{"id":910,"date":"2024-06-24T10:02:23","date_gmt":"2024-06-24T02:02:23","guid":{"rendered":"http:\/\/www.inhhh.com\/blog\/?p=910"},"modified":"2024-06-25T17:39:04","modified_gmt":"2024-06-25T09:39:04","slug":"ai","status":"publish","type":"post","link":"http:\/\/www.inhhh.com\/blog\/?p=910","title":{"rendered":"AI"},"content":{"rendered":"\n

AI \u5165\u95e8\uff0c\u4ece\u96f6\u642d\u5efa\u5b8c\u6574 AI \u5f00\u53d1\u73af\u5883\uff0c\u5e76\u5199\u51fa\u7b2c\u4e00\u4e2a AI \u5e94\u7528<\/p>\n\n\n\n

\u9ebb\u8fa3\u6392\u9aa8\u9762<\/p>\n\n\n\n

\u5df2\u4e8e 2024-05-09 20:08:44 \u4fee\u6539<\/p>\n\n\n\n

\u9605\u8bfb\u91cf2k
\u6536\u85cf 20<\/p>\n\n\n\n

\u70b9\u8d5e\u6570 14
\u6587\u7ae0\u6807\u7b7e\uff1a \u4eba\u5de5\u667a\u80fd neo4j web\u5b89\u5168 redis \u67b6\u6784
\u7248\u6743
\u4eba\u5de5\u667a\u80fd\uff08AI\uff09\u662f\u5f53\u4eca\u79d1\u6280\u9886\u57df\u4e2d\u5907\u53d7\u77a9\u76ee\u7684\u9886\u57df\u4e4b\u4e00\uff0c\u5b83\u6b63\u5728\u6539\u53d8\u6211\u4eec\u7684\u751f\u6d3b\u65b9\u5f0f\u3001\u5de5\u4f5c\u65b9\u5f0f\u4ee5\u53ca\u4e0e\u6280\u672f\u4e92\u52a8\u7684\u65b9\u5f0f\u3002\u672c\u6587\u5c06\u5e26\u60a8\u4ece\u96f6\u5f00\u59cb\uff0c\u4e00\u6b65\u6b65\u642d\u5efa\u4e00\u4e2a\u5b8c\u6574\u7684AI\u5f00\u53d1\u73af\u5883\uff0c\u5e76\u521b\u5efa\u4e00\u4e2a\u7b80\u5355\u7684AI\u5e94\u7528\u7a0b\u5e8f\uff0c\u4ee5\u4fbf\u60a8\u80fd\u591f\u4eb2\u8eab\u4f53\u9a8cAI\u7684\u9b45\u529b\u3002<\/p>\n\n\n\n

\u7b2c\u4e00\u90e8\u5206\uff1a\u51c6\u5907\u5de5\u4f5c
\u5728\u5f00\u59cb\u4e4b\u524d\uff0c\u6211\u4eec\u9700\u8981\u505a\u4e00\u4e9b\u51c6\u5907\u5de5\u4f5c\u3002\u8fd9\u5305\u62ec\u5b89\u88c5\u5fc5\u8981\u7684\u5de5\u5177\u548c\u8bbe\u7f6e\u5f00\u53d1\u73af\u5883\u3002<\/p>\n\n\n\n

1.1. \u5b89\u88c5Python
Python\u662fAI\u9886\u57df\u7684\u9996\u9009\u7f16\u7a0b\u8bed\u8a00\u4e4b\u4e00\uff0c\u6211\u4eec\u5c06\u4f7f\u7528Python\u6765\u7f16\u5199\u6211\u4eec\u7684AI\u5e94\u7528\u3002\u60a8\u53ef\u4ee5\u4ecePython\u5b98\u65b9\u7f51\u7ad9\uff08https:\/\/www.python.org\/\uff09\u4e0b\u8f7d\u5e76\u5b89\u88c5\u6700\u65b0\u7248\u672c\u7684Python\u3002<\/p>\n\n\n\n

\u5b89\u88c5\u5b8c\u6210\u540e\uff0c\u60a8\u53ef\u4ee5\u5728\u547d\u4ee4\u884c\u4e2d\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\u6765\u9a8c\u8bc1Python\u662f\u5426\u6b63\u786e\u5b89\u88c5\uff1a<\/p>\n\n\n\n

python –version
1
1.2. \u5b89\u88c5Anaconda
Anaconda\u662f\u4e00\u4e2a\u5f3a\u5927\u7684Python\u6570\u636e\u79d1\u5b66\u548c\u673a\u5668\u5b66\u4e60\u5e73\u53f0\uff0c\u5b83\u5305\u542b\u4e86\u8bb8\u591a\u5e38\u7528\u7684\u6570\u636e\u79d1\u5b66\u5de5\u5177\u548c\u5e93\u3002\u60a8\u53ef\u4ee5\u4eceAnaconda\u5b98\u65b9\u7f51\u7ad9\uff08https:\/\/www.anaconda.com\/\uff09\u4e0b\u8f7d\u5e76\u5b89\u88c5Anaconda\u3002<\/p>\n\n\n\n

1.3. \u521b\u5efa\u865a\u62df\u73af\u5883
\u4e3a\u4e86\u9694\u79bb\u4e0d\u540c\u9879\u76ee\u7684\u4f9d\u8d56\u5173\u7cfb\uff0c\u6211\u4eec\u5efa\u8bae\u5728Anaconda\u4e2d\u521b\u5efa\u4e00\u4e2a\u865a\u62df\u73af\u5883\u3002\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u521b\u5efa\u4e00\u4e2a\u540d\u4e3aai_env\u7684\u865a\u62df\u73af\u5883\uff1a<\/p>\n\n\n\n

conda create –name ai_env python=3.8
1
\u7136\u540e\u6fc0\u6d3b\u865a\u62df\u73af\u5883\uff1a<\/p>\n\n\n\n

conda activate ai_env
1
1.4. \u5b89\u88c5Jupyter Notebook
Jupyter Notebook\u662f\u4e00\u4e2a\u4ea4\u4e92\u5f0f\u5f00\u53d1\u73af\u5883\uff0c\u975e\u5e38\u9002\u5408\u63a2\u7d22\u6570\u636e\u548c\u7f16\u5199\u4ee3\u7801\u3002\u5728\u6fc0\u6d3b\u865a\u62df\u73af\u5883\u540e\uff0c\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\u6765\u5b89\u88c5Jupyter Notebook\uff1a<\/p>\n\n\n\n

conda install jupyter
1
\u73b0\u5728\uff0c\u6211\u4eec\u5df2\u7ecf\u5b8c\u6210\u4e86\u51c6\u5907\u5de5\u4f5c\uff0c\u53ef\u4ee5\u5f00\u59cb\u6784\u5efa\u6211\u4eec\u7684\u7b2c\u4e00\u4e2aAI\u5e94\u7528\u4e86\u3002<\/p>\n\n\n\n

\u7b2c\u4e8c\u90e8\u5206\uff1a\u521b\u5efa\u4e00\u4e2a\u7b80\u5355\u7684AI\u5e94\u7528
\u6211\u4eec\u5c06\u521b\u5efa\u4e00\u4e2a\u7b80\u5355\u7684AI\u5e94\u7528\uff0c\u8be5\u5e94\u7528\u53ef\u4ee5\u8bc6\u522b\u624b\u5199\u6570\u5b57\u3002\u6211\u4eec\u5c06\u4f7f\u7528Python\u548c\u4e00\u4e2a\u6d41\u884c\u7684\u6df1\u5ea6\u5b66\u4e60\u6846\u67b6\u2014\u2014TensorFlow\u6765\u6784\u5efa\u8fd9\u4e2a\u5e94\u7528\u3002<\/p>\n\n\n\n

2.1. \u5b89\u88c5TensorFlow
\u5728\u6fc0\u6d3b\u865a\u62df\u73af\u5883\u540e\uff0c\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\u6765\u5b89\u88c5TensorFlow\uff1a<\/p>\n\n\n\n

conda install tensorflow
1
2.2. \u5bfc\u5165\u5fc5\u8981\u7684\u5e93
\u9996\u5148\uff0c\u6253\u5f00Jupyter Notebook\u5e76\u521b\u5efa\u4e00\u4e2a\u65b0\u7684Notebook\u3002\u5728Notebook\u4e2d\uff0c\u5bfc\u5165\u4ee5\u4e0b\u5fc5\u8981\u7684\u5e93\uff1a<\/p>\n\n\n\n

import tensorflow as tf from tensorflow import keras import matplotlib.pyplot as plt
1
2.3. \u52a0\u8f7d\u6570\u636e\u96c6
\u6211\u4eec\u5c06\u4f7f\u7528MNIST\u624b\u5199\u6570\u5b57\u6570\u636e\u96c6\uff0c\u8be5\u6570\u636e\u96c6\u5305\u542b\u4e86\u5927\u91cf\u7684\u624b\u5199\u6570\u5b57\u56fe\u50cf\u4ee5\u53ca\u5b83\u4eec\u5bf9\u5e94\u7684\u6807\u7b7e\u3002\u8fd0\u884c\u4ee5\u4e0b\u4ee3\u7801\u6765\u52a0\u8f7d\u6570\u636e\u96c6\uff1a<\/p>\n\n\n\n

mnist = keras.datasets.mnist (train_images, train_labels), (test_images, test_labels) = mnist.load_data()
1
2.4. \u6570\u636e\u9884\u5904\u7406
\u5728\u6784\u5efa\u6a21\u578b\u4e4b\u524d\uff0c\u6211\u4eec\u9700\u8981\u5bf9\u6570\u636e\u8fdb\u884c\u4e00\u4e9b\u9884\u5904\u7406\u3002\u8fd0\u884c\u4ee5\u4e0b\u4ee3\u7801\u5c06\u56fe\u50cf\u6570\u636e\u5f52\u4e00\u5316\u52300\u52301\u4e4b\u95f4\uff1a<\/p>\n\n\n\n

train_images = train_images \/ 255.0 test_images = test_images \/ 255.0
1
2.5. \u6784\u5efa\u795e\u7ecf\u7f51\u7edc\u6a21\u578b
\u6211\u4eec\u5c06\u521b\u5efa\u4e00\u4e2a\u7b80\u5355\u7684\u795e\u7ecf\u7f51\u7edc\u6a21\u578b\uff0c\u5305\u542b\u4e00\u4e2a\u8f93\u5165\u5c42\u3001\u4e00\u4e2a\u9690\u85cf\u5c42\u548c\u4e00\u4e2a\u8f93\u51fa\u5c42\u3002\u8fd0\u884c\u4ee5\u4e0b\u4ee3\u7801\u6765\u6784\u5efa\u6a21\u578b\uff1a<\/p>\n\n\n\n

model = keras.Sequential([ keras.layers.Flatten(input_shape=(28, 28)), keras.layers.Dense(128, activation=’relu’), keras.layers.Dense(10, activation=’softmax’) ])
1
2.6. \u7f16\u8bd1\u6a21\u578b
\u7f16\u8bd1\u6a21\u578b\u65f6\uff0c\u6211\u4eec\u9700\u8981\u6307\u5b9a\u635f\u5931\u51fd\u6570\u3001\u4f18\u5316\u5668\u548c\u8bc4\u4f30\u6307\u6807\u3002\u8fd0\u884c\u4ee5\u4e0b\u4ee3\u7801\u6765\u7f16\u8bd1\u6a21\u578b\uff1a<\/p>\n\n\n\n

model.compile(optimizer=’adam’, loss=’sparse_categorical_crossentropy’, metrics=[‘accuracy’])
1
2.7. \u8bad\u7ec3\u6a21\u578b
\u73b0\u5728\uff0c\u6211\u4eec\u53ef\u4ee5\u5f00\u59cb\u8bad\u7ec3\u6a21\u578b\u4e86\u3002\u8fd0\u884c\u4ee5\u4e0b\u4ee3\u7801\u6765\u8bad\u7ec3\u6a21\u578b\uff1a<\/p>\n\n\n\n

model.fit(train_images, train_labels, epochs=5)
1
2.8. \u8bc4\u4f30\u6a21\u578b
\u8bad\u7ec3\u5b8c\u6210\u540e\uff0c\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528\u6d4b\u8bd5\u96c6\u6765\u8bc4\u4f30\u6a21\u578b\u7684\u6027\u80fd\u3002\u8fd0\u884c\u4ee5\u4e0b\u4ee3\u7801\u6765\u8bc4\u4f30\u6a21\u578b\uff1a<\/p>\n\n\n\n

test_loss, test_acc = model.evaluate(test_images, test_labels, verbose=2) print(‘\\nTest accuracy:’, test_acc)
1
2.9. \u4f7f\u7528\u6a21\u578b\u8fdb\u884c\u9884\u6d4b
\u6700\u540e\uff0c\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528\u8bad\u7ec3\u597d\u7684\u6a21\u578b\u6765\u8fdb\u884c\u9884\u6d4b\u3002\u8fd0\u884c\u4ee5\u4e0b\u4ee3\u7801\u6765\u8fdb\u884c\u9884\u6d4b\uff1a<\/p>\n\n\n\n

predictions = model.predict(test_images)
1
\u60a8\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u4ee3\u7801\u6765\u67e5\u770b\u9884\u6d4b\u7ed3\u679c\uff1a<\/p>\n\n\n\n

print(predictions[0])
1
\u7b2c\u4e09\u90e8\u5206\uff1a\u5b9e\u9645\u5e94\u7528\u6848\u4f8b
\u73b0\u5728\uff0c\u60a8\u5df2\u7ecf\u6210\u529f\u5730\u521b\u5efa\u4e86\u4e00\u4e2a\u7b80\u5355\u7684AI\u5e94\u7528\uff0c\u53ef\u4ee5\u8bc6\u522b\u624b\u5199\u6570\u5b57\u3002\u8fd9\u4e2a\u5e94\u7528\u867d\u7136\u7b80\u5355\uff0c\u4f46\u5c55\u793a\u4e86AI\u7684\u5f3a\u5927\u80fd\u529b\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u5b9e\u9645\u5e94\u7528\u6848\u4f8b\uff1a<\/p>\n\n\n\n

3.1. \u624b\u5199\u6570\u5b57\u8bc6\u522b
\u60a8\u53ef\u4ee5\u5c06\u8fd9\u4e2a\u5e94\u7528\u6269\u5c55\u5230\u66f4\u5927\u89c4\u6a21\u7684\u624b\u5199\u6570\u5b57\u8bc6\u522b\u95ee\u9898\uff0c\u7528\u4e8e\u81ea\u52a8\u8bc6\u522b\u90ae\u653f\u7f16\u7801\u3001\u94f6\u884c\u652f\u7968\u53f7\u7801\u7b49\u3002<\/p>\n\n\n\n

3.2. \u56fe\u50cf\u5206\u7c7b
\u4f7f\u7528\u7c7b\u4f3c\u7684\u65b9\u6cd5\uff0c\u60a8\u53ef\u4ee5\u6784\u5efa\u56fe\u50cf\u5206\u7c7b\u6a21\u578b\uff0c\u7528\u4e8e\u8bc6\u522b\u4e0d\u540c\u79cd\u7c7b\u7684\u56fe\u50cf\uff0c\u5982\u52a8\u7269\u3001\u690d\u7269\u3001\u4ea4\u901a\u6807\u5fd7\u7b49\u3002<\/p>\n\n\n\n

3.3. \u81ea\u7136\u8bed\u8a00\u5904\u7406
\u9664\u4e86\u56fe\u50cf\u8bc6\u522b\uff0cTensorFlow<\/p>\n\n\n\n

\u8fd8\u63d0\u4f9b\u4e86\u81ea\u7136\u8bed\u8a00\u5904\u7406\uff08NLP\uff09\u7684\u529f\u80fd\uff0c\u60a8\u53ef\u4ee5\u6784\u5efa\u6587\u672c\u5206\u7c7b\u3001\u60c5\u611f\u5206\u6790\u3001\u673a\u5668\u7ffb\u8bd1\u7b49\u5e94\u7528\u3002<\/p>\n\n\n\n

\u603b\u7ed3
\u5728\u672c\u6587\u4e2d\uff0c\u6211\u4eec\u4ece\u96f6\u5f00\u59cb\uff0c\u4e00\u6b65\u6b65\u642d\u5efa\u4e86\u4e00\u4e2a\u5b8c\u6574\u7684AI\u5f00\u53d1\u73af\u5883\uff0c\u5e76\u521b\u5efa\u4e86\u4e00\u4e2a\u7b80\u5355\u7684AI\u5e94\u7528\u3002\u60a8\u5b66\u4e60\u4e86\u5982\u4f55\u5b89\u88c5Python\u3001Anaconda\u3001TensorFlow\uff0c\u4ee5\u53ca\u5982\u4f55\u4f7f\u7528Jupyter Notebook\u8fdb\u884c\u5f00\u53d1\u3002\u901a\u8fc7\u8fd9\u4e2a\u7b80\u5355\u7684\u624b\u5199\u6570\u5b57\u8bc6\u522b\u5e94\u7528\uff0c\u60a8\u4e0d\u4ec5\u5b66\u4e60\u4e86AI\u7684\u57fa\u672c\u6982\u5ff5\uff0c\u8fd8\u4eb2\u8eab\u4f53\u9a8c\u4e86AI\u7684\u5e94\u7528\u3002AI\u662f\u4e00\u4e2a\u5e7f\u9614\u800c\u5145\u6ee1\u673a\u9047\u7684\u9886\u57df\uff0c\u5e0c\u671b\u8fd9\u4e2a\u6559\u7a0b\u80fd\u591f\u6fc0\u53d1\u60a8\u7ee7\u7eed\u6df1\u5165\u5b66\u4e60\u548c\u63a2\u7d22AI\u7684\u5174\u8da3\u3002
\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014<\/p>\n\n\n\n

                        \u7248\u6743\u58f0\u660e\uff1a\u672c\u6587\u4e3a\u535a\u4e3b\u539f\u521b\u6587\u7ae0\uff0c\u9075\u5faa CC 4.0 BY-SA \u7248\u6743\u534f\u8bae\uff0c\u8f6c\u8f7d\u8bf7\u9644\u4e0a\u539f\u6587\u51fa\u5904\u94fe\u63a5\u548c\u672c\u58f0\u660e\u3002<\/code><\/pre>\n\n\n\n
\u539f\u6587\u94fe\u63a5\uff1ahttps:\/\/blog.csdn.net\/ytt0523_com\/article\/details\/137588562<\/p>\n","protected":false},"excerpt":{"rendered":"
AI \u5165\u95e8\uff0c\u4ece\u96f6\u642d\u5efa\u5b8c\u6574 AI \u5f00\u53d1\u73af\u5883\uff0c\u5e76\u5199\u51fa\u7b2c\u4e00\u4e2a AI \u5e94\u7528 \u9ebb\u8fa3\u6392\u9aa8\u9762 \u5df2\u4e8e 2024-05-09 2<\/p>\n
\n\t\t\t\t Read More <\/span> <\/i><\/a>\n\t\t\t<\/div>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-910","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=\/wp\/v2\/posts\/910","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=910"}],"version-history":[{"count":1,"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=\/wp\/v2\/posts\/910\/revisions"}],"predecessor-version":[{"id":911,"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=\/wp\/v2\/posts\/910\/revisions\/911"}],"wp:attachment":[{"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=910"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=910"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.inhhh.com\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=910"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}