Demo Project to show Gradient Descent<\/p>\n\n\n\n
import tkinter\nimport matplotlib\nimport matplotlib.pyplot as plt\n\nmatplotlib.use('tkagg')\n# %matplotlib inline\nimport random as random\nimport numpy as np\nimport csv\n\nx_data = [338., 333., 328., 207., 226., 25., 179., 60., 208., 606.]\ny_data = [640., 633., 619., 393., 428., 27., 193., 66., 226., 1591.]\n\nx = np.arange(-200, -100, 1) # bias\ny = np.arange(-5, 5, 0.1) # weight\nZ = np.zeros((len(x), len(y)))\nX, Y = np.meshgrid(x, y)\nfor i in range(len(x)):\n for j in range(len(y)):\n b = x[i]\n w = y[j]\n Z[j][i] = 0\n for n in range(len(x_data)):\n Z[j][i] = Z[j][i] + (y_data[n] - b - w * x_data[n]) ** 2\n Z[j][i] = Z[j][i] \/ len(x_data)\n\n# ydata = b + w * xdata\nb = -120 # initial b\nw = -4 # initial w\nlr = 0.000001 # learning rate\niteration = 100000\n\nb_lr = 0.0\nw_lr = 0.0\n\n# Store initial values for plotting.\nb_history = [b]\nw_history = [w]\n\n# Iterations\nfor i in range(iteration):\n\n b_grad = 0.0\n w_grad = 0.0\n for n in range(len(x_data)):\n b_grad = b_grad - 2.0 * (y_data[n] - b - w * x_data[n]) * 1.0\n w_grad = w_grad - 2.0 * (y_data[n] - b - w * x_data[n]) * x_data[n]\n\n b_lr = b_lr + b_grad ** 2\n w_lr = w_lr + w_grad ** 2\n\n # Update parameters.\n b = b - lr \/ np.sqrt(b_lr) * b_grad\n w = w - lr \/ np.sqrt(w_lr) * w_grad\n\n # Store parameters for plotting\n b_history.append(b)\n w_history.append(w)\n\n# plot the figure\nplt.contourf(x, y, Z, 50, alpha=0.5, cmap=plt.get_cmap('jet'))\nplt.plot([-188.4], [2.67], 'x', ms=12, markeredgewidth=3, color='orange')\nplt.plot(b_history, w_history, 'o-', ms=3, lw=1.5, color='black')\nplt.xlim(-200, -100)\nplt.ylim(-5, 5)\nplt.xlabel(r'$b$', fontsize=16)\nplt.ylabel(r'$w$', fontsize=16)\nplt.show()<\/code><\/pre>\n\n\n\nDEMO 1 task<\/strong>
\u7528 python \u8a08\u7b97\u67d0\u4e00\u500b\u65b0\u805e\u7db2\u9801\u51fa\u73fe Taiwan \u95dc\u9375\u5b57\u7684\u6b21\u6578<\/p>\n\n\n\nwith open('demo_1003.html') as f:\n lines = f.read()\n\nwords = lines.split()\n#print(words[0])\n\nsearch_word = \"Taiwan\"\ncount = words.count(search_word)\nprint(\"\u641c\u5c0b\u7d50\u679c(\", search_word, \")\", count)<\/code><\/pre>\n\n\n\nDEMO 2 task
<\/strong>1. \u767b\u5165
2. \u4fee\u6539\u66b1\u7a31
3. \u4fee\u6539\u5bc6\u78bc
4. \u4fee\u6539\u5927\u982d\u8cbc
\u53ef\u4ee5\u4f7f\u7528 Postman<\/strong>, \u4e0d\u904e\u6700\u597d\u662f\u5690\u8a66\u4f7f\u7528 python \u4f86\u9032\u884c.<\/p>\n\n\n\n\u4eca\u5929 API \u7d50\u679c URL<\/strong>
https:\/\/www.cutexyz.com\/edu\/mlearning2309\/API\/<\/a><\/p>\n\n\n\n\u5982\u679c\u8981\u770b\u900f\u904e\u5b78\u865f\u8207\u4e2d\u6587\u59d3\u540d\u767b\u5165\u7684\u7d50\u679c:
https:\/\/www.cutexyz.com\/edu\/mlearning2309\/API\/index2.php<\/a><\/p>\n\n\n\n\u4eca\u65e5\u6703\u4f7f\u7528\u7684 API List<\/p>\n\n\n\n
API 1<\/strong>: login
Protocol: HTTPS POST
API URL: https:\/\/www.cutexyz.com\/edu\/mlearning2309\/API\/api_login.php
Parameters<\/strong>:
1. student_uid: \u5b78\u865f
2. pwd: \u5bc6\u78bc, \u9810\u8a2d\u70ba cute301, \u6c92\u6709\u5fd8\u8a18\u5bc6\u78bc\u529f\u80fd, \u5982\u679c\u6709\u4eba\u5fd8\u8a18, \u8acb\u5148\u8207\u8001\u5e2b\u901a\u77e5
Return Data ( JSON FORMAT )<\/strong>:
1. result: \u7d50\u679c, OK \u4ee3\u8868\u6c92\u554f\u984c, ERROR \u4ee3\u8868\u932f\u8aa4\u767c\u751f
2. data: \u5167\u542b\u552f\u4e00\u8cc7\u6599 token: \u7528\u65bc\u767b\u5165\u4e4b\u5f8c\u547c\u53eb\u5225\u7684 API \u642d\u914d\u4f7f\u7528
3. msg: \u5982\u679c\u6709\u7279\u6b8a\u88dd\u6846\u9700\u8981\u8a73\u8ff0\u554f\u984c\u6642\u5c31\u6703\u5728\u9019\u908a\u986f\u793a<\/p>\n\n\n\n