#coding=gbk ''' 進階篇：多元回歸：建模問題：Y=x1xx1+x2xw2+x3xw3+...+xnxwn+b,矩陣簡化表示Y=XW+bnumpy庫基礎：整型的一個數字， 不能取得其shape（維度），通過np.array()可以將其轉換成一個標量（數組），可以取得shape；一般問題：需要對重要的特征以及次要特征進行范圍的限定，即需要對數據進行歸一化，有助于模型的收斂。特征值/(最大特征值-最小特征值)~[0,1]特征數據歸一化：特征值/(最大特征值-最小特征值)~[0,1]，標簽不做處理擴展篇 '''from pylab import mpl mpl.rcParams['font.sans-serif'] = ['SimHei'] #設置顯示繪圖顯示中文 mpl.rcParams['axes.unicode_minus'] = False #防止中文亂碼，有時候第一句不能完全避免顯示錯誤#導入tensorflow 模塊 import tensorflow.compat.v1 as tf import matplotlib.pyplot as plt import numpy as np import pandas as pd from sklearn.datasets import load_boston#下載數據 from sklearn.utils import shuffle#用于打亂數據#numpy庫運用 #將一個整型數字轉換成一個數組 s_v = 20 scalar_np = np.array(s_v) print("標量：\n",scalar_np,scalar_np.shape)#向量 vector_v = [1, 2, 3, 4, 5, 6, 7, 8, 9] vector_np = np.array(vector_v) print("向量：\n",vector_np, vector_np.shape)#矩陣 m_v = [[1, 2, 3], [7, 8, 9], [4, 5, 6]] m_np = np.array(m_v) print("矩陣：\n",m_np,m_np.shape)#行向量、列向量的表示 row = np.array([[1, 2, 3]]) print("行向量：\n",row,row.shape)column = np.array([[1], [2], [3]]) print("列向量：\n", column, column.shape)#矩陣的運算 a = np.array([[1, 2, 3], [4, 5, 6]]) print(a) a = a + 7 print(a) a = a * 2 print(a) a = a + a print(a) a = a - 3 print(a) a = a / 2.0 print(a) #對于+，-，*,/,都是對應元素做相應的操作#行列轉置:aij=aji，reshape() m = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]]) print(m.T)#矩陣的點積，對應元素相乘,np.multiply()#矩陣叉乘：兩矩陣維度為 MxN,NxK,使用函數np.matmul(a,b) m_a = np.array([[1, 2, 3]]) m_b = np.array([[3], [2], [1]]) m_c = np.matmul(m_a, m_b) print(m_c)#下載數據，設置return_X_y=True去除摘要只返回數據和標簽 boston_price_data ,_= load_boston(return_X_y=True) print(boston_price_data, boston_price_data.shape)#數據歸一化[0,1] for i in range(12):boston_price_data[:, i] = boston_price_data[:, i] / (boston_price_data[:, i].max() - boston_price_data[:, i].min()) #取出數據 x_d = boston_price_data[:, :12] print(x_d, x_d.shape) print("\n") #取出標簽 y_d = boston_price_data[:, 12] print(y_d, y_d.shape)#模型定義 x = tf.placeholder(tf.float32, [None, 12], name='x') y = tf.placeholder(tf.float32, [None, 1], name='y')#擬合Y=x1xx1+x2xw2+x3xw3+...+xnxwn+b,矩陣簡化表示Y=XW+b with tf.name_scope("model"):#將下面的子圖打包，使計算圖簡介，便于tensorboard查看w = tf.Variable(tf.random_normal([12, 1], stddev=0.01, name='w'))#隨機初始化w數組b = tf.Variable(1.0, name='b') #初始化bdef model(x, w, b):return tf.matmul(x, w) + b #矩陣叉乘predict = model(x, w, b) #預測模型#模型訓練 train_c = 80 learning_rate = 0.01 with tf.name_scope("LossFun"):#打包loss_Fun = tf.reduce_mean(tf.pow(y - predict, 2))#均方誤差 optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss_Fun) #優化器設置sess = tf.Session() init = tf.global_variables_initializer() sess.run(init)loss_l = [] for i in range(train_c):loss_s = 0.0for x_, y_ in zip(x_d, y_d):x_ = x_.reshape(1, 12)#需要將取出來的數據進行維度調整，是x_,y_適合x，y變量維度模型y_=y_.reshape(1,1)_, loss = sess.run([optimizer, loss_Fun], feed_dict={x: x_, y: y_})'''用于可視化的語句_,summary_s, loss = sess.run([optimizer,summary_loss_op, loss_Fun], feed_dict={x: x_, y: y_})write.add_summary(summary_s,i)'''loss_s = loss_s + lossx_v, y_v = shuffle(x_d, y_d) #打亂數據b0 = b.eval(session=sess)w0 = w.eval(session=sess)less_average = loss_s / len(y_d)loss_l.append(less_average)print("train count=", i + 1, "loss=", less_average, "b=", b0, "w=", w0)#模型驗證 x_test = x_d[430] x_test = x_test.reshape(1, 12) p = sess.run(predict, feed_dict={x: x_test}) print("預測值：%f"%p,"標簽值：%f\n"%y_d[430])plt.plot(loss_l) plt.title("損失變化曲線(loss)") plt.show()logdir = "E:/VSCODE/" if tf.gfile.Exists(logdir):tf.gfile.DeleteRecursively(logdir)'''tensorboard可視化數據， summary_loss_op=tf.summary.scalar("loss",loss_Fun)#記錄損失值loss，寫入到tensorboard中的SCALARS欄中 merged=tf.summary.merge_all()#將所有需要的日志文件合并寫入 ''' write = tf.summary.FileWriter(logdir, tf.get_default_graph()) write.close()

附：
本文章學習至中國大學mooc-深度學習應用開發-Tensorflow實戰

總結

以上是生活随笔為你收集整理的Tensorflow深度学习应用(进阶篇)-回归(函数拟合训练)-可视化的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。