![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/1554344732112.png\")
<\/div><\/p>\n\u673a\u5668\u5b66\u4e60\u5df2\u7ecf\u4e0d\u77e5\u4e0d\u89c9\u7684\u8d70\u5165\u4e86\u6211\u4eec\u7684\u751f\u6d3b\uff0c\u6211\u4eec\u53ef\u80fd\u65e0\u6cd5\u5e72\u6d89\u5b83\u7684\u5d1b\u8d77\uff0c\u4e5f\u53ef\u80fd\u65e0\u6cd5\u521b\u5efa\u5982\u5341\u5927\u7b97\u6cd5\u4e00\u6837\u7684\u7b97\u6cd5\u3002\u4f46\u73b0\u5982\u4eca\u751f\u6d3b\u7684\u65b9\u65b9\u9762\u9762\u90fd\u6709\u7740\u673a\u5668\u5b66\u4e60\u7684\u8eab\u5f71\uff0c\u4e3a\u4ec0\u4e48\u6211\u4eec\u4e0d\u8bd5\u7740\u53bb\u8ba4\u8bc6\u5b83\u5462\uff0c\u8ba9\u6211\u4eec\u80fd\u5728\u673a\u5668\u5b66\u4e60\u5f7b\u5e95\u964d\u4e34\u4e4b\u524d\u505a\u597d\u9a7e\u9a6d\u5b83\u7684\u51c6\u5907\u5462\uff1f<\/p>\n
![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/\u5b87\u5b99\u98de\u8239\u9a7e\u9a76\u8bc1.png\")
<\/div><\/p>\n1.1 \u5b66\u4e60\u76ee\u6807<\/h1>\n\n- \u4e86\u89e3\u4eba\u5de5\u667a\u80fd\u3001\u673a\u5668\u5b66\u4e60\u548c\u6df1\u5ea6\u5b66\u4e60\u4e4b\u95f4\u7684\u533a\u522b<\/li>\n
- \u638c\u63e1\u673a\u5668\u5b66\u4e60\u4e2d\u7684\u76d1\u7763\u5b66\u4e60\u548c\u65e0\u76d1\u7763\u5b66\u4e60\u95ee\u9898<\/li>\n<\/ol>\n
1.2 \u4eba\u5de5\u667a\u80fd<\/h1>\n
![\"\"](\"data:image\/svg+xml;base64,PCEtLUFyZ29uTG9hZGluZy0tPgo8c3ZnIHdpZHRoPSIxIiBoZWlnaHQ9IjEiIHhtbG5zPSJodHRwOi8vd3d3LnczLm9yZy8yMDAwL3N2ZyIgc3Ryb2tlPSIjZmZmZmZmMDAiPjxnPjwvZz4KPC9zdmc+\")
<\/div><\/p>\n\u4eba\u5de5\u667a\u80fd\uff08artificial intelligence, AI\uff09\uff1a<\/strong>\u901a\u8fc7\u673a\u5668\u6765\u6a21\u62df\u4eba\u7c7b\u8ba4\u77e5\u7684\u6280\u672f\u3002<\/p>\n \u4eba\u5de5\u667a\u80fd\u88ab\u65b0\u95fb\u5a92\u4f53\u5439\u6367\u7684\u8ba9\u4eba\u9065\u4e0d\u53ef\u53ca\uff0c\u4f46\u662f\u7ec6\u5fc3\u7684\u4f60\u53ef\u80fd\u4f1a\u53d1\u73b0\u4eba\u5de5\u667a\u80fd\u5176\u5b9e\u5df2\u7ecf\u8d70\u5165\u4e86\u6211\u4eec\u7684\u751f\u6d3b\uff0c\u5e76\u4e14\u5df2\u7ecf\u6b65\u5165\u4ee5\u4e0b\u51e0\u4e2a\u9886\u57df\u3002<\/p>\n
\n- \u4eba\u8138\u8bc6\u522b\u5e94\u7528\u4e2d\uff0c\u4ed6\u6839\u636e\u8f93\u5165\u7684\u7167\u7247\uff0c\u5224\u65ad\u7167\u7247\u4e2d\u7684\u4eba\u662f\u8c01<\/li>\n
- \u533b\u7597\u8bca\u65ad\u4e2d\uff0c\u4ed6\u6839\u636e\u8f93\u5165\u7684\u533b\u7597\u5f71\u50cf\uff0c\u5224\u65ad\u75be\u75c5\u7684\u6210\u56e0<\/li>\n
- \u8bed\u97f3\u8bc6\u522b\u4e2d\uff0c\u4ed6\u6839\u636e\u4eba\u8bf4\u8bdd\u7684\u97f3\u9891\u4fe1\u53f7\uff0c\u5224\u65ad\u8bf4\u8bdd\u7684\u5185\u5bb9<\/li>\n
- \u7535\u5b50\u5546\u52a1\u7f51\u7ad9\u4e2d\uff0c\u4ed6\u6839\u636e\u7528\u6237\u66fe\u7ecf\u7684\u8d2d\u4e70\u8bb0\u5f55\uff0c\u9884\u6d4b\u7528\u6237\u611f\u5174\u8da3\u7684\u5546\u54c1\uff0c\u7f51\u7ad9\u8fdb\u800c\u8fbe\u5230\u63a8\u8350\u5546\u54c1\u7684\u76ee\u7684<\/li>\n
- \u81ea\u52a8\u9a7e\u9a76\u5e94\u7528\u4e2d\uff0c\u4ed6\u6839\u636e\u5bf9\u5f53\u524d\u6c7d\u8f66\u6240\u5904\u60c5\u51b5\u7684\u5206\u6790\uff0c\u5224\u65ad\u6c7d\u8f66\u63a5\u4e0b\u6765\u7684\u901f\u5ea6\u548c\u65b9\u5411<\/li>\n
- \u91d1\u878d\u5e94\u7528\u4e2d\uff0c\u4ed6\u6839\u636e\u80a1\u7968\u66fe\u7ecf\u7684\u4ef7\u683c\u548c\u5176\u4ed6\u4ea4\u6613\u4fe1\u606f\uff0c\u9884\u6d4b\u80a1\u7968\u672a\u6765\u7684\u4ef7\u683c\u8d70\u52bf<\/li>\n
- \u56f4\u68cb\u5bf9\u5f08\u4e2d\uff0c\u4ed6\u6839\u636e\u5f53\u524d\u7684\u76d8\u9762\u5f62\u5f0f\uff0c\u9884\u6d4b\u5728\u54ea\u4e2a\u5730\u65b9\u843d\u5b50\u80dc\u7387\u6700\u5927\uff082016\u5e743\u6708\u7684alphago\u5927\u6218\u674e\u4e16\u77f3\uff09<\/li>\n
- \u57fa\u56e0\u6d4b\u5e8f\u5e94\u7528\u4e2d\uff0c\u4ed6\u6839\u636e\u5bf9\u4eba\u7684\u57fa\u56e0\u5e8f\u5217\u7684\u5206\u6790\uff0c\u9884\u6d4b\u8fd9\u4e2a\u4eba\u672a\u6765\u60a3\u75c5\u7684\u53ef\u80fd\u6027<\/li>\n
- \u667a\u80fd\u5bb6\u5c45\u3001\u667a\u80fd\u73a9\u5177\u3001\u7f51\u7edc\u5b89\u5168\u2026\u2026<\/li>\n<\/ol>\n
\u6700\u540e\uff0c\u5f15\u7528\u300a\u7ec8\u6781\u7b97\u6cd5\u300b\u4e2d\u7684\u4e00\u6bb5\u8bdd\uff1a"\u6211\u4eec\u53ef\u80fd\u65e0\u6cd5\u963b\u6321\u4eba\u5de5\u667a\u80fd\u53d1\u5c55\u7684\u8d8b\u52bf\uff0c\u4f46\u662f\u6211\u4eec\u53ef\u4ee5\u5b66\u4f1a\u548c\u4ed6\u505a\u670b\u53cb\u3002"<\/p>\n
1.3 \u673a\u5668\u5b66\u4e60<\/h1>\n
\u673a\u5668\u5b66\u4e60\uff08machine learning\uff09\uff1a<\/strong>\u901a\u8fc7\u5b66\u4e60\u83b7\u5f97\u8fdb\u884c\u9884\u6d4b\u6216\u5224\u65ad\u7684\u80fd\u529b\u7684\u4eba\u5de5\u667a\u80fd\u3002<\/p>\n \u673a\u5668\u5b66\u4e60\u662f\u4eba\u5de5\u667a\u80fd\u5b9e\u73b0\u7684\u4e00\u79cd\u65b9\u6cd5\uff08\u7b97\u6cd5\uff09\u3002\u4ed6\u4e3b\u8981\u662f\u4ece\u5df2\u77e5\u6570\u636e\u4e2d\u53bb\u5b66\u4e60\u6570\u636e\u4e2d\u8574\u542b\u7684\u89c4\u5f8b\u6216\u8005\u5224\u65ad\u89c4\u5219\uff0c\u4e5f\u53ef\u4ee5\u7406\u89e3\u6210\u628a\u65e0\u5e8f\u7684\u4fe1\u606f\u53d8\u5f97\u6709\u5e8f\uff0c\u7136\u540e\u4ed6\u901a\u8fc7\u628a\u8fd9\u79cd\u89c4\u5f8b\u5e94\u7528\u5230\u672a\u6765\u7684\u65b0\u6570\u636e\u4e0a\uff0c\u5e76\u5bf9\u65b0\u6570\u636e\u505a\u51fa\u5224\u65ad\u6216\u9884\u6d4b\u3002<\/p>\n
\u4f8b\u5982\u67d0\u4e2a\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u4ece\u5168\u740370\u4ebf\u4eba\u7c7b\u7684\u6570\u636e\u96c6\u4e2d\u5b66\u5f97\u4e86\u67d0\u79cd\u5224\u65ad\u89c4\u5219\uff0c\u8fd9\u4e2a\u5224\u65ad\u89c4\u5219\u53ef\u4ee5\u901a\u8fc7\u8f93\u5165Nick\u8001\u5e08\u7684\u8eab\u9ad8180cm\u3001\u4f53\u91cd70Kg\uff0c\u5224\u65ad\u4ed6\u662f\u4e2a\u5e05\u54e5\u3002
\n
<\/div><\/p>\n1.3.1 \u673a\u5668\u5b66\u4e60\u57fa\u672c\u672f\u8bed<\/h2>\n# \u673a\u5668\u5b66\u4e60\u57fa\u672c\u672f\u8bed\u56fe\u4f8b\nimport matplotlib.pyplot as plt\nfrom matplotlib.font_manager import FontProperties\n# \u663e\u793aplt.show()\u7684\u56fe\u7247\uff0c\u5982\u82e5\u4e0d\u4f7f\u7528jupyter\uff0c\u8bf7\u6ce8\u91ca\n%matplotlib inline\n# \u4e2d\u6587\u5b57\u4f53\u8bbe\u7f6e\uff0c\u627e\u5230\u4e0e\u4e4b\u5bf9\u5e94\u7684\u4e2d\u6587\u5b57\u4f53\u8def\u5f84\nfont = FontProperties(fname='\/Library\/Fonts\/Heiti.ttc')\n\n# \u6837\u672c\nplt.text(1, 70, s='{\u8eab\u9ad8(180)\uff0c\u4f53\u91cd(70)\uff0c\u5e74\u9f84(19)\uff0c\u4e94\u5b98(\u7cbe\u81f4)',\n fontproperties=font, fontsize=13, color='g')\nplt.annotate(text='\u6837\u672c', xytext=(90, 70), xy=(78, 71), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='\u7279\u5f81', xytext=(30, 90), xy=(10, 73), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 90), xy=(30, 73), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 90), xy=(50, 73), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 90), xy=(70, 73), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\n\nplt.hlines(67, 0, 100, linestyle='--', color='gray')\n\n# \u7279\u5f81\nplt.text(1, 40, s='{\u8eab\u9ad8(180)\uff0c\u4f53\u91cd(70)\uff0c\u5e74\u9f84(19)\uff0c\u4e94\u5b98(\u7cbe\u81f4)\uff0c\u5e05}',\n fontproperties=font, fontsize=10, color='g')\nplt.annotate(text='\u6837\u4f8b(\u5b9e\u4f8b)', xytext=(90, 40), xy=(70, 41), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='\u7279\u5f81', xytext=(30, 60), xy=(8, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 60), xy=(23, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 60), xy=(38, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 60), xy=(53, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='\u6807\u8bb0', xytext=(67, 60), xy=(67, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\n\nplt.hlines(37, 0, 100, linestyle='--', color='gray')\n\n# \u7279\u5f81\u7a7a\u95f4\nplt.text(1, 30, s='$\\{(x_1^{(1)},x_1^{(2)},\\cdots,x_1^{(n)}),(x_2^{(1)},x_2^{(2)},\\cdots,x_2^{(n)}),\\cdots,(x_m^{(1)},x_m^{(2)},\\cdots,x_m^{(n)})\\}$',\n fontproperties=font, fontsize=10, color='g')\nplt.annotate(text='\u7279\u5f81\u7a7a\u95f4($m$\u4e2a\u6837\u672c)', xytext=(90, 15), xy=(92, 30), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='\u7279\u5f81\u5411\u91cf($n$\u7ef4\u7279\u5f81)', xytext=(15, 15), xy=(15, 28), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\n\nplt.xlim(0, 100)\nplt.ylim(10, 100)\nplt.title('\u673a\u5668\u5b66\u4e60\u57fa\u672c\u672f\u8bed\u56fe\u4f8b', fontproperties=font, fontsize=20)\nplt.show()<\/code><\/pre>\n<\/div><\/p>\n\u901a\u8fc7\u4e0a\u56fe\uff0c\u6211\u4eec\u53ef\u4ee5\u4e60\u5f97\u4ee5\u4e0b\u673a\u5668\u5b66\u4e60\u7684\u57fa\u672c\u672f\u8bed\uff1a<\/p>\n
\n- \u7279\u5f81\uff08feature\uff09\uff1a\u63cf\u8ff0\u4e00\u4ef6\u4e8b\u7269\u7684\u7279\u6027\uff0c\u5982\u4e00\u4e2a\u4eba\u7684\u8eab\u9ad8\u3001\u4f53\u91cd\u3001\u5e74\u9f84\u548c\u4e94\u5b98\u3002<\/li>\n
- \u6837\u672c\uff08sample\uff09\uff1a\u7531\u4e00\u4e2a\u4eba\u7684\u7279\u5f81\u7ec4\u6210\u7684\u6570\u636e\uff0c\u5982${180,70,19,\u7cbe\u81f4}$\u3002<\/li>\n
- \u6807\u8bb0\uff08label\uff09\uff1a\u63cf\u8ff0\u4e00\u4ef6\u4e8b\u7269\u7684\u7279\u6027\uff0c\u5982\u4e00\u4e2a\u4eba\u5e05\u6216\u4e11\u3001\u4e00\u4e2a\u4eba\u7684\u8d22\u5bcc\u6570\u91cf\u3002\u6ce8\uff1a\u7279\u5f81\u548c\u6807\u8bb0\u6ca1\u6709\u660e\u786e\u7684\u5212\u5206\uff0c\u7531\u4e8e\u95ee\u9898\u7684\u4e0d\u540c\u53ef\u80fd\u5bfc\u81f4A\u95ee\u9898\u7684\u7279\u5f81\u662fB\u95ee\u9898\u7684\u6807\u8bb0\uff0cB\u95ee\u9898\u7684\u6807\u8bb0\u662fA\u95ee\u9898\u7684\u7279\u5f81\u3002<\/strong><\/li>\n
- \u6837\u4f8b\uff08example\uff09\uff1a\u7531\u4e00\u4e2a\u4eba\u7684\u7279\u5f81\u548c\u6807\u8bb0\u7ec4\u6210\u7684\u6570\u636e\uff0c\u5982${180,70,19,\u7cbe\u81f4,\u5e05}$\u3002<\/li>\n
- \u7279\u5f81\u7a7a\u95f4\uff08feature space\uff09\uff1a\u7531${x^{(1)},x^{(2)},\\cdots,x^{(n)}}$\u8fd9n\u4e2a\u7279\u5f81\u5f20\u6210\u7684n\u7ef4\u7a7a\u95f4\uff0c\u5982\u4ee5\u8eab\u9ad8\u5f20\u6210\u7684\u4e00\u7ef4\u7a7a\u95f4\uff08\u7ebf\uff09\uff1b\u4ee5\u8eab\u9ad8\u548c\u4f53\u91cd\u5f20\u6210\u7684\u4e8c\u7ef4\u7a7a\u95f4\uff08\u9762\uff09\uff1b\u4ee5\u8eab\u9ad8\u3001\u4f53\u91cd\u548c\u5e74\u9f84\u5f20\u6210\u7684\u4e09\u7ef4\u7a7a\u95f4\uff08\u4f53\uff09\u3002<\/li>\n
- \u7279\u5f81\u5411\u91cf\uff08feature vector\uff09\uff1a\u7279\u5f81\u7a7a\u95f4\u5185\u7684\u67d0\u4e00\u4e2a\u5177\u4f53\u7684\u5411\u91cf\uff0c\u5373\u7279\u79cd\u7a7a\u95f4\u4e2d\u7684\u67d0\u4e00\u4e2a\u5177\u4f53\u7684\u70b9\uff0c
\n<\/div>
\n\u5176\u4e2d$x_m^{(n)}$\u8868\u793a\u7b2cm\u4e2a\u4eba\u7684\u7b2cn\u4e2a\u7279\u5f81\u3002\u5982\u8eab\u9ad8\u3001\u4f53\u91cd\u3001\u5e74\u9f84\u5f20\u6210\u7684\u4e09\u7ef4\u7a7a\u95f4\u4e2d\u7684\u67d0\u4e00\u4e2a\u5177\u4f53\u7684\u70b9$(180,70,19)$\u3002<\/li>\n<\/ul>\n1.4 \u6df1\u5ea6\u5b66\u4e60<\/h1>\n
\u6df1\u5ea6\u5b66\u4e60\uff08deep learning\uff09\uff1a<\/strong>\u901a\u8fc7\u7ec4\u5408\u4f4e\u5c42\u5c5e\u6027\u7279\u5f81\u5f62\u6210\u66f4\u52a0\u62bd\u8c61\u7684\u9ad8\u5c42\u5c5e\u6027\u7279\u5f81\u3002\u5982\u5b66\u82f1\u8bed\u7684\u65f6\u5019\uff0c\u201cNick handsome\u201d\u8fd912\u4e2a\u82f1\u6587\u5355\u8bcd\u662f\u4f4e\u5c42\u5c5e\u6027\u7279\u5f81\uff0c\u800c\u201cNick handsome\u201d\u7684\u8bed\u4e49\u5219\u662f\u62bd\u8c61\u7684\u9ad8\u5c42\u5c5e\u6027\u7279\u5f81\u3002<\/p>\n\u63d2\u56fe\uff1a\u6076\u641e\u56fe03
\n
<\/div><\/p>\n \u6df1\u5ea6\u5b66\u4e60\u7684\u6982\u5ff5\u6e90\u4e8e\u4eba\u5de5\u795e\u7ecf\u7f51\u7edc\u7684\u7814\u7a76\uff0c\u5b83\u5c5e\u4e8e\u673a\u5668\u5b66\u4e60\u4e2d\u7684\u67d0\u4e00\u4e2a\u65b9\u6cd5\uff0c\u5176\u4e2d\u6df1\u5ea6\u5b66\u4e60\u4e2d\u7684\u201c\u6df1\u5ea6\u201d\u662f\u6307\u795e\u7ecf\u7f51\u7edc\u7684\u5c42\u6570\u3002<\/p>\n
<\/div><\/p>\n1.5 \u673a\u5668\u5b66\u4e60\u5206\u7c7b<\/h1>\n
\u673a\u5668\u5b66\u4e60\u4f9d\u636e\u6570\u636e\u96c6\u6570\u636e\u683c\u5f0f\u7684\u4e0d\u540c\uff0c\u53ef\u4ee5\u5212\u5206\u6210\u76d1\u7763\u5b66\u4e60\u548c\u65e0\u76d1\u7763\u5b66\u4e60\uff1b\u4f9d\u636e\u7b97\u6cd5\u6a21\u5f0f\u7684\u4e0d\u540c\u53ef\u4ee5\u5212\u5206\u4e3a\u76d1\u7763\u5b66\u4e60\u3001\u65e0\u76d1\u7763\u5b66\u4e60\u3001\u534a\u76d1\u7763\u5b66\u4e60\u548c\u5f3a\u5316\u5b66\u4e60\u3002<\/p>\n
<\/div><\/p>\n1.5.1 \u76d1\u7763\u5b66\u4e60<\/h2>\n
\u76d1\u7763\u5b66\u4e60\uff08supervised learning\uff09\uff1a<\/strong>\u901a\u8fc7\u8f93\u5165\u4e00\u7ec4\u5df2\u77e5\u6807\u8bb0\u7684\u6837\u672c\uff0c\u8f93\u51fa\u4e00\u4e2a\u6a21\u578b\uff08model\uff09\uff0c\u7136\u540e\u901a\u8fc7\u8fd9\u4e2a\u6a21\u578b\u9884\u6d4b\u672a\u6765\u65b0\u6570\u636e\u7684\u9884\u6d4b\u503c\u6216\u9884\u6d4b\u7c7b\u522b\u3002<\/p>\n \u76d1\u7763\u5b66\u4e60\u7684\u6d41\u7a0b\uff1a<\/p>\n
\n- \u8f93\u5165\uff1a\u4e00\u7ec4\u5df2\u77e5\u7c7b\u522b\u7684\u6837\u672c<\/li>\n
- \u8f93\u51fa\uff1a\u4e00\u4e2a\u6a21\u578b<\/li>\n
- \u9884\u6d4b\u672a\u6765\u65b0\u6570\u636e\u7684\u9884\u6d4b\u503c\u6216\u9884\u6d4b\u7c7b\u522b<\/li>\n<\/ol>\n
\u7531\u4e8e\u6570\u636e\u6807\u8bb0\u7684\u7c7b\u578b\u4e0d\u540c\uff0c\u76d1\u7763\u5b66\u4e60\u95ee\u9898\u53c8\u5206\u4e3a\u56de\u5f52\u95ee\u9898\u548c\u5206\u7c7b\u95ee\u9898\uff1a<\/p>\n
\n- \u5982\u679c\u6a21\u578b\u7684\u8f93\u51fa\u4e3a\u8fde\u7eed\u503c\uff0c\u5982\u80a1\u7968\u4ef7\u683c$2.2,2.4,2.6,3\u2026$\u7c7b\u578b\u7684\u6570\u636e\u79f0\u4e3a\u8fde\u7eed\u503c\uff0c\u5219\u8be5\u76d1\u7763\u5b66\u4e60\u95ee\u9898\u79f0\u4f5c\u56de\u5f52\u95ee\u9898<\/li>\n
- \u5982\u679c\u6a21\u578b\u7684\u8f93\u51fa\u4e3a\u79bb\u6563\u503c\uff0c\u5982$\u963f\u732b,\u963f\u72d7,\u2026$\u6216${0,1}$\uff08\u8f93\u51fa\u4e0d\u662f$0$\u5c31\u662f$1$\uff09\uff0c\u5219\u8be5\u76d1\u7763\u5b66\u4e60\u95ee\u9898\u79f0\u4f5c\u5206\u7c7b\u95ee\u9898\u3002<\/li>\n<\/ul>\n
1.5.1.2 \u56de\u5f52\u95ee\u9898<\/h3>\n
\n\n\n\u5e8f\u53f7<\/th>\n \u623f\u5b50\u9762\u79ef(m$^2$)<\/th>\n \u623f\u4ef7(\u5143)<\/th>\n<\/tr>\n<\/thead>\n \n\n1<\/td>\n 10<\/td>\n 100<\/td>\n<\/tr>\n \n2<\/td>\n 20<\/td>\n 200<\/td>\n<\/tr>\n \n3<\/td>\n 40<\/td>\n 400<\/td>\n<\/tr>\n \n4<\/td>\n 100<\/td>\n 1000<\/td>\n<\/tr>\n \n5<\/td>\n 200<\/td>\n 2000<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n# \u56de\u5f52\u95ee\u9898\u56fe\u4f8b\nimport numpy as np\nimport matplotlib.pyplot as plt\nfrom matplotlib.font_manager import FontProperties\n%matplotlib inline\nfont = FontProperties(fname='\/Library\/Fonts\/Heiti.ttc')\n\nx = np.linspace(0, 200, 10)\ny = 10*x\n\nplt.plot(x, y)\nplt.xlabel('\u623f\u5b50\u9762\u79ef', fontproperties=font, fontsize=15)\nplt.ylabel('\u623f\u4ef7', fontproperties=font, fontsize=15)\nplt.title('\u56de\u5f52\u95ee\u9898\u56fe\u4f8b', fontproperties=font, fontsize=20)\nplt.show()<\/code><\/pre>\n<\/div><\/p>\n \u901a\u8fc7\u4e0a\u8868\u7ed9\u51fa\u7684\u6570\u636e\u96c6\uff08\u4e00\u7ec4\u5df2\u77e5\u6807\u8bb0\u7684\u6837\u672c\uff09\u4e3e\u4f8b\uff0c\u901a\u8fc7\u6570\u636e\u96c6\u4e2d\u7684\u6570\u636e\uff0c\u6211\u4eec\u53ef\u4ee5\u5047\u8bbe\u623f\u5b50\u9762\u79ef\u548c\u623f\u4ef7\u4e4b\u95f4\u5b58\u5728\u67d0\u79cd\u5173\u7cfb\uff08\u6a21\u578b\uff09\u4e3a
\n$$
\n\u623f\u4ef7=10\u623f\u5b50\u9762\u79ef
\n$$
\n \u5047\u8bbe\u4f60\u5728\u5f97\u5230\u623f\u5b50\u9762\u79ef\u548c\u623f\u4ef7\u4e4b\u95f4\u7684\u5173\u7cfb\u540e\uff0c\u5f97\u5230\u4e86\u4e0a\u6d77\u4e8c\u73af\u67d0\u4e00\u6240\u623f\u5b50$H_1$\u7684\u9762\u79ef\u4e3a$1000m^2$\u7684\u6d88\u606f\uff08\u672a\u6765\u65b0\u6570\u636e\uff09\uff0c\u8fd9\u4e2a\u65f6\u5019\u4f60\u5c31\u53ef\u4ee5\u901a\u8fc7$H_1$\u7684\u623f\u5b50\u9762\u79ef\u548c\u623f\u4ef7\u4e4b\u95f4\u7684\u5173\u7cfb\u5f97\u5230\u8fd9\u6240\u623f\u5b50\u7684\u4ef7\u683c\u4e3a
\n$$
\nH_1\u7684\u623f\u4ef7=10<\/em>H_1\u7684\u623f\u5b50\u9762\u79ef=10*1000=10000(\u5143)\\quad\\text{\u9884\u6d4b\u503c}
\n$$<\/p>\n\u5728\u8fd9\u4e2a\u56de\u5f52\u95ee\u9898\u4e2d\uff1a<\/p>\n
\n- \u623f\u5b50\u9762\u79ef\u79f0\u4e3a\u7279\u5f81\uff08feature\uff09<\/li>\n
- \u623f\u4ef7\u79f0\u4e3a\u6807\u8bb0\uff08label\uff09<\/li>\n<\/ol>\n<\/div><\/p>\n
1.5.1.3 \u5206\u7c7b\u95ee\u9898<\/h3>\n
\n\n\n\u5e8f\u53f7<\/th>\n \u623f\u5b50\u9762\u79ef(m$^2$)<\/th>\n \u623f\u5b50\u6240\u5728\u533a\u57df<\/th>\n<\/tr>\n<\/thead>\n \n\n1<\/td>\n 10<\/td>\n A\u533a<\/td>\n<\/tr>\n \n2<\/td>\n 20<\/td>\n A\u533a<\/td>\n<\/tr>\n \n3<\/td>\n 40<\/td>\n A\u533a<\/td>\n<\/tr>\n \n4<\/td>\n 100<\/td>\n B\u533a<\/td>\n<\/tr>\n \n5<\/td>\n 200<\/td>\n B\u533a<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n# \u5206\u7c7b\u95ee\u9898\u56fe\u4f8b\nimport numpy as np\nimport matplotlib.pyplot as plt\nfrom matplotlib.font_manager import FontProperties\n%matplotlib inline\nfont = FontProperties(fname='\/Library\/Fonts\/Heiti.ttc')\n\na_area = [10, 20, 40]\na_y = [0, 0, 0]\n\nb_area = [100, 200]\nb_y = [0, 0]\n\nplt.scatter(a_area, a_y, label='A', color='r')\nplt.scatter(b_area, b_y, label='B', color='g')\nplt.vlines(50, ymin=-0.015, ymax=0.015, color='gray', linestyles='--')\nplt.xlabel('\u623f\u5b50\u9762\u79ef', fontproperties=font, fontsize=15)\nplt.title('\u5206\u7c7b\u95ee\u9898\u56fe\u4f8b', fontproperties=font, fontsize=20)\nplt.legend()\nplt.show()<\/code><\/pre>\n<\/div><\/p>\n \u540c\u6837\u901a\u8fc7\u4e0a\u8868\u7ed9\u51fa\u7684\u6570\u636e\u96c6\u4e3e\u4f8b\uff0c\u901a\u8fc7\u6570\u636e\u96c6\u7684\u6570\u636e\uff0c\u6211\u4eec\u53ef\u4ee5\u5047\u8bbe\u623f\u5b50\u9762\u79ef\u548c\u623f\u5b50\u6240\u5728\u533a\u57df\u5b58\u5728\u67d0\u79cd\u5173\u7cfb\u4e3a
\n
<\/div><\/p>\n \u5176\u4e2d$P(A\u533a|\u623f\u5b50\u9762\u79ef<50m^2)$\u4e3a\u6761\u4ef6\u6982\u7387\uff0c\u5373\u5f53\u6ee1\u8db3$\u623f\u5b50\u9762\u79ef<50m^2$\u6761\u4ef6\u65f6\u623f\u5b50\u6240\u5728\u533a\u57df\u4e3a$A\u533a$\u7684\u6982\u7387\uff1b$P(B\u533a|\u623f\u5b50\u9762\u79ef>50m^2)$\u4e5f\u4e3a\u6761\u4ef6\u6982\u7387\uff0c\u5f53\u6ee1\u8db3$\u623f\u5b50\u9762\u79ef>50m^2$\u6761\u4ef6\u65f6\u623f\u5b50\u6240\u5728\u533a\u57df\u4e3a$B$\u533a\u7684\u6982\u7387\u3002<\/p>\n
\u5047\u8bbe\u4f60\u5728\u5f97\u5230\u623f\u5b50\u9762\u79ef\u548c\u623f\u5b50\u6240\u5728\u533a\u57df\u7684\u5173\u7cfb\u540e\uff0c\u5f97\u5230\u4e86\u4e0a\u6d77\u4e8c\u73af\u5185\u4e24\u6240\u623f\u5b50$H_1$\u548c$H_2$\u9762\u79ef\u5206\u522b\u4e3a$30m^2$\u548c$1000m^2$\u7684\u6d88\u606f\uff0c\u8fd9\u4e2a\u65f6\u5019\u4f60\u5c31\u53ef\u4ee5\u901a\u8fc7\u623f\u5b50\u9762\u79ef\u548c\u623f\u5b50\u6240\u5728\u533a\u57df\u4e4b\u95f4\u7684\u5173\u7cfb\u5f97\u5230\u8fd9\u4e24\u6240\u623f\u5b50\u6240\u5728\u7684\u533a\u57df
\n
<\/div><\/p>\n\u5728\u8fd9\u4e2a\u5206\u7c7b\u95ee\u9898\u4e2d\uff1a<\/p>\n
\n- \u623f\u5b50\u9762\u79ef\u79f0\u4e3a\u7279\u5f81<\/li>\n
- \u623f\u5b50\u6240\u5728\u533a\u57df\u79f0\u4e3a\u6807\u8bb0<\/li>\n<\/ol>\n<\/div><\/p>\n
1.5.2 \u65e0\u76d1\u7763\u5b66\u4e60<\/h2>\n
\u65e0\u76d1\u7763\u5b66\u4e60\uff08unsupervised learning\uff09\uff1a<\/strong>\u901a\u8fc7\u8f93\u5165\u4e00\u7ec4\u672a\u77e5\u6807\u8bb0\u7684\u6837\u672c\uff0c\u53ef\u4ee5\u901a\u8fc7\u805a\u7c7b\uff08clustering\uff09\u7684\u65b9\u6cd5\uff0c\u5c06\u6570\u636e\u5206\u6210\u591a\u4e2a\u7c07\uff08cluster\uff09\u3002<\/p>\n\u65e0\u76d1\u7763\u5b66\u4e60\u7684\u6d41\u7a0b\uff1a<\/p>\n
\n- \u8f93\u5165\uff1a\u4e00\u7ec4\u672a\u77e5\u7c7b\u522b\u7684\u6837\u672c<\/li>\n
- \u8f93\u51fa\uff1a\u5206\u6210\u591a\u4e2a\u7c07\u7684\u4e00\u7ec4\u6837\u672c<\/li>\n<\/ol>\n
1.5.2.1 \u805a\u7c7b<\/h3>\n
\u805a\u7c7b\uff08clustering\uff09\uff1a<\/strong>\u5c06\u6570\u636e\u96c6\u5206\u6210\u7531\u7c7b\u4f3c\u7684\u6570\u636e\u7ec4\u6210\u7684\u591a\u4e2a\u7c7b\u7684\u8fc7\u7a0b\u88ab\u79f0\u4e3a\u805a\u7c7b\u3002\u7531\u805a\u7c7b\u6240\u751f\u6210\u7684\u7c07\u662f\u4e00\u7ec4\u6570\u636e\u7684\u96c6\u5408\uff0c\u540c\u4e00\u4e2a\u7c07\u4e2d\u7684\u6570\u636e\u5f7c\u6b64\u76f8\u4f3c\uff0c\u4e0d\u540c\u7c07\u4e2d\u7684\u5bf9\u8c61\u76f8\u5f02\u3002<\/p>\n \u9700\u8981\u6ce8\u610f\u7684\u662f\uff0c\u7ec4\u6210\u591a\u4e2a\u7c07\u7684\u6982\u5ff5\u662f\u7531\u67d0\u79cd\u975e\u4eba\u4e3a\u5b9a\u4e49\u7684\u6f5c\u5728\u6982\u5ff5\u81ea\u52a8\u751f\u6210\u7684\uff0c\u800c\u6211\u4eec\u5728\u76d1\u7763\u5b66\u4e60\u4e2d\u7684\u5206\u7c7b\u90fd\u6709\u660e\u786e\u7684\u5b9a\u4e49\uff0c\u5373\u6309\u7167\u623f\u5b50\u9762\u79ef\u548c\u623f\u4ef7\u7684\u5173\u7cfb\u3001\u623f\u5b50\u9762\u79ef\u548c\u623f\u5b50\u6240\u5728\u533a\u57df\u7684\u5173\u7cfb\u3002<\/p>\n
\n\n\n\u5e8f\u53f7<\/th>\n \u623f\u5b50\u9762\u79ef(m$^2$)<\/th>\n<\/tr>\n<\/thead>\n \n\n1<\/td>\n 10<\/td>\n<\/tr>\n \n2<\/td>\n 20<\/td>\n<\/tr>\n \n3<\/td>\n 40<\/td>\n<\/tr>\n \n4<\/td>\n 100<\/td>\n<\/tr>\n \n5<\/td>\n 200<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n \u901a\u8fc7\u4e0a\u8868\u7ed9\u51fa\u7684\u6570\u636e\u96c6\u4e3e\u4f8b\uff0c\u901a\u8fc7\u805a\u7c7b\u53ef\u80fd\u628a\u6570\u636e\u96c6\u4e2d\u76845\u4e2a\u6570\u636e\u5206\u6210\u4e24\u4e2a\u7c07\uff0c\u5373\u623f\u5b50\u9762\u79ef\u5927\u4e8e50\u7684\u5206\u4e3a\u4e00\u4e2a\u7c07\uff0c\u623f\u5b50\u9762\u79ef\u5c0f\u4e8e50\u7684\u4e3a\u53e6\u4e00\u4e2a\u7c07\uff0c\u800c\u8fd9\u79cd\u901a\u8fc7\u9762\u79ef\u5206\u7c7b\u5e76\u4e0d\u662f\u6211\u4eec\u4e8b\u5148\u8bbe\u5b9a\u7684\uff0c\u800c\u662f\u7531\u805a\u7c7b\u7b97\u6cd5\u81ea\u5df1\u5b9e\u73b0\u7684\u3002\u4e5f\u5c31\u662f\u8bf4\u901a\u8fc7\u805a\u7c7b\u628a\u4e0a\u8868\u7684\u6570\u636e\u96c6\u5206\u4e3a\u4e24\u7c07\uff0c\u4f46\u662f\u6211\u4eec\u9700\u8981\u5728\u628a\u6570\u636e\u96c6\u5206\u6210\u4e24\u4e2a\u7c07\u540e\u518d\u53bb\u51b3\u5b9a\u7528\u4ec0\u4e48\u6982\u5ff5\u5b9a\u4e49\u8fd9\u4e24\u4e2a\u7c07\u7684\u5212\u5206\u3002<\/p>\n
<\/div><\/p>\n1.5.3 \u534a\u76d1\u7763\u5b66\u4e60<\/h2>\n
\u534a\u76d1\u7763\u5b66\u4e60\uff08semi-supervised learning\uff09\uff1a<\/strong>\u76d1\u7763\u5b66\u4e60\u4e0e\u65e0\u76d1\u7763\u5b66\u4e60\u76f8\u7ed3\u5408\u7684\u4e00\u79cd\u5b66\u4e60\u65b9\u6cd5\u3002<\/p>\n \u76d1\u7763\u5b66\u4e60\u548c\u65e0\u76d1\u7763\u5b66\u4e60\u7684\u533a\u522b\u4e3b\u8981\u5728\u4e8e\u6570\u636e\u96c6\u4e2d\u7684\u6570\u636e\u662f\u5426\u5177\u6709\u6807\u8bb0\uff0c\u5f53\u67d0\u4e2a\u95ee\u9898\u4e2d\u7684\u6570\u636e\u96c6\u4e2d\u7684\u6570\u636e\u5177\u6709\u6807\u8bb0\u65f6\uff0c\u6211\u4eec\u79f0\u4e4b\u4e3a\u76d1\u7763\u5b66\u4e60\uff1b\u53cd\u4e4b\uff0c\u6211\u4eec\u79f0\u4e4b\u4e3a\u65e0\u76d1\u7763\u5b66\u4e60\u3002<\/p>\n
\u76ee\u524d\uff0c\u7531\u4e8e\u65e0\u76d1\u7763\u5b66\u4e60\u7b97\u6cd5\u7684\u4e0d\u7a33\u5b9a\u6027\uff0c\u5b83\u4e00\u822c\u4f5c\u4e3a\u4e2d\u95f4\u7b97\u6cd5\uff0c\u5de5\u4e1a\u4e0a\u4f7f\u7528\u8f83\u591a\u7684\u662f\u76d1\u7763\u5b66\u4e60\u7b97\u6cd5\u3002\u4f46\u662f\u76ee\u524d\u5de5\u4e1a\u4e0a\u7d2f\u79ef\u8f83\u591a\u7684\u662f\u4e0d\u5e26\u6807\u8bb0\u7684\u6570\u636e\uff0c\u56e0\u6b64\u65e0\u76d1\u7763\u5b66\u4e60\u5728\u672a\u6765\u5c06\u662f\u4e00\u4e2a\u6280\u672f\u7a81\u7834\u7684\u91cd\u70b9\u76ee\u6807\u3002\u4e3a\u4e86\u5904\u7406\u8fd9\u6279\u4e0d\u5e26\u6807\u8bb0\u7684\u6570\u636e\uff0c\u4e5f\u6709\u79d1\u5b66\u5bb6\u63d0\u51fa\u4e86\u534a\u76d1\u7763\u5b66\u4e60\uff0c\u5b83\u53ef\u4ee5\u7b80\u5355\u7684\u7406\u89e3\u4e3a\u76d1\u7763\u5b66\u4e60\u548c\u65e0\u76d1\u7763\u5b66\u4e60\u7684\u7efc\u5408\u8fd0\u7528\uff0c\u5b83\u5373\u4f7f\u7528\u5e26\u6709\u6807\u8bb0\u7684\u6570\u636e\u4e5f\u4f7f\u7528\u4e0d\u5e26\u6709\u6807\u8bb0\u7684\u6570\u636e\u3002<\/p>\n
<\/div><\/p>\n1.5.4 \u5f3a\u5316\u5b66\u4e60<\/h2>\n
\u5f3a\u5316\u5b66\u4e60\uff08reinforcement learning\uff09\uff1a<\/strong>\u5b66\u4e60\u7cfb\u7edf\u4ece\u73af\u5883\u5230\u884c\u4e3a\u6620\u5c04\u7684\u5b66\u4e60\uff0c\u76ee\u6807\u662f\u4f7f\u667a\u80fd\u4f53\u83b7\u5f97\u6700\u5927\u7684\u5956\u52b1\uff0c\u5373\u5f3a\u5316\u4fe1\u53f7\u6700\u5927\u3002<\/p>\n \u5f3a\u5316\u5b66\u4e60\u7c7b\u4f3c\u4e8e\u65e0\u76d1\u7763\u5b66\u4e60\uff0c\u4f46\u662f\u53c8\u4e0d\u540c\u4e8e\u65e0\u76d1\u7763\u5b66\u4e60\uff0c\u5f3a\u5316\u5b66\u4e60\u9488\u5bf9\u7684\u662f\u65e0\u6807\u8bb0\u7684\u6570\u636e\uff0c\u4f46\u662f\u5728\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u5bf9\u65e0\u6807\u8bb0\u7684\u6570\u636e\u8fdb\u884c\u5904\u7406\u7684\u540c\u65f6\u4f1a\u7ed9\u51fa\u4e00\u4e2a\u7279\u5b9a\u7684\u76ee\u6807\uff0c\u5982\u679c\u7b97\u6cd5\u7684\u5904\u7406\u7ed3\u679c\u504f\u79bb\u76ee\u6807\uff0c\u5219\u4f1a\u7ed9\u51fa\u60e9\u7f5a\uff1b\u53cd\u4e4b\uff0c\u7ed9\u51fa\u5956\u52b1\uff0c\u800c\u8be5\u7b97\u6cd5\u7684\u76ee\u6807\u5c31\u662f\u4e3a\u4e86\u83b7\u5f97\u6700\u5927\u5956\u52b1\u3002\u5f3a\u5316\u5b66\u4e60\u4e00\u822c\u7528\u4e8e\u65e0\u4eba\u9a7e\u9a76\u3002<\/p>\n
1.6 Python\u8bed\u8a00\u7684\u4f18\u52bf<\/h1>\n
\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7Python\u3001R\u3001C\u3001Java\u3001MATLAB\u3001Octave\u7b49\u5de5\u5177\u5b9e\u73b0\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u3002\u672c\u6587\u63a5\u4e0b\u6765\u5c06\u901a\u8fc7Python\u5b9e\u73b0\uff0c\u56e0\u6b64\u5728\u8fd9\u91cc\u7b80\u5355\u8bf4\u8bf4Python\u7684\u4f18\u52bf\uff1a\u8bed\u6cd5\u6e05\u6670\u7b80\u5355\u3001\u6613\u4e8e\u64cd\u4f5c\u6587\u672c\u6587\u4ef6\u3001\u4f18\u79c0\u7684\u5f00\u6e90\u793e\u533a\u63d0\u4f9b\u4e86\u5927\u91cf\u7684\u5f00\u6e90\u8f6f\u4ef6\u5e93\uff0c\u5982scikit-learn\u3001tensorflow\u7b49\u3002<\/p>\n
\u672c\u6587\u57fa\u4e8ePython3.6+\u7248\u672c\uff0c\u4e0d\u5efa\u8bae\u4f7f\u7528Python2.7\u7248\u672c\u3002\u53ef\u4ee5\u4ecehttps:\/\/www.python.org\/downloads\/<\/a>\u4e0b\u8f7dPython\u3002<\/p>\n
1.2 \u4eba\u5de5\u667a\u80fd<\/h1>\n
![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/\u4eba\u5de5\u667a\u80fd-\u673a\u5668\u5b66\u4e60-\u6df1\u5ea6\u5b66\u4e60.png\")
<\/div><\/p>\n\u4eba\u5de5\u667a\u80fd\uff08artificial intelligence, AI\uff09\uff1a<\/strong>\u901a\u8fc7\u673a\u5668\u6765\u6a21\u62df\u4eba\u7c7b\u8ba4\u77e5\u7684\u6280\u672f\u3002<\/p>\n \u4eba\u5de5\u667a\u80fd\u88ab\u65b0\u95fb\u5a92\u4f53\u5439\u6367\u7684\u8ba9\u4eba\u9065\u4e0d\u53ef\u53ca\uff0c\u4f46\u662f\u7ec6\u5fc3\u7684\u4f60\u53ef\u80fd\u4f1a\u53d1\u73b0\u4eba\u5de5\u667a\u80fd\u5176\u5b9e\u5df2\u7ecf\u8d70\u5165\u4e86\u6211\u4eec\u7684\u751f\u6d3b\uff0c\u5e76\u4e14\u5df2\u7ecf\u6b65\u5165\u4ee5\u4e0b\u51e0\u4e2a\u9886\u57df\u3002<\/p>\n \u6700\u540e\uff0c\u5f15\u7528\u300a\u7ec8\u6781\u7b97\u6cd5\u300b\u4e2d\u7684\u4e00\u6bb5\u8bdd\uff1a"\u6211\u4eec\u53ef\u80fd\u65e0\u6cd5\u963b\u6321\u4eba\u5de5\u667a\u80fd\u53d1\u5c55\u7684\u8d8b\u52bf\uff0c\u4f46\u662f\u6211\u4eec\u53ef\u4ee5\u5b66\u4f1a\u548c\u4ed6\u505a\u670b\u53cb\u3002"<\/p>\n \u673a\u5668\u5b66\u4e60\uff08machine learning\uff09\uff1a<\/strong>\u901a\u8fc7\u5b66\u4e60\u83b7\u5f97\u8fdb\u884c\u9884\u6d4b\u6216\u5224\u65ad\u7684\u80fd\u529b\u7684\u4eba\u5de5\u667a\u80fd\u3002<\/p>\n \u673a\u5668\u5b66\u4e60\u662f\u4eba\u5de5\u667a\u80fd\u5b9e\u73b0\u7684\u4e00\u79cd\u65b9\u6cd5\uff08\u7b97\u6cd5\uff09\u3002\u4ed6\u4e3b\u8981\u662f\u4ece\u5df2\u77e5\u6570\u636e\u4e2d\u53bb\u5b66\u4e60\u6570\u636e\u4e2d\u8574\u542b\u7684\u89c4\u5f8b\u6216\u8005\u5224\u65ad\u89c4\u5219\uff0c\u4e5f\u53ef\u4ee5\u7406\u89e3\u6210\u628a\u65e0\u5e8f\u7684\u4fe1\u606f\u53d8\u5f97\u6709\u5e8f\uff0c\u7136\u540e\u4ed6\u901a\u8fc7\u628a\u8fd9\u79cd\u89c4\u5f8b\u5e94\u7528\u5230\u672a\u6765\u7684\u65b0\u6570\u636e\u4e0a\uff0c\u5e76\u5bf9\u65b0\u6570\u636e\u505a\u51fa\u5224\u65ad\u6216\u9884\u6d4b\u3002<\/p>\n \u4f8b\u5982\u67d0\u4e2a\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u4ece\u5168\u740370\u4ebf\u4eba\u7c7b\u7684\u6570\u636e\u96c6\u4e2d\u5b66\u5f97\u4e86\u67d0\u79cd\u5224\u65ad\u89c4\u5219\uff0c\u8fd9\u4e2a\u5224\u65ad\u89c4\u5219\u53ef\u4ee5\u901a\u8fc7\u8f93\u5165Nick\u8001\u5e08\u7684\u8eab\u9ad8180cm\u3001\u4f53\u91cd70Kg\uff0c\u5224\u65ad\u4ed6\u662f\u4e2a\u5e05\u54e5\u3002 \u901a\u8fc7\u4e0a\u56fe\uff0c\u6211\u4eec\u53ef\u4ee5\u4e60\u5f97\u4ee5\u4e0b\u673a\u5668\u5b66\u4e60\u7684\u57fa\u672c\u672f\u8bed\uff1a<\/p>\n \u6df1\u5ea6\u5b66\u4e60\uff08deep learning\uff09\uff1a<\/strong>\u901a\u8fc7\u7ec4\u5408\u4f4e\u5c42\u5c5e\u6027\u7279\u5f81\u5f62\u6210\u66f4\u52a0\u62bd\u8c61\u7684\u9ad8\u5c42\u5c5e\u6027\u7279\u5f81\u3002\u5982\u5b66\u82f1\u8bed\u7684\u65f6\u5019\uff0c\u201cNick handsome\u201d\u8fd912\u4e2a\u82f1\u6587\u5355\u8bcd\u662f\u4f4e\u5c42\u5c5e\u6027\u7279\u5f81\uff0c\u800c\u201cNick handsome\u201d\u7684\u8bed\u4e49\u5219\u662f\u62bd\u8c61\u7684\u9ad8\u5c42\u5c5e\u6027\u7279\u5f81\u3002<\/p>\n \u63d2\u56fe\uff1a\u6076\u641e\u56fe03 \u6df1\u5ea6\u5b66\u4e60\u7684\u6982\u5ff5\u6e90\u4e8e\u4eba\u5de5\u795e\u7ecf\u7f51\u7edc\u7684\u7814\u7a76\uff0c\u5b83\u5c5e\u4e8e\u673a\u5668\u5b66\u4e60\u4e2d\u7684\u67d0\u4e00\u4e2a\u65b9\u6cd5\uff0c\u5176\u4e2d\u6df1\u5ea6\u5b66\u4e60\u4e2d\u7684\u201c\u6df1\u5ea6\u201d\u662f\u6307\u795e\u7ecf\u7f51\u7edc\u7684\u5c42\u6570\u3002<\/p>\n \u673a\u5668\u5b66\u4e60\u4f9d\u636e\u6570\u636e\u96c6\u6570\u636e\u683c\u5f0f\u7684\u4e0d\u540c\uff0c\u53ef\u4ee5\u5212\u5206\u6210\u76d1\u7763\u5b66\u4e60\u548c\u65e0\u76d1\u7763\u5b66\u4e60\uff1b\u4f9d\u636e\u7b97\u6cd5\u6a21\u5f0f\u7684\u4e0d\u540c\u53ef\u4ee5\u5212\u5206\u4e3a\u76d1\u7763\u5b66\u4e60\u3001\u65e0\u76d1\u7763\u5b66\u4e60\u3001\u534a\u76d1\u7763\u5b66\u4e60\u548c\u5f3a\u5316\u5b66\u4e60\u3002<\/p>\n \u76d1\u7763\u5b66\u4e60\uff08supervised learning\uff09\uff1a<\/strong>\u901a\u8fc7\u8f93\u5165\u4e00\u7ec4\u5df2\u77e5\u6807\u8bb0\u7684\u6837\u672c\uff0c\u8f93\u51fa\u4e00\u4e2a\u6a21\u578b\uff08model\uff09\uff0c\u7136\u540e\u901a\u8fc7\u8fd9\u4e2a\u6a21\u578b\u9884\u6d4b\u672a\u6765\u65b0\u6570\u636e\u7684\u9884\u6d4b\u503c\u6216\u9884\u6d4b\u7c7b\u522b\u3002<\/p>\n \u76d1\u7763\u5b66\u4e60\u7684\u6d41\u7a0b\uff1a<\/p>\n \u7531\u4e8e\u6570\u636e\u6807\u8bb0\u7684\u7c7b\u578b\u4e0d\u540c\uff0c\u76d1\u7763\u5b66\u4e60\u95ee\u9898\u53c8\u5206\u4e3a\u56de\u5f52\u95ee\u9898\u548c\u5206\u7c7b\u95ee\u9898\uff1a<\/p>\n \u901a\u8fc7\u4e0a\u8868\u7ed9\u51fa\u7684\u6570\u636e\u96c6\uff08\u4e00\u7ec4\u5df2\u77e5\u6807\u8bb0\u7684\u6837\u672c\uff09\u4e3e\u4f8b\uff0c\u901a\u8fc7\u6570\u636e\u96c6\u4e2d\u7684\u6570\u636e\uff0c\u6211\u4eec\u53ef\u4ee5\u5047\u8bbe\u623f\u5b50\u9762\u79ef\u548c\u623f\u4ef7\u4e4b\u95f4\u5b58\u5728\u67d0\u79cd\u5173\u7cfb\uff08\u6a21\u578b\uff09\u4e3a \u5728\u8fd9\u4e2a\u56de\u5f52\u95ee\u9898\u4e2d\uff1a<\/p>\n \u540c\u6837\u901a\u8fc7\u4e0a\u8868\u7ed9\u51fa\u7684\u6570\u636e\u96c6\u4e3e\u4f8b\uff0c\u901a\u8fc7\u6570\u636e\u96c6\u7684\u6570\u636e\uff0c\u6211\u4eec\u53ef\u4ee5\u5047\u8bbe\u623f\u5b50\u9762\u79ef\u548c\u623f\u5b50\u6240\u5728\u533a\u57df\u5b58\u5728\u67d0\u79cd\u5173\u7cfb\u4e3a \u5176\u4e2d$P(A\u533a|\u623f\u5b50\u9762\u79ef<50m^2)$\u4e3a\u6761\u4ef6\u6982\u7387\uff0c\u5373\u5f53\u6ee1\u8db3$\u623f\u5b50\u9762\u79ef<50m^2$\u6761\u4ef6\u65f6\u623f\u5b50\u6240\u5728\u533a\u57df\u4e3a$A\u533a$\u7684\u6982\u7387\uff1b$P(B\u533a|\u623f\u5b50\u9762\u79ef>50m^2)$\u4e5f\u4e3a\u6761\u4ef6\u6982\u7387\uff0c\u5f53\u6ee1\u8db3$\u623f\u5b50\u9762\u79ef>50m^2$\u6761\u4ef6\u65f6\u623f\u5b50\u6240\u5728\u533a\u57df\u4e3a$B$\u533a\u7684\u6982\u7387\u3002<\/p>\n \u5047\u8bbe\u4f60\u5728\u5f97\u5230\u623f\u5b50\u9762\u79ef\u548c\u623f\u5b50\u6240\u5728\u533a\u57df\u7684\u5173\u7cfb\u540e\uff0c\u5f97\u5230\u4e86\u4e0a\u6d77\u4e8c\u73af\u5185\u4e24\u6240\u623f\u5b50$H_1$\u548c$H_2$\u9762\u79ef\u5206\u522b\u4e3a$30m^2$\u548c$1000m^2$\u7684\u6d88\u606f\uff0c\u8fd9\u4e2a\u65f6\u5019\u4f60\u5c31\u53ef\u4ee5\u901a\u8fc7\u623f\u5b50\u9762\u79ef\u548c\u623f\u5b50\u6240\u5728\u533a\u57df\u4e4b\u95f4\u7684\u5173\u7cfb\u5f97\u5230\u8fd9\u4e24\u6240\u623f\u5b50\u6240\u5728\u7684\u533a\u57df \u5728\u8fd9\u4e2a\u5206\u7c7b\u95ee\u9898\u4e2d\uff1a<\/p>\n \u65e0\u76d1\u7763\u5b66\u4e60\uff08unsupervised learning\uff09\uff1a<\/strong>\u901a\u8fc7\u8f93\u5165\u4e00\u7ec4\u672a\u77e5\u6807\u8bb0\u7684\u6837\u672c\uff0c\u53ef\u4ee5\u901a\u8fc7\u805a\u7c7b\uff08clustering\uff09\u7684\u65b9\u6cd5\uff0c\u5c06\u6570\u636e\u5206\u6210\u591a\u4e2a\u7c07\uff08cluster\uff09\u3002<\/p>\n \u65e0\u76d1\u7763\u5b66\u4e60\u7684\u6d41\u7a0b\uff1a<\/p>\n \u805a\u7c7b\uff08clustering\uff09\uff1a<\/strong>\u5c06\u6570\u636e\u96c6\u5206\u6210\u7531\u7c7b\u4f3c\u7684\u6570\u636e\u7ec4\u6210\u7684\u591a\u4e2a\u7c7b\u7684\u8fc7\u7a0b\u88ab\u79f0\u4e3a\u805a\u7c7b\u3002\u7531\u805a\u7c7b\u6240\u751f\u6210\u7684\u7c07\u662f\u4e00\u7ec4\u6570\u636e\u7684\u96c6\u5408\uff0c\u540c\u4e00\u4e2a\u7c07\u4e2d\u7684\u6570\u636e\u5f7c\u6b64\u76f8\u4f3c\uff0c\u4e0d\u540c\u7c07\u4e2d\u7684\u5bf9\u8c61\u76f8\u5f02\u3002<\/p>\n \u9700\u8981\u6ce8\u610f\u7684\u662f\uff0c\u7ec4\u6210\u591a\u4e2a\u7c07\u7684\u6982\u5ff5\u662f\u7531\u67d0\u79cd\u975e\u4eba\u4e3a\u5b9a\u4e49\u7684\u6f5c\u5728\u6982\u5ff5\u81ea\u52a8\u751f\u6210\u7684\uff0c\u800c\u6211\u4eec\u5728\u76d1\u7763\u5b66\u4e60\u4e2d\u7684\u5206\u7c7b\u90fd\u6709\u660e\u786e\u7684\u5b9a\u4e49\uff0c\u5373\u6309\u7167\u623f\u5b50\u9762\u79ef\u548c\u623f\u4ef7\u7684\u5173\u7cfb\u3001\u623f\u5b50\u9762\u79ef\u548c\u623f\u5b50\u6240\u5728\u533a\u57df\u7684\u5173\u7cfb\u3002<\/p>\n \u901a\u8fc7\u4e0a\u8868\u7ed9\u51fa\u7684\u6570\u636e\u96c6\u4e3e\u4f8b\uff0c\u901a\u8fc7\u805a\u7c7b\u53ef\u80fd\u628a\u6570\u636e\u96c6\u4e2d\u76845\u4e2a\u6570\u636e\u5206\u6210\u4e24\u4e2a\u7c07\uff0c\u5373\u623f\u5b50\u9762\u79ef\u5927\u4e8e50\u7684\u5206\u4e3a\u4e00\u4e2a\u7c07\uff0c\u623f\u5b50\u9762\u79ef\u5c0f\u4e8e50\u7684\u4e3a\u53e6\u4e00\u4e2a\u7c07\uff0c\u800c\u8fd9\u79cd\u901a\u8fc7\u9762\u79ef\u5206\u7c7b\u5e76\u4e0d\u662f\u6211\u4eec\u4e8b\u5148\u8bbe\u5b9a\u7684\uff0c\u800c\u662f\u7531\u805a\u7c7b\u7b97\u6cd5\u81ea\u5df1\u5b9e\u73b0\u7684\u3002\u4e5f\u5c31\u662f\u8bf4\u901a\u8fc7\u805a\u7c7b\u628a\u4e0a\u8868\u7684\u6570\u636e\u96c6\u5206\u4e3a\u4e24\u7c07\uff0c\u4f46\u662f\u6211\u4eec\u9700\u8981\u5728\u628a\u6570\u636e\u96c6\u5206\u6210\u4e24\u4e2a\u7c07\u540e\u518d\u53bb\u51b3\u5b9a\u7528\u4ec0\u4e48\u6982\u5ff5\u5b9a\u4e49\u8fd9\u4e24\u4e2a\u7c07\u7684\u5212\u5206\u3002<\/p>\n \u534a\u76d1\u7763\u5b66\u4e60\uff08semi-supervised learning\uff09\uff1a<\/strong>\u76d1\u7763\u5b66\u4e60\u4e0e\u65e0\u76d1\u7763\u5b66\u4e60\u76f8\u7ed3\u5408\u7684\u4e00\u79cd\u5b66\u4e60\u65b9\u6cd5\u3002<\/p>\n \u76d1\u7763\u5b66\u4e60\u548c\u65e0\u76d1\u7763\u5b66\u4e60\u7684\u533a\u522b\u4e3b\u8981\u5728\u4e8e\u6570\u636e\u96c6\u4e2d\u7684\u6570\u636e\u662f\u5426\u5177\u6709\u6807\u8bb0\uff0c\u5f53\u67d0\u4e2a\u95ee\u9898\u4e2d\u7684\u6570\u636e\u96c6\u4e2d\u7684\u6570\u636e\u5177\u6709\u6807\u8bb0\u65f6\uff0c\u6211\u4eec\u79f0\u4e4b\u4e3a\u76d1\u7763\u5b66\u4e60\uff1b\u53cd\u4e4b\uff0c\u6211\u4eec\u79f0\u4e4b\u4e3a\u65e0\u76d1\u7763\u5b66\u4e60\u3002<\/p>\n \u76ee\u524d\uff0c\u7531\u4e8e\u65e0\u76d1\u7763\u5b66\u4e60\u7b97\u6cd5\u7684\u4e0d\u7a33\u5b9a\u6027\uff0c\u5b83\u4e00\u822c\u4f5c\u4e3a\u4e2d\u95f4\u7b97\u6cd5\uff0c\u5de5\u4e1a\u4e0a\u4f7f\u7528\u8f83\u591a\u7684\u662f\u76d1\u7763\u5b66\u4e60\u7b97\u6cd5\u3002\u4f46\u662f\u76ee\u524d\u5de5\u4e1a\u4e0a\u7d2f\u79ef\u8f83\u591a\u7684\u662f\u4e0d\u5e26\u6807\u8bb0\u7684\u6570\u636e\uff0c\u56e0\u6b64\u65e0\u76d1\u7763\u5b66\u4e60\u5728\u672a\u6765\u5c06\u662f\u4e00\u4e2a\u6280\u672f\u7a81\u7834\u7684\u91cd\u70b9\u76ee\u6807\u3002\u4e3a\u4e86\u5904\u7406\u8fd9\u6279\u4e0d\u5e26\u6807\u8bb0\u7684\u6570\u636e\uff0c\u4e5f\u6709\u79d1\u5b66\u5bb6\u63d0\u51fa\u4e86\u534a\u76d1\u7763\u5b66\u4e60\uff0c\u5b83\u53ef\u4ee5\u7b80\u5355\u7684\u7406\u89e3\u4e3a\u76d1\u7763\u5b66\u4e60\u548c\u65e0\u76d1\u7763\u5b66\u4e60\u7684\u7efc\u5408\u8fd0\u7528\uff0c\u5b83\u5373\u4f7f\u7528\u5e26\u6709\u6807\u8bb0\u7684\u6570\u636e\u4e5f\u4f7f\u7528\u4e0d\u5e26\u6709\u6807\u8bb0\u7684\u6570\u636e\u3002<\/p>\n \u5f3a\u5316\u5b66\u4e60\uff08reinforcement learning\uff09\uff1a<\/strong>\u5b66\u4e60\u7cfb\u7edf\u4ece\u73af\u5883\u5230\u884c\u4e3a\u6620\u5c04\u7684\u5b66\u4e60\uff0c\u76ee\u6807\u662f\u4f7f\u667a\u80fd\u4f53\u83b7\u5f97\u6700\u5927\u7684\u5956\u52b1\uff0c\u5373\u5f3a\u5316\u4fe1\u53f7\u6700\u5927\u3002<\/p>\n \u5f3a\u5316\u5b66\u4e60\u7c7b\u4f3c\u4e8e\u65e0\u76d1\u7763\u5b66\u4e60\uff0c\u4f46\u662f\u53c8\u4e0d\u540c\u4e8e\u65e0\u76d1\u7763\u5b66\u4e60\uff0c\u5f3a\u5316\u5b66\u4e60\u9488\u5bf9\u7684\u662f\u65e0\u6807\u8bb0\u7684\u6570\u636e\uff0c\u4f46\u662f\u5728\u5f3a\u5316\u5b66\u4e60\u7b97\u6cd5\u5bf9\u65e0\u6807\u8bb0\u7684\u6570\u636e\u8fdb\u884c\u5904\u7406\u7684\u540c\u65f6\u4f1a\u7ed9\u51fa\u4e00\u4e2a\u7279\u5b9a\u7684\u76ee\u6807\uff0c\u5982\u679c\u7b97\u6cd5\u7684\u5904\u7406\u7ed3\u679c\u504f\u79bb\u76ee\u6807\uff0c\u5219\u4f1a\u7ed9\u51fa\u60e9\u7f5a\uff1b\u53cd\u4e4b\uff0c\u7ed9\u51fa\u5956\u52b1\uff0c\u800c\u8be5\u7b97\u6cd5\u7684\u76ee\u6807\u5c31\u662f\u4e3a\u4e86\u83b7\u5f97\u6700\u5927\u5956\u52b1\u3002\u5f3a\u5316\u5b66\u4e60\u4e00\u822c\u7528\u4e8e\u65e0\u4eba\u9a7e\u9a76\u3002<\/p>\n \u6211\u4eec\u53ef\u4ee5\u901a\u8fc7Python\u3001R\u3001C\u3001Java\u3001MATLAB\u3001Octave\u7b49\u5de5\u5177\u5b9e\u73b0\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u3002\u672c\u6587\u63a5\u4e0b\u6765\u5c06\u901a\u8fc7Python\u5b9e\u73b0\uff0c\u56e0\u6b64\u5728\u8fd9\u91cc\u7b80\u5355\u8bf4\u8bf4Python\u7684\u4f18\u52bf\uff1a\u8bed\u6cd5\u6e05\u6670\u7b80\u5355\u3001\u6613\u4e8e\u64cd\u4f5c\u6587\u672c\u6587\u4ef6\u3001\u4f18\u79c0\u7684\u5f00\u6e90\u793e\u533a\u63d0\u4f9b\u4e86\u5927\u91cf\u7684\u5f00\u6e90\u8f6f\u4ef6\u5e93\uff0c\u5982scikit-learn\u3001tensorflow\u7b49\u3002<\/p>\n \u672c\u6587\u57fa\u4e8ePython3.6+\u7248\u672c\uff0c\u4e0d\u5efa\u8bae\u4f7f\u7528Python2.7\u7248\u672c\u3002\u53ef\u4ee5\u4ecehttps:\/\/www.python.org\/downloads\/<\/a>\u4e0b\u8f7dPython\u3002<\/p>\n\n
1.3 \u673a\u5668\u5b66\u4e60<\/h1>\n
\n![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/06-01-\u4e1c\u65b9\u4e0d\u8d25.jpg\")
<\/div><\/p>\n1.3.1 \u673a\u5668\u5b66\u4e60\u57fa\u672c\u672f\u8bed<\/h2>\n
# \u673a\u5668\u5b66\u4e60\u57fa\u672c\u672f\u8bed\u56fe\u4f8b\nimport matplotlib.pyplot as plt\nfrom matplotlib.font_manager import FontProperties\n# \u663e\u793aplt.show()\u7684\u56fe\u7247\uff0c\u5982\u82e5\u4e0d\u4f7f\u7528jupyter\uff0c\u8bf7\u6ce8\u91ca\n%matplotlib inline\n# \u4e2d\u6587\u5b57\u4f53\u8bbe\u7f6e\uff0c\u627e\u5230\u4e0e\u4e4b\u5bf9\u5e94\u7684\u4e2d\u6587\u5b57\u4f53\u8def\u5f84\nfont = FontProperties(fname='\/Library\/Fonts\/Heiti.ttc')\n\n# \u6837\u672c\nplt.text(1, 70, s='{\u8eab\u9ad8(180)\uff0c\u4f53\u91cd(70)\uff0c\u5e74\u9f84(19)\uff0c\u4e94\u5b98(\u7cbe\u81f4)',\n fontproperties=font, fontsize=13, color='g')\nplt.annotate(text='\u6837\u672c', xytext=(90, 70), xy=(78, 71), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='\u7279\u5f81', xytext=(30, 90), xy=(10, 73), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 90), xy=(30, 73), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 90), xy=(50, 73), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 90), xy=(70, 73), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\n\nplt.hlines(67, 0, 100, linestyle='--', color='gray')\n\n# \u7279\u5f81\nplt.text(1, 40, s='{\u8eab\u9ad8(180)\uff0c\u4f53\u91cd(70)\uff0c\u5e74\u9f84(19)\uff0c\u4e94\u5b98(\u7cbe\u81f4)\uff0c\u5e05}',\n fontproperties=font, fontsize=10, color='g')\nplt.annotate(text='\u6837\u4f8b(\u5b9e\u4f8b)', xytext=(90, 40), xy=(70, 41), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='\u7279\u5f81', xytext=(30, 60), xy=(8, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 60), xy=(23, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 60), xy=(38, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='', xytext=(30, 60), xy=(53, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='\u6807\u8bb0', xytext=(67, 60), xy=(67, 43), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\n\nplt.hlines(37, 0, 100, linestyle='--', color='gray')\n\n# \u7279\u5f81\u7a7a\u95f4\nplt.text(1, 30, s='$\\{(x_1^{(1)},x_1^{(2)},\\cdots,x_1^{(n)}),(x_2^{(1)},x_2^{(2)},\\cdots,x_2^{(n)}),\\cdots,(x_m^{(1)},x_m^{(2)},\\cdots,x_m^{(n)})\\}$',\n fontproperties=font, fontsize=10, color='g')\nplt.annotate(text='\u7279\u5f81\u7a7a\u95f4($m$\u4e2a\u6837\u672c)', xytext=(90, 15), xy=(92, 30), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\nplt.annotate(text='\u7279\u5f81\u5411\u91cf($n$\u7ef4\u7279\u5f81)', xytext=(15, 15), xy=(15, 28), ha='center', arrowprops=dict(\n arrowstyle="<-", color='b'), fontproperties=font, fontsize=15, color='r')\n\nplt.xlim(0, 100)\nplt.ylim(10, 100)\nplt.title('\u673a\u5668\u5b66\u4e60\u57fa\u672c\u672f\u8bed\u56fe\u4f8b', fontproperties=font, fontsize=20)\nplt.show()<\/code><\/pre>\n![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/01-01-\u673a\u5668\u5b66\u4e60_11_0-1.png\")
<\/div><\/p>\n\n
\n![\"file\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/image-1675220161073.png\")
<\/div>
\n\u5176\u4e2d$x_m^{(n)}$\u8868\u793a\u7b2cm\u4e2a\u4eba\u7684\u7b2cn\u4e2a\u7279\u5f81\u3002\u5982\u8eab\u9ad8\u3001\u4f53\u91cd\u3001\u5e74\u9f84\u5f20\u6210\u7684\u4e09\u7ef4\u7a7a\u95f4\u4e2d\u7684\u67d0\u4e00\u4e2a\u5177\u4f53\u7684\u70b9$(180,70,19)$\u3002<\/li>\n<\/ul>\n1.4 \u6df1\u5ea6\u5b66\u4e60<\/h1>\n
\n![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/\u6076\u641e\u56fe03.png\")
<\/div><\/p>\n![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/A-Z\u5b57\u6bcd\u8868.jpg\")
<\/div><\/p>\n1.5 \u673a\u5668\u5b66\u4e60\u5206\u7c7b<\/h1>\n
![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/\u76d1\u7763\u5b66\u4e60\u548c\u65e0\u76d1\u7763\u5b66\u4e601.png\")
<\/div><\/p>\n1.5.1 \u76d1\u7763\u5b66\u4e60<\/h2>\n
\n
\n
1.5.1.2 \u56de\u5f52\u95ee\u9898<\/h3>\n
\n\n
\n \n\u5e8f\u53f7<\/th>\n \u623f\u5b50\u9762\u79ef(m$^2$)<\/th>\n \u623f\u4ef7(\u5143)<\/th>\n<\/tr>\n<\/thead>\n \n 1<\/td>\n 10<\/td>\n 100<\/td>\n<\/tr>\n \n 2<\/td>\n 20<\/td>\n 200<\/td>\n<\/tr>\n \n 3<\/td>\n 40<\/td>\n 400<\/td>\n<\/tr>\n \n 4<\/td>\n 100<\/td>\n 1000<\/td>\n<\/tr>\n \n 5<\/td>\n 200<\/td>\n 2000<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n # \u56de\u5f52\u95ee\u9898\u56fe\u4f8b\nimport numpy as np\nimport matplotlib.pyplot as plt\nfrom matplotlib.font_manager import FontProperties\n%matplotlib inline\nfont = FontProperties(fname='\/Library\/Fonts\/Heiti.ttc')\n\nx = np.linspace(0, 200, 10)\ny = 10*x\n\nplt.plot(x, y)\nplt.xlabel('\u623f\u5b50\u9762\u79ef', fontproperties=font, fontsize=15)\nplt.ylabel('\u623f\u4ef7', fontproperties=font, fontsize=15)\nplt.title('\u56de\u5f52\u95ee\u9898\u56fe\u4f8b', fontproperties=font, fontsize=20)\nplt.show()<\/code><\/pre>\n![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/01-01-\u673a\u5668\u5b66\u4e60_23_0-1.png\")
<\/div><\/p>\n
\n$$
\n\u623f\u4ef7=10\u623f\u5b50\u9762\u79ef
\n$$
\n \u5047\u8bbe\u4f60\u5728\u5f97\u5230\u623f\u5b50\u9762\u79ef\u548c\u623f\u4ef7\u4e4b\u95f4\u7684\u5173\u7cfb\u540e\uff0c\u5f97\u5230\u4e86\u4e0a\u6d77\u4e8c\u73af\u67d0\u4e00\u6240\u623f\u5b50$H_1$\u7684\u9762\u79ef\u4e3a$1000m^2$\u7684\u6d88\u606f\uff08\u672a\u6765\u65b0\u6570\u636e\uff09\uff0c\u8fd9\u4e2a\u65f6\u5019\u4f60\u5c31\u53ef\u4ee5\u901a\u8fc7$H_1$\u7684\u623f\u5b50\u9762\u79ef\u548c\u623f\u4ef7\u4e4b\u95f4\u7684\u5173\u7cfb\u5f97\u5230\u8fd9\u6240\u623f\u5b50\u7684\u4ef7\u683c\u4e3a
\n$$
\nH_1\u7684\u623f\u4ef7=10<\/em>H_1\u7684\u623f\u5b50\u9762\u79ef=10*1000=10000(\u5143)\\quad\\text{\u9884\u6d4b\u503c}
\n$$<\/p>\n\n
![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/06-02-\u623f\u4ef7.jpg\")
<\/div><\/p>\n1.5.1.3 \u5206\u7c7b\u95ee\u9898<\/h3>\n
\n\n
\n \n\u5e8f\u53f7<\/th>\n \u623f\u5b50\u9762\u79ef(m$^2$)<\/th>\n \u623f\u5b50\u6240\u5728\u533a\u57df<\/th>\n<\/tr>\n<\/thead>\n \n 1<\/td>\n 10<\/td>\n A\u533a<\/td>\n<\/tr>\n \n 2<\/td>\n 20<\/td>\n A\u533a<\/td>\n<\/tr>\n \n 3<\/td>\n 40<\/td>\n A\u533a<\/td>\n<\/tr>\n \n 4<\/td>\n 100<\/td>\n B\u533a<\/td>\n<\/tr>\n \n 5<\/td>\n 200<\/td>\n B\u533a<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n # \u5206\u7c7b\u95ee\u9898\u56fe\u4f8b\nimport numpy as np\nimport matplotlib.pyplot as plt\nfrom matplotlib.font_manager import FontProperties\n%matplotlib inline\nfont = FontProperties(fname='\/Library\/Fonts\/Heiti.ttc')\n\na_area = [10, 20, 40]\na_y = [0, 0, 0]\n\nb_area = [100, 200]\nb_y = [0, 0]\n\nplt.scatter(a_area, a_y, label='A', color='r')\nplt.scatter(b_area, b_y, label='B', color='g')\nplt.vlines(50, ymin=-0.015, ymax=0.015, color='gray', linestyles='--')\nplt.xlabel('\u623f\u5b50\u9762\u79ef', fontproperties=font, fontsize=15)\nplt.title('\u5206\u7c7b\u95ee\u9898\u56fe\u4f8b', fontproperties=font, fontsize=20)\nplt.legend()\nplt.show()<\/code><\/pre>\n![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/01-01-\u673a\u5668\u5b66\u4e60_27_0-1.png\")
<\/div><\/p>\n
\n![\"file\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/image-1675219877817.png\")
<\/div><\/p>\n
\n![\"file\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/image-1675219939481.png\")
<\/div><\/p>\n\n
![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/06-03-\u5893\u5730.jpg\")
<\/div><\/p>\n1.5.2 \u65e0\u76d1\u7763\u5b66\u4e60<\/h2>\n
\n
1.5.2.1 \u805a\u7c7b<\/h3>\n
\n\n
\n \n\u5e8f\u53f7<\/th>\n \u623f\u5b50\u9762\u79ef(m$^2$)<\/th>\n<\/tr>\n<\/thead>\n \n 1<\/td>\n 10<\/td>\n<\/tr>\n \n 2<\/td>\n 20<\/td>\n<\/tr>\n \n 3<\/td>\n 40<\/td>\n<\/tr>\n \n 4<\/td>\n 100<\/td>\n<\/tr>\n \n 5<\/td>\n 200<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n ![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2022\/02\/\u6076\u641e\u56fe04.png\")
<\/div><\/p>\n1.5.3 \u534a\u76d1\u7763\u5b66\u4e60<\/h2>\n
<\/div><\/p>\n1.5.4 \u5f3a\u5316\u5b66\u4e60<\/h2>\n
1.6 Python\u8bed\u8a00\u7684\u4f18\u52bf<\/h1>\n
![\"\"](\"https:\/\/egonlin.com\/wp-content\/uploads\/2021\/08\/\u6076\u641e\u56fe01-18.png\")
<\/div><\/p>\n","protected":false},"excerpt":{"rendered":"