLayerNetExtended(input_size, [hidden_layer1_size, hidden_layer2_size], output_size, \n", " activation='ReLU', \n", " initializer='N2',\n", " optimizer='AdaGrad', learning_rate=learning_rate,\n", " use_batch_normalization=False, \n", " use_weight_decay=False, weight_decay_lambda=0.0)\n", " elif key == \"N2, AdaGrad, Batch_Norm, lambda=0.0\":\n", " networks[key] = MultiLayerNetExtended(input_size, [hidden_layer1_size, hidden_layer2_size], output_size, \n", " activation='ReLU', \n", " initializer='N2',\n", " optimizer='AdaGrad', learning_rate=learning_rate,\n", " use_batch_normalization=True,\n", " use_weight_decay=False, weight_decay_lambda=0.0)\n", " elif key == \"N2, AdaGrad, Batch_Norm, lambda=0.1\":\n", " networks[key] = MultiLayerNetExtended(input_size, [hidden_layer1_size, hidden_layer2_size], output_size, \n", " activation='ReLU', \n", " initializer='N2',\n", " optimizer='AdaGrad', learning_rate=learning_rate,\n", " use_batch_normalization=True,\n", " use_weight_decay=True, weight_decay_lambda=0.1) \n", " elif key == \"He, AdaGrad, No_Batch_Norm, lambda=0.0\":\n", " networks[key] = MultiLayerNetExtended(input_size, [hidden_layer1_size, hidden_layer2_size], output_size, \n", " activation='ReLU', \n", " initializer='He',\n", " optimizer='AdaGrad', learning_rate=learning_rate,\n", " use_batch_normalization=False,\n", " use_weight_decay=False, weight_decay_lambda=0.0)\n", " elif key == \"He, AdaGrad, Batch_Norm, lambda=0.0\":\n", " networks[key] = MultiLayerNetExtended(input_size, [hidden_layer1_size, hidden_layer2_size], output_size, \n", " activation='ReLU', \n", " initializer='He',\n", " optimizer='AdaGrad', learning_rate=learning_rate,\n", " use_batch_normalization=True,\n", " use_weight_decay=False, weight_decay_lambda=0.0)\n", " elif key == \"He, AdaGrad, Batch_Norm, lambda=0.1\":\n", " networks[key] = MultiLayerNetExtended(input_size, [hidden_layer1_size, hidden_layer2_size], output_size, \n", " activation='ReLU', \n", " initializer='He',\n", " optimizer='AdaGrad', learning_rate=learning_rate,\n", " use_batch_normalization=True,\n", " use_weight_decay=True, weight_decay_lambda=0.1) \n", " \n", " train_errors[key] = [] \n", " validation_errors[key] = []\n", " test_accuracy_values[key] = []\n", " max_test_accuracy_epoch[key] = 0\n", " max_test_accuracy_value[key] = 0.0" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 0, Train Err.:0.29420, Validation Err.:0.27180, Test Accuracy:0.90270, Max Test Accuracy:0.90270\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 0, Train Err.:0.09101, Validation Err.:0.11990, Test Accuracy:0.95630, Max Test Accuracy:0.95630\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 0, Train Err.:1.17575, Validation Err.:1.21322, Test Accuracy:0.77510, Max Test Accuracy:0.77510\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 0, Train Err.:0.24563, Validation Err.:0.21871, Test Accuracy:0.91570, Max Test Accuracy:0.91570\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 0, Train Err.:0.07612, Validation Err.:0.11275, Test Accuracy:0.95690, Max Test Accuracy:0.95690\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 0, Train Err.:1.06066, Validation Err.:1.08043, Test Accuracy:0.81700, Max Test Accuracy:0.81700\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 1, Train Err.:0.18584, Validation Err.:0.19378, Test Accuracy:0.93030, Max Test Accuracy:0.93030\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 1, Train Err.:0.04513, Validation Err.:0.09069, Test Accuracy:0.96600, Max Test Accuracy:0.96600\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 1, Train Err.:0.76966, Validation Err.:0.80488, Test Accuracy:0.87300, Max Test Accuracy:0.87300\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 1, Train Err.:0.15998, Validation Err.:0.15808, Test Accuracy:0.93800, Max Test Accuracy:0.93800\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 1, Train Err.:0.03489, Validation Err.:0.08802, Test Accuracy:0.96620, Max Test Accuracy:0.96620\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 1, Train Err.:0.78362, Validation Err.:0.80771, Test Accuracy:0.87490, Max Test Accuracy:0.87490\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 2, Train Err.:0.13651, Validation Err.:0.17047, Test Accuracy:0.93700, Max Test Accuracy:0.93700\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 2, Train Err.:0.02978, Validation Err.:0.08239, Test Accuracy:0.97130, Max Test Accuracy:0.97130\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 2, Train Err.:0.57870, Validation Err.:0.62202, Test Accuracy:0.90760, Max Test Accuracy:0.90760\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 2, Train Err.:0.12686, Validation Err.:0.13883, Test Accuracy:0.94540, Max Test Accuracy:0.94540\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 2, Train Err.:0.02114, Validation Err.:0.07969, Test Accuracy:0.97090, Max Test Accuracy:0.97090\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 2, Train Err.:0.63156, Validation Err.:0.64712, Test Accuracy:0.89290, Max Test Accuracy:0.89290\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 3, Train Err.:0.11136, Validation Err.:0.15961, Test Accuracy:0.94220, Max Test Accuracy:0.94220\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 3, Train Err.:0.02140, Validation Err.:0.07951, Test Accuracy:0.97220, Max Test Accuracy:0.97220\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 3, Train Err.:0.50559, Validation Err.:0.53505, Test Accuracy:0.91840, Max Test Accuracy:0.91840\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 3, Train Err.:0.10503, Validation Err.:0.13074, Test Accuracy:0.94950, Max Test Accuracy:0.94950\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 3, Train Err.:0.01383, Validation Err.:0.07697, Test Accuracy:0.97280, Max Test Accuracy:0.97280\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 3, Train Err.:0.53976, Validation Err.:0.55938, Test Accuracy:0.90210, Max Test Accuracy:0.90210\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 4, Train Err.:0.09677, Validation Err.:0.15059, Test Accuracy:0.94720, Max Test Accuracy:0.94720\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 4, Train Err.:0.01558, Validation Err.:0.08008, Test Accuracy:0.97440, Max Test Accuracy:0.97440\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 4, Train Err.:0.46726, Validation Err.:0.49332, Test Accuracy:0.91740, Max Test Accuracy:0.91840\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 4, Train Err.:0.09079, Validation Err.:0.12461, Test Accuracy:0.95370, Max Test Accuracy:0.95370\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 4, Train Err.:0.01008, Validation Err.:0.07600, Test Accuracy:0.97380, Max Test Accuracy:0.97380\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 4, Train Err.:0.49064, Validation Err.:0.51690, Test Accuracy:0.90960, Max Test Accuracy:0.90960\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 5, Train Err.:0.08653, Validation Err.:0.14609, Test Accuracy:0.95010, Max Test Accuracy:0.95010\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 5, Train Err.:0.01040, Validation Err.:0.08287, Test Accuracy:0.97530, Max Test Accuracy:0.97530\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 5, Train Err.:0.43375, Validation Err.:0.47160, Test Accuracy:0.92680, Max Test Accuracy:0.92680\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 5, Train Err.:0.08114, Validation Err.:0.12357, Test Accuracy:0.95470, Max Test Accuracy:0.95470\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 5, Train Err.:0.00715, Validation Err.:0.07806, Test Accuracy:0.97380, Max Test Accuracy:0.97380\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 5, Train Err.:0.46675, Validation Err.:0.48959, Test Accuracy:0.91550, Max Test Accuracy:0.91550\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 6, Train Err.:0.07786, Validation Err.:0.14373, Test Accuracy:0.95180, Max Test Accuracy:0.95180\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 6, Train Err.:0.00665, Validation Err.:0.08641, Test Accuracy:0.97520, Max Test Accuracy:0.97530\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 6, Train Err.:0.42105, Validation Err.:0.45514, Test Accuracy:0.92440, Max Test Accuracy:0.92680\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 6, Train Err.:0.07369, Validation Err.:0.12240, Test Accuracy:0.95640, Max Test Accuracy:0.95640\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 6, Train Err.:0.00540, Validation Err.:0.07887, Test Accuracy:0.97410, Max Test Accuracy:0.97410\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 6, Train Err.:0.44519, Validation Err.:0.47436, Test Accuracy:0.92050, Max Test Accuracy:0.92050\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 7, Train Err.:0.07099, Validation Err.:0.13986, Test Accuracy:0.95340, Max Test Accuracy:0.95340\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 7, Train Err.:0.00474, Validation Err.:0.09089, Test Accuracy:0.97570, Max Test Accuracy:0.97570\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 7, Train Err.:0.40149, Validation Err.:0.43570, Test Accuracy:0.92450, Max Test Accuracy:0.92680\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 7, Train Err.:0.06394, Validation Err.:0.11814, Test Accuracy:0.95890, Max Test Accuracy:0.95890\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 7, Train Err.:0.00421, Validation Err.:0.08036, Test Accuracy:0.97460, Max Test Accuracy:0.97460\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 7, Train Err.:0.40480, Validation Err.:0.44352, Test Accuracy:0.92420, Max Test Accuracy:0.92420\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 8, Train Err.:0.06634, Validation Err.:0.13857, Test Accuracy:0.95490, Max Test Accuracy:0.95490\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 8, Train Err.:0.00367, Validation Err.:0.09489, Test Accuracy:0.97560, Max Test Accuracy:0.97570\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 8, Train Err.:0.38733, Validation Err.:0.42702, Test Accuracy:0.92470, Max Test Accuracy:0.92680\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 8, Train Err.:0.05801, Validation Err.:0.11635, Test Accuracy:0.96050, Max Test Accuracy:0.96050\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 8, Train Err.:0.00336, Validation Err.:0.08210, Test Accuracy:0.97450, Max Test Accuracy:0.97460\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 8, Train Err.:0.38972, Validation Err.:0.42517, Test Accuracy:0.92570, Max Test Accuracy:0.92570\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 9, Train Err.:0.06232, Validation Err.:0.13746, Test Accuracy:0.95530, Max Test Accuracy:0.95530\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 9, Train Err.:0.00282, Validation Err.:0.09751, Test Accuracy:0.97520, Max Test Accuracy:0.97570\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 9, Train Err.:0.36445, Validation Err.:0.41003, Test Accuracy:0.92980, Max Test Accuracy:0.92980\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 9, Train Err.:0.05172, Validation Err.:0.11300, Test Accuracy:0.96190, Max Test Accuracy:0.96190\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 9, Train Err.:0.00272, Validation Err.:0.08507, Test Accuracy:0.97490, Max Test Accuracy:0.97490\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 9, Train Err.:0.38392, Validation Err.:0.41876, Test Accuracy:0.92540, Max Test Accuracy:0.92570\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 10, Train Err.:0.06004, Validation Err.:0.13731, Test Accuracy:0.95630, Max Test Accuracy:0.95630\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 10, Train Err.:0.00224, Validation Err.:0.09908, Test Accuracy:0.97550, Max Test Accuracy:0.97570\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 10, Train Err.:0.34985, Validation Err.:0.39745, Test Accuracy:0.92970, Max Test Accuracy:0.92980\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 10, Train Err.:0.04633, Validation Err.:0.11245, Test Accuracy:0.96250, Max Test Accuracy:0.96250\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 10, Train Err.:0.00211, Validation Err.:0.08720, Test Accuracy:0.97500, Max Test Accuracy:0.97500\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 10, Train Err.:0.37163, Validation Err.:0.40899, Test Accuracy:0.92710, Max Test Accuracy:0.92710\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 11, Train Err.:0.05818, Validation Err.:0.13716, Test Accuracy:0.95700, Max Test Accuracy:0.95700\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 11, Train Err.:0.00182, Validation Err.:0.10058, Test Accuracy:0.97610, Max Test Accuracy:0.97610\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 11, Train Err.:0.35032, Validation Err.:0.39271, Test Accuracy:0.92890, Max Test Accuracy:0.92980\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 11, Train Err.:0.04288, Validation Err.:0.11264, Test Accuracy:0.96290, Max Test Accuracy:0.96290\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 11, Train Err.:0.00184, Validation Err.:0.09045, Test Accuracy:0.97570, Max Test Accuracy:0.97570\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 11, Train Err.:0.35933, Validation Err.:0.40072, Test Accuracy:0.92710, Max Test Accuracy:0.92710\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 12, Train Err.:0.05542, Validation Err.:0.13980, Test Accuracy:0.95750, Max Test Accuracy:0.95750\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 12, Train Err.:0.00151, Validation Err.:0.10199, Test Accuracy:0.97660, Max Test Accuracy:0.97660\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 12, Train Err.:0.34691, Validation Err.:0.38471, Test Accuracy:0.93040, Max Test Accuracy:0.93040\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 12, Train Err.:0.03984, Validation Err.:0.10996, Test Accuracy:0.96310, Max Test Accuracy:0.96310\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 12, Train Err.:0.00156, Validation Err.:0.09259, Test Accuracy:0.97620, Max Test Accuracy:0.97620\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 12, Train Err.:0.34823, Validation Err.:0.39000, Test Accuracy:0.92910, Max Test Accuracy:0.92910\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 13, Train Err.:0.05578, Validation Err.:0.13774, Test Accuracy:0.95750, Max Test Accuracy:0.95750\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 13, Train Err.:0.00130, Validation Err.:0.10361, Test Accuracy:0.97660, Max Test Accuracy:0.97660\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 13, Train Err.:0.32934, Validation Err.:0.37521, Test Accuracy:0.93140, Max Test Accuracy:0.93140\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 13, Train Err.:0.03662, Validation Err.:0.10900, Test Accuracy:0.96360, Max Test Accuracy:0.96360\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 13, Train Err.:0.00133, Validation Err.:0.09494, Test Accuracy:0.97620, Max Test Accuracy:0.97620\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 13, Train Err.:0.33641, Validation Err.:0.38260, Test Accuracy:0.92810, Max Test Accuracy:0.92910\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 14, Train Err.:0.05166, Validation Err.:0.13853, Test Accuracy:0.95770, Max Test Accuracy:0.95770\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 14, Train Err.:0.00111, Validation Err.:0.10442, Test Accuracy:0.97650, Max Test Accuracy:0.97660\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 14, Train Err.:0.32834, Validation Err.:0.36650, Test Accuracy:0.93290, Max Test Accuracy:0.93290\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 14, Train Err.:0.03324, Validation Err.:0.10699, Test Accuracy:0.96430, Max Test Accuracy:0.96430\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 14, Train Err.:0.00117, Validation Err.:0.09660, Test Accuracy:0.97620, Max Test Accuracy:0.97620\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 14, Train Err.:0.32340, Validation Err.:0.37252, Test Accuracy:0.93150, Max Test Accuracy:0.93150\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 15, Train Err.:0.05007, Validation Err.:0.14017, Test Accuracy:0.95740, Max Test Accuracy:0.95770\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 15, Train Err.:0.00097, Validation Err.:0.10582, Test Accuracy:0.97690, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 15, Train Err.:0.32052, Validation Err.:0.36358, Test Accuracy:0.93420, Max Test Accuracy:0.93420\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 15, Train Err.:0.02984, Validation Err.:0.10748, Test Accuracy:0.96380, Max Test Accuracy:0.96430\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 15, Train Err.:0.00104, Validation Err.:0.09814, Test Accuracy:0.97600, Max Test Accuracy:0.97620\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 15, Train Err.:0.31628, Validation Err.:0.36752, Test Accuracy:0.93160, Max Test Accuracy:0.93160\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 16, Train Err.:0.04876, Validation Err.:0.14164, Test Accuracy:0.95790, Max Test Accuracy:0.95790\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 16, Train Err.:0.00086, Validation Err.:0.10693, Test Accuracy:0.97680, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 16, Train Err.:0.31383, Validation Err.:0.35600, Test Accuracy:0.93450, Max Test Accuracy:0.93450\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 16, Train Err.:0.02816, Validation Err.:0.10686, Test Accuracy:0.96450, Max Test Accuracy:0.96450\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 16, Train Err.:0.00092, Validation Err.:0.09970, Test Accuracy:0.97590, Max Test Accuracy:0.97620\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 16, Train Err.:0.32000, Validation Err.:0.36064, Test Accuracy:0.93310, Max Test Accuracy:0.93310\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 17, Train Err.:0.04692, Validation Err.:0.14243, Test Accuracy:0.95790, Max Test Accuracy:0.95790\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 17, Train Err.:0.00076, Validation Err.:0.10844, Test Accuracy:0.97680, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 17, Train Err.:0.30872, Validation Err.:0.35081, Test Accuracy:0.93570, Max Test Accuracy:0.93570\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 17, Train Err.:0.02668, Validation Err.:0.10748, Test Accuracy:0.96550, Max Test Accuracy:0.96550\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 17, Train Err.:0.00082, Validation Err.:0.10096, Test Accuracy:0.97630, Max Test Accuracy:0.97630\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 17, Train Err.:0.31683, Validation Err.:0.35991, Test Accuracy:0.93290, Max Test Accuracy:0.93310\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 18, Train Err.:0.04615, Validation Err.:0.14246, Test Accuracy:0.95960, Max Test Accuracy:0.95960\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 18, Train Err.:0.00069, Validation Err.:0.10958, Test Accuracy:0.97680, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 18, Train Err.:0.30258, Validation Err.:0.34818, Test Accuracy:0.93560, Max Test Accuracy:0.93570\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 18, Train Err.:0.02500, Validation Err.:0.10849, Test Accuracy:0.96520, Max Test Accuracy:0.96550\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 18, Train Err.:0.00075, Validation Err.:0.10225, Test Accuracy:0.97660, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 18, Train Err.:0.30635, Validation Err.:0.35130, Test Accuracy:0.93340, Max Test Accuracy:0.93340\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 19, Train Err.:0.04534, Validation Err.:0.14372, Test Accuracy:0.95910, Max Test Accuracy:0.95960\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 19, Train Err.:0.00062, Validation Err.:0.11080, Test Accuracy:0.97690, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 19, Train Err.:0.29958, Validation Err.:0.34209, Test Accuracy:0.93620, Max Test Accuracy:0.93620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 19, Train Err.:0.02391, Validation Err.:0.10896, Test Accuracy:0.96550, Max Test Accuracy:0.96550\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 19, Train Err.:0.00068, Validation Err.:0.10331, Test Accuracy:0.97640, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 19, Train Err.:0.30547, Validation Err.:0.35055, Test Accuracy:0.93330, Max Test Accuracy:0.93340\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 20, Train Err.:0.04435, Validation Err.:0.14378, Test Accuracy:0.95940, Max Test Accuracy:0.95960\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 20, Train Err.:0.00057, Validation Err.:0.11184, Test Accuracy:0.97670, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 20, Train Err.:0.29679, Validation Err.:0.33832, Test Accuracy:0.93700, Max Test Accuracy:0.93700\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 20, Train Err.:0.02276, Validation Err.:0.11006, Test Accuracy:0.96590, Max Test Accuracy:0.96590\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 20, Train Err.:0.00062, Validation Err.:0.10449, Test Accuracy:0.97630, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 20, Train Err.:0.29915, Validation Err.:0.34496, Test Accuracy:0.93540, Max Test Accuracy:0.93540\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 21, Train Err.:0.04319, Validation Err.:0.14539, Test Accuracy:0.96030, Max Test Accuracy:0.96030\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 21, Train Err.:0.00052, Validation Err.:0.11285, Test Accuracy:0.97670, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 21, Train Err.:0.28954, Validation Err.:0.33464, Test Accuracy:0.93670, Max Test Accuracy:0.93700\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 21, Train Err.:0.02176, Validation Err.:0.11107, Test Accuracy:0.96610, Max Test Accuracy:0.96610\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 21, Train Err.:0.00057, Validation Err.:0.10559, Test Accuracy:0.97630, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 21, Train Err.:0.29590, Validation Err.:0.34388, Test Accuracy:0.93430, Max Test Accuracy:0.93540\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 22, Train Err.:0.04199, Validation Err.:0.14508, Test Accuracy:0.96040, Max Test Accuracy:0.96040\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 22, Train Err.:0.00048, Validation Err.:0.11385, Test Accuracy:0.97680, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 22, Train Err.:0.28989, Validation Err.:0.33250, Test Accuracy:0.93680, Max Test Accuracy:0.93700\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 22, Train Err.:0.02073, Validation Err.:0.11237, Test Accuracy:0.96580, Max Test Accuracy:0.96610\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 22, Train Err.:0.00053, Validation Err.:0.10661, Test Accuracy:0.97620, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 22, Train Err.:0.29062, Validation Err.:0.33910, Test Accuracy:0.93560, Max Test Accuracy:0.93560\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 23, Train Err.:0.04078, Validation Err.:0.14599, Test Accuracy:0.96060, Max Test Accuracy:0.96060\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 23, Train Err.:0.00045, Validation Err.:0.11484, Test Accuracy:0.97690, Max Test Accuracy:0.97690\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 23, Train Err.:0.28247, Validation Err.:0.32833, Test Accuracy:0.93870, Max Test Accuracy:0.93870\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 23, Train Err.:0.01981, Validation Err.:0.11324, Test Accuracy:0.96630, Max Test Accuracy:0.96630\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 23, Train Err.:0.00050, Validation Err.:0.10755, Test Accuracy:0.97630, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 23, Train Err.:0.28678, Validation Err.:0.33179, Test Accuracy:0.93570, Max Test Accuracy:0.93570\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 24, Train Err.:0.03917, Validation Err.:0.14634, Test Accuracy:0.96050, Max Test Accuracy:0.96060\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 24, Train Err.:0.00041, Validation Err.:0.11577, Test Accuracy:0.97700, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 24, Train Err.:0.27577, Validation Err.:0.32034, Test Accuracy:0.93940, Max Test Accuracy:0.93940\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 24, Train Err.:0.01866, Validation Err.:0.11512, Test Accuracy:0.96590, Max Test Accuracy:0.96630\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 24, Train Err.:0.00046, Validation Err.:0.10848, Test Accuracy:0.97620, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 24, Train Err.:0.28297, Validation Err.:0.33251, Test Accuracy:0.93570, Max Test Accuracy:0.93570\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 25, Train Err.:0.03928, Validation Err.:0.15092, Test Accuracy:0.96030, Max Test Accuracy:0.96060\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 25, Train Err.:0.00039, Validation Err.:0.11670, Test Accuracy:0.97690, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 25, Train Err.:0.27189, Validation Err.:0.31887, Test Accuracy:0.93930, Max Test Accuracy:0.93940\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 25, Train Err.:0.01798, Validation Err.:0.11687, Test Accuracy:0.96620, Max Test Accuracy:0.96630\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 25, Train Err.:0.00043, Validation Err.:0.10926, Test Accuracy:0.97620, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 25, Train Err.:0.27871, Validation Err.:0.32801, Test Accuracy:0.93720, Max Test Accuracy:0.93720\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 26, Train Err.:0.03778, Validation Err.:0.15127, Test Accuracy:0.96050, Max Test Accuracy:0.96060\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 26, Train Err.:0.00036, Validation Err.:0.11755, Test Accuracy:0.97680, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 26, Train Err.:0.27410, Validation Err.:0.32080, Test Accuracy:0.93830, Max Test Accuracy:0.93940\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 26, Train Err.:0.01710, Validation Err.:0.11880, Test Accuracy:0.96700, Max Test Accuracy:0.96700\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 26, Train Err.:0.00041, Validation Err.:0.11009, Test Accuracy:0.97610, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 26, Train Err.:0.27968, Validation Err.:0.32543, Test Accuracy:0.93670, Max Test Accuracy:0.93720\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 27, Train Err.:0.03660, Validation Err.:0.15307, Test Accuracy:0.96020, Max Test Accuracy:0.96060\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 27, Train Err.:0.00034, Validation Err.:0.11839, Test Accuracy:0.97660, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 27, Train Err.:0.26602, Validation Err.:0.31758, Test Accuracy:0.93840, Max Test Accuracy:0.93940\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 27, Train Err.:0.01626, Validation Err.:0.11943, Test Accuracy:0.96720, Max Test Accuracy:0.96720\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 27, Train Err.:0.00038, Validation Err.:0.11084, Test Accuracy:0.97620, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 27, Train Err.:0.27474, Validation Err.:0.32402, Test Accuracy:0.93710, Max Test Accuracy:0.93720\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 28, Train Err.:0.03512, Validation Err.:0.15302, Test Accuracy:0.96080, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 28, Train Err.:0.00032, Validation Err.:0.11921, Test Accuracy:0.97660, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 28, Train Err.:0.25895, Validation Err.:0.31177, Test Accuracy:0.94010, Max Test Accuracy:0.94010\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 28, Train Err.:0.01539, Validation Err.:0.12042, Test Accuracy:0.96740, Max Test Accuracy:0.96740\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 28, Train Err.:0.00036, Validation Err.:0.11162, Test Accuracy:0.97610, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 28, Train Err.:0.27454, Validation Err.:0.32350, Test Accuracy:0.93620, Max Test Accuracy:0.93720\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 29, Train Err.:0.03440, Validation Err.:0.15515, Test Accuracy:0.96060, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 29, Train Err.:0.00030, Validation Err.:0.11998, Test Accuracy:0.97660, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 29, Train Err.:0.27028, Validation Err.:0.31340, Test Accuracy:0.93690, Max Test Accuracy:0.94010\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 29, Train Err.:0.01432, Validation Err.:0.12217, Test Accuracy:0.96710, Max Test Accuracy:0.96740\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 29, Train Err.:0.00034, Validation Err.:0.11234, Test Accuracy:0.97610, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 29, Train Err.:0.26981, Validation Err.:0.31980, Test Accuracy:0.93780, Max Test Accuracy:0.93780\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 30, Train Err.:0.03342, Validation Err.:0.15591, Test Accuracy:0.96050, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 30, Train Err.:0.00029, Validation Err.:0.12071, Test Accuracy:0.97660, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 30, Train Err.:0.26814, Validation Err.:0.31344, Test Accuracy:0.93920, Max Test Accuracy:0.94010\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 30, Train Err.:0.01303, Validation Err.:0.12385, Test Accuracy:0.96680, Max Test Accuracy:0.96740\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 30, Train Err.:0.00033, Validation Err.:0.11306, Test Accuracy:0.97610, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 30, Train Err.:0.26723, Validation Err.:0.31677, Test Accuracy:0.93850, Max Test Accuracy:0.93850\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 31, Train Err.:0.03243, Validation Err.:0.15623, Test Accuracy:0.96070, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 31, Train Err.:0.00027, Validation Err.:0.12153, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 31, Train Err.:0.25137, Validation Err.:0.30525, Test Accuracy:0.94090, Max Test Accuracy:0.94090\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 31, Train Err.:0.01225, Validation Err.:0.12443, Test Accuracy:0.96740, Max Test Accuracy:0.96740\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 31, Train Err.:0.00031, Validation Err.:0.11376, Test Accuracy:0.97620, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 31, Train Err.:0.26592, Validation Err.:0.31338, Test Accuracy:0.93860, Max Test Accuracy:0.93860\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 32, Train Err.:0.03175, Validation Err.:0.15853, Test Accuracy:0.96050, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 32, Train Err.:0.00026, Validation Err.:0.12219, Test Accuracy:0.97640, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 32, Train Err.:0.25264, Validation Err.:0.30131, Test Accuracy:0.94210, Max Test Accuracy:0.94210\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 32, Train Err.:0.01159, Validation Err.:0.12526, Test Accuracy:0.96740, Max Test Accuracy:0.96740\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 32, Train Err.:0.00029, Validation Err.:0.11450, Test Accuracy:0.97610, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 32, Train Err.:0.26338, Validation Err.:0.31205, Test Accuracy:0.93760, Max Test Accuracy:0.93860\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 33, Train Err.:0.03032, Validation Err.:0.15880, Test Accuracy:0.96060, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 33, Train Err.:0.00025, Validation Err.:0.12286, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 33, Train Err.:0.25775, Validation Err.:0.29989, Test Accuracy:0.94080, Max Test Accuracy:0.94210\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 33, Train Err.:0.01096, Validation Err.:0.12662, Test Accuracy:0.96760, Max Test Accuracy:0.96760\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 33, Train Err.:0.00028, Validation Err.:0.11509, Test Accuracy:0.97610, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 33, Train Err.:0.26122, Validation Err.:0.30823, Test Accuracy:0.93990, Max Test Accuracy:0.93990\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 34, Train Err.:0.02934, Validation Err.:0.16047, Test Accuracy:0.96040, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 34, Train Err.:0.00024, Validation Err.:0.12351, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 34, Train Err.:0.25318, Validation Err.:0.29770, Test Accuracy:0.94240, Max Test Accuracy:0.94240\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 34, Train Err.:0.01021, Validation Err.:0.12845, Test Accuracy:0.96740, Max Test Accuracy:0.96760\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 34, Train Err.:0.00027, Validation Err.:0.11575, Test Accuracy:0.97600, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 34, Train Err.:0.25927, Validation Err.:0.30368, Test Accuracy:0.93970, Max Test Accuracy:0.93990\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 35, Train Err.:0.02863, Validation Err.:0.16114, Test Accuracy:0.96080, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 35, Train Err.:0.00023, Validation Err.:0.12418, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 35, Train Err.:0.24809, Validation Err.:0.29516, Test Accuracy:0.94240, Max Test Accuracy:0.94240\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 35, Train Err.:0.00977, Validation Err.:0.12905, Test Accuracy:0.96780, Max Test Accuracy:0.96780\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 35, Train Err.:0.00026, Validation Err.:0.11632, Test Accuracy:0.97600, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 35, Train Err.:0.25839, Validation Err.:0.30716, Test Accuracy:0.93870, Max Test Accuracy:0.93990\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 36, Train Err.:0.02715, Validation Err.:0.16195, Test Accuracy:0.96040, Max Test Accuracy:0.96080\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 36, Train Err.:0.00022, Validation Err.:0.12471, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 36, Train Err.:0.24926, Validation Err.:0.29458, Test Accuracy:0.94320, Max Test Accuracy:0.94320\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 36, Train Err.:0.00930, Validation Err.:0.13014, Test Accuracy:0.96770, Max Test Accuracy:0.96780\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 36, Train Err.:0.00025, Validation Err.:0.11704, Test Accuracy:0.97600, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 36, Train Err.:0.25514, Validation Err.:0.30490, Test Accuracy:0.93850, Max Test Accuracy:0.93990\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 37, Train Err.:0.02647, Validation Err.:0.16439, Test Accuracy:0.96100, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 37, Train Err.:0.00021, Validation Err.:0.12533, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 37, Train Err.:0.23976, Validation Err.:0.28905, Test Accuracy:0.94420, Max Test Accuracy:0.94420\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 37, Train Err.:0.00892, Validation Err.:0.13128, Test Accuracy:0.96760, Max Test Accuracy:0.96780\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 37, Train Err.:0.00024, Validation Err.:0.11755, Test Accuracy:0.97610, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 37, Train Err.:0.25090, Validation Err.:0.30167, Test Accuracy:0.93870, Max Test Accuracy:0.93990\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 38, Train Err.:0.02544, Validation Err.:0.16335, Test Accuracy:0.96070, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 38, Train Err.:0.00020, Validation Err.:0.12588, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 38, Train Err.:0.24853, Validation Err.:0.28956, Test Accuracy:0.94340, Max Test Accuracy:0.94420\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 38, Train Err.:0.00854, Validation Err.:0.13268, Test Accuracy:0.96780, Max Test Accuracy:0.96780\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 38, Train Err.:0.00023, Validation Err.:0.11807, Test Accuracy:0.97600, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 38, Train Err.:0.25251, Validation Err.:0.30084, Test Accuracy:0.93950, Max Test Accuracy:0.93990\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 39, Train Err.:0.02501, Validation Err.:0.16580, Test Accuracy:0.96090, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 39, Train Err.:0.00019, Validation Err.:0.12646, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 39, Train Err.:0.24411, Validation Err.:0.28804, Test Accuracy:0.94440, Max Test Accuracy:0.94440\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 39, Train Err.:0.00805, Validation Err.:0.13392, Test Accuracy:0.96770, Max Test Accuracy:0.96780\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 39, Train Err.:0.00022, Validation Err.:0.11859, Test Accuracy:0.97580, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 39, Train Err.:0.24540, Validation Err.:0.29953, Test Accuracy:0.94030, Max Test Accuracy:0.94030\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 40, Train Err.:0.02445, Validation Err.:0.16672, Test Accuracy:0.96090, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 40, Train Err.:0.00019, Validation Err.:0.12700, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 40, Train Err.:0.23861, Validation Err.:0.28453, Test Accuracy:0.94490, Max Test Accuracy:0.94490\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 40, Train Err.:0.00774, Validation Err.:0.13473, Test Accuracy:0.96770, Max Test Accuracy:0.96780\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 40, Train Err.:0.00021, Validation Err.:0.11910, Test Accuracy:0.97590, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 40, Train Err.:0.24690, Validation Err.:0.29990, Test Accuracy:0.94120, Max Test Accuracy:0.94120\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 41, Train Err.:0.02432, Validation Err.:0.16874, Test Accuracy:0.96000, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 41, Train Err.:0.00018, Validation Err.:0.12749, Test Accuracy:0.97640, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 41, Train Err.:0.23502, Validation Err.:0.28314, Test Accuracy:0.94530, Max Test Accuracy:0.94530\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 41, Train Err.:0.00752, Validation Err.:0.13642, Test Accuracy:0.96820, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 41, Train Err.:0.00020, Validation Err.:0.11965, Test Accuracy:0.97590, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 41, Train Err.:0.24877, Validation Err.:0.29509, Test Accuracy:0.94160, Max Test Accuracy:0.94160\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 42, Train Err.:0.02285, Validation Err.:0.16934, Test Accuracy:0.96020, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 42, Train Err.:0.00017, Validation Err.:0.12807, Test Accuracy:0.97630, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 42, Train Err.:0.23065, Validation Err.:0.28077, Test Accuracy:0.94620, Max Test Accuracy:0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 42, Train Err.:0.00703, Validation Err.:0.13766, Test Accuracy:0.96770, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 42, Train Err.:0.00020, Validation Err.:0.12013, Test Accuracy:0.97560, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 42, Train Err.:0.24649, Validation Err.:0.29687, Test Accuracy:0.94010, Max Test Accuracy:0.94160\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 43, Train Err.:0.02299, Validation Err.:0.16936, Test Accuracy:0.96040, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 43, Train Err.:0.00017, Validation Err.:0.12855, Test Accuracy:0.97640, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 43, Train Err.:0.23461, Validation Err.:0.28135, Test Accuracy:0.94460, Max Test Accuracy:0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 43, Train Err.:0.00681, Validation Err.:0.13921, Test Accuracy:0.96770, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 43, Train Err.:0.00019, Validation Err.:0.12063, Test Accuracy:0.97570, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 43, Train Err.:0.24317, Validation Err.:0.29660, Test Accuracy:0.94070, Max Test Accuracy:0.94160\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 44, Train Err.:0.02193, Validation Err.:0.17098, Test Accuracy:0.96100, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 44, Train Err.:0.00016, Validation Err.:0.12907, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 44, Train Err.:0.23058, Validation Err.:0.28014, Test Accuracy:0.94460, Max Test Accuracy:0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 44, Train Err.:0.00652, Validation Err.:0.13996, Test Accuracy:0.96760, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 44, Train Err.:0.00018, Validation Err.:0.12114, Test Accuracy:0.97560, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 44, Train Err.:0.24141, Validation Err.:0.29775, Test Accuracy:0.93990, Max Test Accuracy:0.94160\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 45, Train Err.:0.02191, Validation Err.:0.17240, Test Accuracy:0.96070, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 45, Train Err.:0.00016, Validation Err.:0.12953, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 45, Train Err.:0.23597, Validation Err.:0.27892, Test Accuracy:0.94560, Max Test Accuracy:0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 45, Train Err.:0.00627, Validation Err.:0.14153, Test Accuracy:0.96720, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 45, Train Err.:0.00018, Validation Err.:0.12168, Test Accuracy:0.97560, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 45, Train Err.:0.24047, Validation Err.:0.29621, Test Accuracy:0.93930, Max Test Accuracy:0.94160\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 46, Train Err.:0.02130, Validation Err.:0.17170, Test Accuracy:0.96090, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 46, Train Err.:0.00015, Validation Err.:0.13001, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 46, Train Err.:0.23316, Validation Err.:0.27729, Test Accuracy:0.94490, Max Test Accuracy:0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 46, Train Err.:0.00601, Validation Err.:0.14269, Test Accuracy:0.96710, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 46, Train Err.:0.00017, Validation Err.:0.12198, Test Accuracy:0.97560, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 46, Train Err.:0.23721, Validation Err.:0.29099, Test Accuracy:0.94170, Max Test Accuracy:0.94170\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 47, Train Err.:0.02095, Validation Err.:0.17358, Test Accuracy:0.96080, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 47, Train Err.:0.00015, Validation Err.:0.13046, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 47, Train Err.:0.23210, Validation Err.:0.27684, Test Accuracy:0.94560, Max Test Accuracy:0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 47, Train Err.:0.00566, Validation Err.:0.14375, Test Accuracy:0.96770, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 47, Train Err.:0.00017, Validation Err.:0.12249, Test Accuracy:0.97530, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 47, Train Err.:0.23684, Validation Err.:0.29381, Test Accuracy:0.93980, Max Test Accuracy:0.94170\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 48, Train Err.:0.02009, Validation Err.:0.17410, Test Accuracy:0.96010, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 48, Train Err.:0.00014, Validation Err.:0.13089, Test Accuracy:0.97650, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 48, Train Err.:0.23556, Validation Err.:0.27700, Test Accuracy:0.94560, Max Test Accuracy:0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 48, Train Err.:0.00546, Validation Err.:0.14534, Test Accuracy:0.96750, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 48, Train Err.:0.00016, Validation Err.:0.12290, Test Accuracy:0.97530, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 48, Train Err.:0.23525, Validation Err.:0.28893, Test Accuracy:0.94160, Max Test Accuracy:0.94170\n", "\n", "N2, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 49, Train Err.:0.02023, Validation Err.:0.17624, Test Accuracy:0.96070, Max Test Accuracy:0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 49, Train Err.:0.00014, Validation Err.:0.13131, Test Accuracy:0.97660, Max Test Accuracy:0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 49, Train Err.:0.22922, Validation Err.:0.27559, Test Accuracy:0.94510, Max Test Accuracy:0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0-Epoch: 49, Train Err.:0.00524, Validation Err.:0.14633, Test Accuracy:0.96740, Max Test Accuracy:0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 -Epoch: 49, Train Err.:0.00016, Validation Err.:0.12329, Test Accuracy:0.97540, Max Test Accuracy:0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 -Epoch: 49, Train Err.:0.23311, Validation Err.:0.28942, Test Accuracy:0.94110, Max Test Accuracy:0.94170\n", "\n" ] } ], "source": [ "epoch_list = []\n", "\n", "num_batch = math.ceil(train_size / batch_size)\n", "\n", "for i in range(num_epochs):\n", " epoch_list.append(i)\n", " for key in markers.keys():\n", " for k in range(num_batch):\n", " x_batch = img_train[k * batch_size : k * batch_size + batch_size]\n", " t_batch = label_train[k * batch_size : k * batch_size + batch_size]\n", " networks[key].learning(x_batch, t_batch)\n", "\n", " train_loss = networks[key].loss(x_batch, t_batch, is_train=True)\n", " train_errors[key].append(train_loss)\n", "\n", " validation_loss = networks[key].loss(img_validation, label_validation, is_train=False)\n", " validation_errors[key].append(validation_loss) \n", "\n", " test_accuracy = networks[key].accuracy(img_test, label_test)\n", " test_accuracy_values[key].append(test_accuracy)\n", " if test_accuracy > max_test_accuracy_value[key]:\n", " max_test_accuracy_epoch[key] = i\n", " max_test_accuracy_value[key] = test_accuracy\n", " print(\"{0:38s}-Epoch:{1:3d}, Train Err.:{2:7.5f}, Validation Err.:{3:7.5f}, Test Accuracy:{4:7.5f}, Max Test Accuracy:{5:7.5f}\".format(\n", " key,\n", " i,\n", " train_loss,\n", " validation_loss,\n", " test_accuracy,\n", " max_test_accuracy_value[key]\n", " ))\n", " print() " ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABJkAAALJCAYAAAAXlPLmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "f, axarr = plt.subplots(2, 2, figsize=(20, 12))\n", "for key in markers.keys():\n", " axarr[0, 0].plot(epoch_list[1:], train_errors[key][1:], marker=markers[key], markevery=2, label=key)\n", "axarr[0, 0].set_ylabel('Train - Total Error')\n", "axarr[0, 0].set_xlabel('Epochs')\n", "axarr[0, 0].grid(True)\n", "axarr[0, 0].set_title('Train Error')\n", "axarr[0, 0].legend(loc='upper right')\n", "\n", "for key in markers.keys():\n", " axarr[0, 1].plot(epoch_list[1:], validation_errors[key][1:], marker=markers[key], markevery=2, label=key)\n", "axarr[0, 1].set_ylabel('Validation - Total Error')\n", "axarr[0, 1].set_xlabel('Epochs')\n", "axarr[0, 1].grid(True)\n", "axarr[0, 1].set_title('Validation Error')\n", "axarr[0, 1].legend(loc='upper right')\n", "\n", "for key in markers.keys():\n", " axarr[1, 0].plot(epoch_list[1:], train_errors[key][1:], marker=markers[key], markevery=2, label=key)\n", "axarr[1, 0].set_ylabel('Train - Total Error')\n", "axarr[1, 0].set_xlabel('Epochs')\n", "axarr[1, 0].grid(True)\n", "axarr[1, 0].set_ylim(0, 0.1)\n", "axarr[1, 0].set_title('Train Error (0.00 ~ 0.10)')\n", "axarr[1, 0].legend(loc='upper right')\n", "\n", "for key in markers.keys():\n", " axarr[1, 1].plot(epoch_list[1:], validation_errors[key][1:], marker=markers[key], markevery=2, label=key)\n", "axarr[1, 1].set_ylabel('Validation - Total Error')\n", "axarr[1, 1].set_xlabel('Epochs')\n", "axarr[1, 1].grid(True)\n", "axarr[1, 1].set_ylim(0.05, 0.2)\n", "axarr[1, 1].set_title('Validation Error (0.05 ~ 0.20)')\n", "axarr[1, 1].legend(loc='upper right')\n", "\n", "f.subplots_adjust(hspace=0.3)\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4IAAAJdCAYAAABnFskvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "f, axarr = plt.subplots(2, 1, figsize=(15,10))\n", "for key in markers.keys():\n", " axarr[0].plot(epoch_list[1:], test_accuracy_values[key][1:], marker=markers[key], markevery=1, label=key)\n", "axarr[0].set_ylabel('Test Accuracy')\n", "axarr[0].set_xlabel('Epochs')\n", "axarr[0].grid(True)\n", "axarr[0].set_title('Test Accuracy')\n", "axarr[0].legend(loc='lower right')\n", "\n", "for key in markers.keys():\n", " axarr[1].plot(epoch_list[1:], test_accuracy_values[key][1:], marker=markers[key], markevery=1, label=key)\n", "axarr[1].set_ylabel('Test Accuracy')\n", "axarr[1].set_xlabel('Epochs')\n", "axarr[1].grid(True)\n", "axarr[1].set_ylim(0.92, 0.99)\n", "axarr[1].set_title('Test Accuracy (0.92 ~ 0.99)')\n", "axarr[1].legend(loc='lower right')\n", "\n", "f.subplots_adjust(hspace=0.3)\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "N2, AdaGrad, No_Batch_Norm, lambda=0.0 - Epoch: 37, Max Test Accuracy: 0.96100\n", "N2, AdaGrad, Batch_Norm, lambda=0.0 - Epoch: 24, Max Test Accuracy: 0.97700\n", "N2, AdaGrad, Batch_Norm, lambda=0.1 - Epoch: 42, Max Test Accuracy: 0.94620\n", "He, AdaGrad, No_Batch_Norm, lambda=0.0 - Epoch: 41, Max Test Accuracy: 0.96820\n", "He, AdaGrad, Batch_Norm, lambda=0.0 - Epoch: 18, Max Test Accuracy: 0.97660\n", "He, AdaGrad, Batch_Norm, lambda=0.1 - Epoch: 46, Max Test Accuracy: 0.94170\n" ] } ], "source": [ "for key in markers.keys():\n", " print(\"{0:26s} - Epoch:{1:3d}, Max Test Accuracy: {2:7.5f}\".format(key, max_test_accuracy_epoch[key], max_test_accuracy_value[key]))" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.0" } }, "nbformat": 4, "nbformat_minor": 0 }