文章目錄

• 目錄
• • 1.神經(jīng)網(wǎng)絡(luò)知識(shí)概覽
  • • 1.1深度學(xué)習(xí)頂會(huì)
    • 1.2相關(guān)比賽
    • 1.3神經(jīng)網(wǎng)絡(luò)知識(shí)概覽
    • 1.4神經(jīng)網(wǎng)絡(luò)編程一般實(shí)現(xiàn)過(guò)程
  • 2.簡(jiǎn)單神經(jīng)網(wǎng)絡(luò)ANN
  • • 2.1 數(shù)據(jù)集：
    • 2.2 網(wǎng)絡(luò)結(jié)構(gòu)：
    • 2.3 代碼實(shí)現(xiàn)
    • • 2.3.1 讀取數(shù)據(jù)，并做處理
      • 2.3.2 構(gòu)建網(wǎng)絡(luò)結(jié)構(gòu)
      • 2.3.3 訓(xùn)練網(wǎng)絡(luò)

目錄

1.神經(jīng)網(wǎng)絡(luò)知識(shí)概覽

1.1深度學(xué)習(xí)頂會(huì)

• CVPR : IEEE Conference on Computer Vision and Pattern Recognition
  • CVPR是計(jì)算機(jī)視覺(jué)與模式識(shí)別頂會(huì)
• ICCV：IEEE International Conference on Computer Vision
  • ICCV論文錄用率非常低，是三大會(huì)議中公認(rèn)級(jí)別最高的
• ECCV：European Conference on Computer Vision

1.2相關(guān)比賽

1.ImageNet

• ImageNet 數(shù)據(jù)集最初由斯坦福大學(xué)李飛飛等人在 CVPR 2009 的一篇論文中推出
• 
  
  2.webvision
  

1.3神經(jīng)網(wǎng)絡(luò)知識(shí)概覽

1.4神經(jīng)網(wǎng)絡(luò)編程一般實(shí)現(xiàn)過(guò)程

1.數(shù)據(jù)預(yù)處理
2.定義神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)
3.初始化網(wǎng)絡(luò)模型中的參數(shù)
4.開(kāi)始訓(xùn)練模型

loop(number_iterations):forward propagationcompute costbackward propagationupdate parameters

5.對(duì)新的數(shù)據(jù)進(jìn)行預(yù)測(cè)

2.簡(jiǎn)單神經(jīng)網(wǎng)絡(luò)ANN

2.1 數(shù)據(jù)集：

• 訓(xùn)練集 + 測(cè)試集
• 訓(xùn)練集：訓(xùn)練集 + 評(píng)估集

• 數(shù)據(jù)信息：
  

2.2 網(wǎng)絡(luò)結(jié)構(gòu)：

• 網(wǎng)絡(luò)結(jié)構(gòu) linear -> relu -> linear -> relu -> linear -> softmax
• 網(wǎng)絡(luò)結(jié)構(gòu)12288 -> 25 -> 12 -> 6
• 迭代次數(shù)1000，學(xué)習(xí)率0.0001，minibatch_size=32，優(yōu)化算法Adam
• 將RGB圖片轉(zhuǎn)換為向量（損失空間結(jié)構(gòu)信息）
• 出現(xiàn)過(guò)擬合，應(yīng)該使用正則化（L2、Dropout、早停）

2.3 代碼實(shí)現(xiàn)

2.3.1 讀取數(shù)據(jù)，并做處理

import math import h5py import numpy as np import tensorflow as tf import matplotlib.pyplot as plt import scipy from PIL import Image from scipy import ndimage from tensorflow.python.framework import ops from improv_utils import *%matplotlib inline np.random.seed(1)# 下載數(shù)據(jù) X_train_orig, Y_train_orig, X_test_orig, Y_test_orig, classes = load_dataset()# 顯示圖片 index = 2 plt.imshow(X_train_orig[index]) plt.show() print("y = " + str(np.squeeze(Y_train_orig[:, index])))# 將數(shù)據(jù)平鋪，歸一化，標(biāo)簽one-hot X_train_flatten = X_train_orig.reshape(X_train_orig.shape[0], -1).T X_test_flatten = X_test_orig.reshape(X_test_orig.shape[0], -1).TX_train = X_train_flatten/255. X_test = X_test_flatten/255.Y_train = convert_to_one_hot(Y_train_orig, 6) Y_test = convert_to_one_hot(Y_test_orig, 6)print ("number of training examples = " + str(X_train.shape[1])) print ("number of test examples = " + str(X_test.shape[1])) print ("X_train shape: " + str(X_train.shape)) print ("Y_train shape: " + str(Y_train.shape)) print ("X_test shape: " + str(X_test.shape)) print ("Y_test shape: " + str(Y_test.shape))

y = 2
number of training examples = 1080
number of test examples = 120
X_train shape: (12288, 1080)
Y_train shape: (6, 1080)
X_test shape: (12288, 120)
Y_test shape: (6, 120)

2.3.2 構(gòu)建網(wǎng)絡(luò)結(jié)構(gòu)

# 1-1、創(chuàng)建占位符 def create_placeholders(n_x, n_y):"""Creates the placeholders for the tensorflow session.Arguments:n_x -- scalar, size of an image vector (num_px * num_px = 64 * 64 * 3 = 12288)n_y -- scalar, number of classes (from 0 to 5, so -> 6)Returns:X -- placeholder for the data input, of shape [n_x, None] and dtype "float"Y -- placeholder for the input labels, of shape [n_y, None] and dtype "float"Tips:- You will use None because it let's us be flexible on the number of examples you will for the placeholders.In fact, the number of examples during test/train is different."""X = tf.placeholder(tf.float32, shape = [n_x, None])Y = tf.placeholder(tf.float32, shape = [n_y, None])return X, Y# 1-2、初始化參數(shù) def initialize_parameters():"""Initializes parameters to build a neural network with tensorflow. The shapes are:W1 : [25, 12288]b1 : [25, 1]W2 : [12, 25]b2 : [12, 1]W3 : [6, 12]b3 : [6, 1]Returns:parameters -- a dictionary of tensors containing W1, b1, W2, b2, W3, b3"""tf.set_random_seed(1) # so that your "random" numbers match oursW1 = tf.get_variable("W1", [25,12288], initializer = tf.contrib.layers.xavier_initializer(seed = 1))b1 = tf.get_variable("b1", [25,1], initializer = tf.zeros_initializer())W2 = tf.get_variable("W2", [12,25], initializer = tf.contrib.layers.xavier_initializer(seed = 1))b2 = tf.get_variable("b2", [12,1], initializer = tf.zeros_initializer())W3 = tf.get_variable("W3", [6,12], initializer = tf.contrib.layers.xavier_initializer(seed = 1))b3 = tf.get_variable("b3", [6,1], initializer = tf.zeros_initializer())parameters = {"W1": W1,"b1": b1,"W2": W2,"b2": b2,"W3": W3,"b3": b3}return parameters# 1-3、TensorFlow中的前向傳播 # tf中前向傳播停止在z3，是因?yàn)閠f中最后的線性層輸出是被作為輸入計(jì)算loss，不需要a3 def forward_propagation(X, parameters):"""Implements the forward propagation for the model: LINEAR -> RELU -> LINEAR -> RELU -> LINEAR -> SOFTMAXArguments:X -- input dataset placeholder, of shape (input size, number of examples)parameters -- python dictionary containing your parameters "W1", "b1", "W2", "b2", "W3", "b3"the shapes are given in initialize_parametersReturns:Z3 -- the output of the last LINEAR unit"""W1 = parameters['W1']b1 = parameters['b1']W2 = parameters['W2']b2 = parameters['b2']W3 = parameters['W3']b3 = parameters['b3']Z1 = tf.add(tf.matmul(W1, X), b1) # Z1 = np.dot(W1, X) + b1A1 = tf.nn.relu(Z1) # A1 = relu(Z1)Z2 = tf.add(tf.matmul(W2, A1), b2) # Z2 = np.dot(W2, a1) + b2A2 = tf.nn.relu(Z2) # A2 = relu(Z2)Z3 = tf.add(tf.matmul(W3, A2), b3) # Z3 = np.dot(W3,Z2) + b3return Z3# 1-4、計(jì)算成本函數(shù) def compute_cost(Z3, Y):"""Computes the costArguments:Z3 -- output of forward propagation (output of the last LINEAR unit), of shape (6, number of examples)Y -- "true" labels vector placeholder, same shape as Z3Returns:cost - Tensor of the cost function"""# to fit the tensorflow requirement for tf.nn.softmax_cross_entropy_with_logits(...,...)logits = tf.transpose(Z3)labels = tf.transpose(Y)# 函數(shù)輸入：shape =（樣本數(shù)，類(lèi)數(shù)）# tf.reduce_mean()cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits = logits, labels = labels))return cost def predict(X, parameters):W1 = tf.convert_to_tensor(parameters["W1"])b1 = tf.convert_to_tensor(parameters["b1"])W2 = tf.convert_to_tensor(parameters["W2"])b2 = tf.convert_to_tensor(parameters["b2"])W3 = tf.convert_to_tensor(parameters["W3"])b3 = tf.convert_to_tensor(parameters["b3"])params = {"W1": W1,"b1": b1,"W2": W2,"b2": b2,"W3": W3,"b3": b3}x = tf.placeholder("float", [12288, 1])z3 = forward_propagation(x, params)p = tf.argmax(z3)with tf.Session() as sess:prediction = sess.run(p, feed_dict = {x: X})return prediction

2.3.3 訓(xùn)練網(wǎng)絡(luò)

my_image = "my_image.jpg" fname = "images/" + my_imageimage = np.array(ndimage.imread(fname, flatten=False)) my_image = scipy.misc.imresize(image, size=(64,64)).reshape((1, 64*64*3)).T parameters = model(X_train, Y_train, X_test, Y_test)plt.imshow(image) plt.show()my_image_prediction = predict(my_image, parameters) print("Your algorithm predicts: y = " + str(np.squeeze(my_image_prediction)))

總結(jié)

以上是生活随笔為你收集整理的深度学习（02）-- ANN学习的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問(wèn)題。

如果覺(jué)得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。