{"id":985992,"date":"2024-12-27T07:41:11","date_gmt":"2024-12-26T23:41:11","guid":{"rendered":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/985992.html"},"modified":"2024-12-27T07:41:13","modified_gmt":"2024-12-26T23:41:13","slug":"python%e5%a6%82%e4%bd%95%e7%ad%9b%e9%80%89%e5%a4%a7%e9%87%8f%e6%95%b0%e6%8d%ae","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/985992.html","title":{"rendered":"python\u5982\u4f55\u7b5b\u9009\u5927\u91cf\u6570\u636e"},"content":{"rendered":"<p style=\"text-align:center;\" ><img decoding=\"async\" src=\"https:\/\/cdn-kb.worktile.com\/kb\/wp-content\/uploads\/2024\/04\/25062755\/8f82e21b-3586-4522-b911-1b3730c83220.webp\" alt=\"python\u5982\u4f55\u7b5b\u9009\u5927\u91cf\u6570\u636e\" \/><\/p>\n<p><p> \u4e00\u3001Python\u7b5b\u9009\u5927\u91cf\u6570\u636e\u7684\u65b9\u6cd5\u6982\u8ff0<\/p>\n<\/p>\n<p><p>\u5728Python\u4e2d\uff0c\u7b5b\u9009\u5927\u91cf\u6570\u636e\u7684\u5e38\u7528\u65b9\u6cd5\u5305\u62ec\uff1a<strong>\u4f7f\u7528Pandas\u5e93\u5904\u7406\u6570\u636e\u3001\u5229\u7528Numpy\u8fdb\u884c\u6570\u503c\u8ba1\u7b97\u3001\u8fd0\u7528SQLalchemy\u4e0e\u6570\u636e\u5e93\u7ed3\u5408\u8fdb\u884c\u67e5\u8be2\u3001\u7ed3\u5408Dask\u5e93\u8fdb\u884c\u5927\u6570\u636e\u5904\u7406<\/strong>\u3002\u5176\u4e2d\uff0cPandas\u5e93\u662f\u6700\u5e38\u7528\u7684\u65b9\u6cd5\u4e4b\u4e00\uff0c\u56e0\u4e3a\u5176\u5177\u5907\u5904\u7406\u8868\u683c\u6570\u636e\u7684\u5f3a\u5927\u529f\u80fd\uff0c\u4e14\u6613\u4e8e\u5b66\u4e60\u548c\u4f7f\u7528\u3002\u901a\u8fc7Pandas\uff0c\u7528\u6237\u53ef\u4ee5\u8f7b\u677e\u5730\u8fdb\u884c\u6570\u636e\u7684\u8bfb\u53d6\u3001\u6e05\u6d17\u3001\u7b5b\u9009\u548c\u8f6c\u6362\u3002\u63a5\u4e0b\u6765\uff0c\u6211\u4eec\u5c06\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528Pandas\u5e93\u6765\u9ad8\u6548\u5730\u7b5b\u9009\u5927\u91cf\u6570\u636e\u3002<\/p>\n<\/p>\n<p><h3>\u4f7f\u7528Pandas\u5e93\u5904\u7406\u6570\u636e<\/h3>\n<\/p>\n<p><p>Pandas\u662f\u4e00\u4e2a\u5f3a\u5927\u7684Python\u5e93\uff0c\u4e13\u95e8\u7528\u4e8e\u6570\u636e\u64cd\u4f5c\u548c\u5206\u6790\u3002\u901a\u8fc7DataFrame\u5bf9\u8c61\uff0cPandas\u80fd\u591f\u4ee5\u884c\u548c\u5217\u7684\u5f62\u5f0f\u5b58\u50a8\u6570\u636e\uff0c\u7c7b\u4f3c\u4e8eExcel\u8868\u683c\u3002\u8981\u7b5b\u9009\u6570\u636e\uff0c\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528\u5e03\u5c14\u7d22\u5f15\u3001\u6761\u4ef6\u7b5b\u9009\u7b49\u591a\u79cd\u65b9\u6cd5\u3002<\/p>\n<\/p>\n<p><h4>1. \u8bfb\u53d6\u6570\u636e<\/h4>\n<\/p>\n<p><p>\u5728\u5f00\u59cb\u7b5b\u9009\u6570\u636e\u4e4b\u524d\uff0c\u9996\u5148\u9700\u8981\u5c06\u6570\u636e\u5bfc\u5165Pandas\u3002Pandas\u652f\u6301\u591a\u79cd\u6570\u636e\u683c\u5f0f\u7684\u8bfb\u53d6\uff0c\u5982CSV\u3001Excel\u3001SQL\u6570\u636e\u5e93\u7b49\u3002\u4ee5\u4e0b\u662f\u4eceCSV\u6587\u4ef6\u4e2d\u8bfb\u53d6\u6570\u636e\u7684\u793a\u4f8b\u4ee3\u7801\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import pandas as pd<\/p>\n<h2><strong>\u8bfb\u53d6CSV\u6587\u4ef6<\/strong><\/h2>\n<p>data = pd.read_csv(&#39;data.csv&#39;)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h4>2. \u57fa\u672c\u6570\u636e\u7b5b\u9009<\/h4>\n<\/p>\n<p><p>Pandas\u5141\u8bb8\u901a\u8fc7\u5e03\u5c14\u7d22\u5f15\u6765\u7b5b\u9009\u6570\u636e\u3002\u5e03\u5c14\u7d22\u5f15\u662f\u6839\u636e\u6761\u4ef6\u751f\u6210\u7684\u5e03\u5c14\u6570\u7ec4\uff0c\u7528\u4e8e\u7b5b\u9009\u6ee1\u8db3\u6761\u4ef6\u7684\u6570\u636e\u884c\u3002\u4f8b\u5982\uff0c\u7b5b\u9009\u51fa\u67d0\u5217\u6570\u636e\u5927\u4e8e\u67d0\u4e2a\u503c\u7684\u884c\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u7b5b\u9009\u51fa&#39;Sales&#39;\u5217\u5927\u4e8e500\u7684\u6570\u636e\u884c<\/p>\n<p>filtered_data = data[data[&#39;Sales&#39;] &gt; 500]<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h4>3. \u591a\u6761\u4ef6\u7b5b\u9009<\/h4>\n<\/p>\n<p><p>\u5f53\u9700\u8981\u6839\u636e\u591a\u4e2a\u6761\u4ef6\u7b5b\u9009\u6570\u636e\u65f6\uff0c\u53ef\u4ee5\u4f7f\u7528\u903b\u8f91\u8fd0\u7b97\u7b26\uff08&amp;\u3001|\u3001~\uff09\u7ed3\u5408\u5e03\u5c14\u7d22\u5f15\u3002\u4f8b\u5982\uff0c\u7b5b\u9009\u51fa&#39;Sales&#39;\u5217\u5927\u4e8e500\u4e14&#39;Category&#39;\u5217\u4e3a&#39;Electronics&#39;\u7684\u6570\u636e\u884c\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u7b5b\u9009\u51fa\u6ee1\u8db3\u591a\u4e2a\u6761\u4ef6\u7684\u6570\u636e\u884c<\/p>\n<p>filtered_data = data[(data[&#39;Sales&#39;] &gt; 500) &amp; (data[&#39;Category&#39;] == &#39;Electronics&#39;)]<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h4>4. \u4f7f\u7528\u67e5\u8be2\u65b9\u6cd5<\/h4>\n<\/p>\n<p><p>Pandas\u8fd8\u63d0\u4f9b\u4e86<code>query()<\/code>\u65b9\u6cd5\uff0c\u53ef\u4ee5\u901a\u8fc7\u5b57\u7b26\u4e32\u8868\u8fbe\u5f0f\u8fdb\u884c\u6570\u636e\u7b5b\u9009\u3002\u8fd9\u79cd\u65b9\u6cd5\u5728\u5904\u7406\u590d\u6742\u6761\u4ef6\u65f6\u66f4\u52a0\u76f4\u89c2\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u4f7f\u7528query\u65b9\u6cd5\u8fdb\u884c\u7b5b\u9009<\/p>\n<p>filtered_data = data.query(&#39;Sales &gt; 500 and Category == &quot;Electronics&quot;&#39;)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>\u5229\u7528Numpy\u8fdb\u884c\u6570\u503c\u8ba1\u7b97<\/h3>\n<\/p>\n<p><p>Numpy\u662fPython\u4e2d\u8fdb\u884c\u6570\u503c\u8ba1\u7b97\u7684\u57fa\u7840\u5e93\uff0c\u9002\u5408\u5904\u7406\u5927\u89c4\u6a21\u7684\u6570\u503c\u6570\u7ec4\u3002\u5728\u6570\u636e\u7b5b\u9009\u65b9\u9762\uff0cNumpy\u7684\u5e03\u5c14\u7d22\u5f15\u529f\u80fd\u53ef\u4ee5\u7528\u4e8e\u5feb\u901f\u7b5b\u9009\u6ee1\u8db3\u6761\u4ef6\u7684\u6570\u636e\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import numpy as np<\/p>\n<h2><strong>\u521b\u5efa\u4e00\u4e2a\u968f\u673a\u6570\u7ec4<\/strong><\/h2>\n<p>array = np.random.randint(0, 1000, size=(10000,))<\/p>\n<h2><strong>\u7b5b\u9009\u51fa\u5927\u4e8e500\u7684\u5143\u7d20<\/strong><\/h2>\n<p>filtered_array = array[array &gt; 500]<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>\u8fd0\u7528SQLalchemy\u4e0e\u6570\u636e\u5e93\u7ed3\u5408\u8fdb\u884c\u67e5\u8be2<\/h3>\n<\/p>\n<p><p>\u5bf9\u4e8e\u5b58\u50a8\u5728\u6570\u636e\u5e93\u4e2d\u7684\u5927\u89c4\u6a21\u6570\u636e\uff0cSQLalchemy\u63d0\u4f9b\u4e86\u4e00\u4e2aPythonic\u7684\u65b9\u5f0f\u6765\u8fdb\u884c\u6570\u636e\u5e93\u64cd\u4f5c\u3002\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528SQLalchemy\u6784\u5efaSQL\u67e5\u8be2\u8bed\u53e5\uff0c\u76f4\u63a5\u5728\u6570\u636e\u5e93\u4e2d\u8fdb\u884c\u6570\u636e\u7b5b\u9009\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">from sqlalchemy import create_engine, select, Table, MetaData<\/p>\n<h2><strong>\u521b\u5efa\u6570\u636e\u5e93\u8fde\u63a5<\/strong><\/h2>\n<p>engine = create_engine(&#39;sqlite:\/\/\/example.db&#39;)<\/p>\n<p>metadata = MetaData(bind=engine)<\/p>\n<h2><strong>\u53cd\u5c04\u8868\u7ed3\u6784<\/strong><\/h2>\n<p>table = Table(&#39;sales_data&#39;, metadata, autoload=True)<\/p>\n<h2><strong>\u6784\u5efa\u67e5\u8be2<\/strong><\/h2>\n<p>query = select([table]).where(table.c.sales &gt; 500)<\/p>\n<h2><strong>\u6267\u884c\u67e5\u8be2<\/strong><\/h2>\n<p>with engine.connect() as connection:<\/p>\n<p>    result = connection.execute(query)<\/p>\n<p>    filtered_data = result.fetchall()<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>\u7ed3\u5408Dask\u5e93\u8fdb\u884c\u5927\u6570\u636e\u5904\u7406<\/h3>\n<\/p>\n<p><p>\u5f53\u6570\u636e\u91cf\u975e\u5e38\u5927\uff0c\u4ee5\u81f3\u4e8e\u65e0\u6cd5\u5168\u90e8\u52a0\u8f7d\u5230\u5185\u5b58\u4e2d\u65f6\uff0cDask\u662f\u4e00\u4e2a\u5f88\u597d\u7684\u9009\u62e9\u3002Dask\u80fd\u591f\u5c06\u6570\u636e\u5212\u5206\u4e3a\u591a\u4e2a\u5206\u533a\uff0c\u5206\u6279\u52a0\u8f7d\u5230\u5185\u5b58\u4e2d\u8fdb\u884c\u5904\u7406\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import dask.dataframe as dd<\/p>\n<h2><strong>\u8bfb\u53d6\u5927\u89c4\u6a21CSV\u6587\u4ef6<\/strong><\/h2>\n<p>dask_df = dd.read_csv(&#39;large_data.csv&#39;)<\/p>\n<h2><strong>\u7b5b\u9009\u6570\u636e<\/strong><\/h2>\n<p>filtered_dask_df = dask_df[dask_df[&#39;Sales&#39;] &gt; 500]<\/p>\n<h2><strong>\u8ba1\u7b97\u7ed3\u679c<\/strong><\/h2>\n<p>result = filtered_dask_df.compute()<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u4e8c\u3001Pandas\u7684\u9ad8\u7ea7\u7b5b\u9009\u6280\u5de7<\/p>\n<\/p>\n<p><p>\u5728\u6570\u636e\u5206\u6790\u4e2d\uff0c\u9664\u4e86\u57fa\u672c\u7684\u6761\u4ef6\u7b5b\u9009\u5916\uff0cPandas\u8fd8\u63d0\u4f9b\u4e86\u4e00\u4e9b\u9ad8\u7ea7\u7684\u7b5b\u9009\u6280\u5de7\uff0c\u8fd9\u4e9b\u6280\u5de7\u53ef\u4ee5\u5e2e\u52a9\u6211\u4eec\u66f4\u9ad8\u6548\u5730\u5904\u7406\u590d\u6742\u7684\u6570\u636e\u7b5b\u9009\u4efb\u52a1\u3002<\/p>\n<\/p>\n<p><h3>1. \u4f7f\u7528<code>isin()<\/code>\u65b9\u6cd5<\/h3>\n<\/p>\n<p><p>\u5f53\u9700\u8981\u6839\u636e\u4e00\u7ec4\u503c\u7b5b\u9009\u6570\u636e\u65f6\uff0c<code>isin()<\/code>\u65b9\u6cd5\u975e\u5e38\u6709\u7528\u3002\u5b83\u53ef\u4ee5\u7528\u4e8e\u7b5b\u9009\u51fa\u67d0\u5217\u6570\u636e\u5c5e\u4e8e\u6307\u5b9a\u96c6\u5408\u7684\u884c\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u7b5b\u9009\u51fa&#39;Category&#39;\u5217\u5c5e\u4e8e\u6307\u5b9a\u96c6\u5408\u7684\u6570\u636e\u884c<\/p>\n<p>categories = [&#39;Electronics&#39;, &#39;Furniture&#39;]<\/p>\n<p>filtered_data = data[data[&#39;Category&#39;].isin(categories)]<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>2. \u4f7f\u7528<code>between()<\/code>\u65b9\u6cd5<\/h3>\n<\/p>\n<p><p><code>between()<\/code>\u65b9\u6cd5\u7528\u4e8e\u7b5b\u9009\u51fa\u67d0\u5217\u6570\u636e\u5728\u6307\u5b9a\u8303\u56f4\u5185\u7684\u884c\u3002\u8fd9\u79cd\u65b9\u6cd5\u5bf9\u4e8e\u6570\u503c\u8303\u56f4\u7b5b\u9009\u975e\u5e38\u65b9\u4fbf\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u7b5b\u9009\u51fa&#39;Sales&#39;\u5217\u5728100\u5230500\u4e4b\u95f4\u7684\u6570\u636e\u884c<\/p>\n<p>filtered_data = data[data[&#39;Sales&#39;].between(100, 500)]<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>3. \u4f7f\u7528\u5b57\u7b26\u4e32\u65b9\u6cd5<\/h3>\n<\/p>\n<p><p>Pandas\u63d0\u4f9b\u4e86\u4e00\u7ec4\u5b57\u7b26\u4e32\u65b9\u6cd5\uff0c\u53ef\u4ee5\u7528\u4e8e\u5b57\u7b26\u4e32\u6570\u636e\u7684\u7b5b\u9009\u3002\u4f8b\u5982\uff0c\u7b5b\u9009\u51fa\u67d0\u5217\u4ee5\u7279\u5b9a\u524d\u7f00\u5f00\u5934\u7684\u6570\u636e\u884c\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u7b5b\u9009\u51fa&#39;Product&#39;\u5217\u4ee5&#39;Pro&#39;\u5f00\u5934\u7684\u6570\u636e\u884c<\/p>\n<p>filtered_data = data[data[&#39;Product&#39;].str.startswith(&#39;Pro&#39;)]<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>4. \u4f7f\u7528\u6b63\u5219\u8868\u8fbe\u5f0f<\/h3>\n<\/p>\n<p><p>Pandas\u7684<code>str.cont<a href=\"https:\/\/docs.pingcode.com\/blog\/59162.html\" target=\"_blank\">AI<\/a>ns()<\/code>\u65b9\u6cd5\u5141\u8bb8\u4f7f\u7528\u6b63\u5219\u8868\u8fbe\u5f0f\u8fdb\u884c\u6570\u636e\u7b5b\u9009\u3002\u4f8b\u5982\uff0c\u7b5b\u9009\u51fa\u67d0\u5217\u5305\u542b\u7279\u5b9a\u6a21\u5f0f\u7684\u6570\u636e\u884c\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u7b5b\u9009\u51fa&#39;Product&#39;\u5217\u5305\u542b&#39;Pro&#39;\u7684\u884c\uff0c\u5ffd\u7565\u5927\u5c0f\u5199<\/p>\n<p>filtered_data = data[data[&#39;Product&#39;].str.contains(&#39;Pro&#39;, case=False, regex=True)]<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>5. \u4f7f\u7528<code>apply()<\/code>\u51fd\u6570<\/h3>\n<\/p>\n<p><p>\u5f53\u9700\u8981\u57fa\u4e8e\u590d\u6742\u903b\u8f91\u8fdb\u884c\u7b5b\u9009\u65f6\uff0c\u53ef\u4ee5\u4f7f\u7528<code>apply()<\/code>\u51fd\u6570\u3002<code>apply()<\/code>\u51fd\u6570\u5141\u8bb8\u5bf9DataFrame\u7684\u6bcf\u4e00\u884c\u6216\u6bcf\u4e00\u5217\u5e94\u7528\u81ea\u5b9a\u4e49\u51fd\u6570\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u5b9a\u4e49\u81ea\u5b9a\u4e49\u7b5b\u9009\u51fd\u6570<\/p>\n<p>def custom_filter(row):<\/p>\n<p>    return row[&#39;Sales&#39;] &gt; 500 and row[&#39;Category&#39;] == &#39;Electronics&#39;<\/p>\n<h2><strong>\u4f7f\u7528apply\u51fd\u6570\u8fdb\u884c\u7b5b\u9009<\/strong><\/h2>\n<p>filtered_data = data[data.apply(custom_filter, axis=1)]<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u4e09\u3001\u63d0\u9ad8\u6570\u636e\u7b5b\u9009\u6027\u80fd\u7684\u6280\u5de7<\/p>\n<\/p>\n<p><p>\u5728\u5904\u7406\u5927\u89c4\u6a21\u6570\u636e\u65f6\uff0c\u6027\u80fd\u662f\u4e00\u4e2a\u91cd\u8981\u7684\u8003\u91cf\u56e0\u7d20\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u63d0\u9ad8\u6570\u636e\u7b5b\u9009\u6027\u80fd\u7684\u6280\u5de7\uff1a<\/p>\n<\/p>\n<p><h3>1. \u4f7f\u7528\u77e2\u91cf\u5316\u64cd\u4f5c<\/h3>\n<\/p>\n<p><p>Pandas\u548cNumpy\u90fd\u652f\u6301\u77e2\u91cf\u5316\u64cd\u4f5c\uff0c\u80fd\u591f\u5728\u4e0d\u4f7f\u7528\u5faa\u73af\u7684\u60c5\u51b5\u4e0b\u5bf9\u6570\u636e\u8fdb\u884c\u6279\u91cf\u5904\u7406\u3002\u77e2\u91cf\u5316\u64cd\u4f5c\u901a\u5e38\u6bd4\u5faa\u73af\u66f4\u9ad8\u6548\u3002<\/p>\n<\/p>\n<p><h3>2. \u51cf\u5c11DataFrame\u7684\u590d\u5236<\/h3>\n<\/p>\n<p><p>\u5728\u8fdb\u884c\u6570\u636e\u7b5b\u9009\u65f6\uff0c\u5c3d\u91cf\u907f\u514d\u4e0d\u5fc5\u8981\u7684DataFrame\u590d\u5236\uff0c\u56e0\u4e3a\u590d\u5236\u4f1a\u589e\u52a0\u5185\u5b58\u6d88\u8017\u5e76\u964d\u4f4e\u6027\u80fd\u3002\u53ef\u4ee5\u901a\u8fc7\u94fe\u5f0f\u64cd\u4f5c\u6765\u51cf\u5c11\u590d\u5236\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u94fe\u5f0f\u64cd\u4f5c<\/p>\n<p>filtered_data = data[data[&#39;Sales&#39;] &gt; 500].copy()<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>3. \u4f7f\u7528\u5206\u5757\u5904\u7406<\/h3>\n<\/p>\n<p><p>\u5bf9\u4e8e\u65e0\u6cd5\u4e00\u6b21\u6027\u52a0\u8f7d\u5230\u5185\u5b58\u4e2d\u7684\u5927\u6570\u636e\u96c6\uff0c\u53ef\u4ee5\u4f7f\u7528\u5206\u5757\u5904\u7406\u3002Pandas\u7684<code>read_csv()<\/code>\u51fd\u6570\u652f\u6301\u5206\u5757\u8bfb\u53d6\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\"># \u5206\u5757\u8bfb\u53d6\u5927\u6570\u636e\u96c6<\/p>\n<p>chunk_iter = pd.read_csv(&#39;large_data.csv&#39;, chunksize=10000)<\/p>\n<h2><strong>\u5bf9\u6bcf\u4e2a\u5757\u8fdb\u884c\u7b5b\u9009<\/strong><\/h2>\n<p>filtered_chunks = [chunk[chunk[&#39;Sales&#39;] &gt; 500] for chunk in chunk_iter]<\/p>\n<h2><strong>\u5408\u5e76\u7ed3\u679c<\/strong><\/h2>\n<p>filtered_data = pd.concat(filtered_chunks)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>4. \u4f7f\u7528Dask\u8fdb\u884c\u5e76\u884c\u5904\u7406<\/h3>\n<\/p>\n<p><p>Dask\u901a\u8fc7\u5e76\u884c\u5904\u7406\u63d0\u9ad8\u4e86\u5927\u6570\u636e\u96c6\u7684\u5904\u7406\u6027\u80fd\u3002\u5b83\u5c06\u6570\u636e\u5212\u5206\u4e3a\u591a\u4e2a\u5206\u533a\uff0c\u5e76\u5728\u591a\u4e2aCPU\u6838\u5fc3\u4e0a\u540c\u65f6\u8fdb\u884c\u8ba1\u7b97\u3002<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import dask.dataframe as dd<\/p>\n<h2><strong>\u8bfb\u53d6\u5927\u89c4\u6a21CSV\u6587\u4ef6<\/strong><\/h2>\n<p>dask_df = dd.read_csv(&#39;large_data.csv&#39;)<\/p>\n<h2><strong>\u7b5b\u9009\u6570\u636e<\/strong><\/h2>\n<p>filtered_dask_df = dask_df[dask_df[&#39;Sales&#39;] &gt; 500]<\/p>\n<h2><strong>\u8ba1\u7b97\u7ed3\u679c<\/strong><\/h2>\n<p>result = filtered_dask_df.compute()<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u56db\u3001\u7ed3\u5408Python\u751f\u6001\u7cfb\u7edf\u7684\u5176\u4ed6\u5de5\u5177<\/p>\n<\/p>\n<p><p>Python\u7684\u751f\u6001\u7cfb\u7edf\u4e2d\u6709\u8bb8\u591a\u5de5\u5177\u53ef\u4ee5\u5e2e\u52a9\u6211\u4eec\u66f4\u9ad8\u6548\u5730\u7b5b\u9009\u548c\u5904\u7406\u5927\u89c4\u6a21\u6570\u636e\u3002<\/p>\n<\/p>\n<p><h3>1. \u4f7f\u7528SQLalchemy\u4e0e\u6570\u636e\u5e93\u7ed3\u5408<\/h3>\n<\/p>\n<p><p>\u5bf9\u4e8e\u5b58\u50a8\u5728\u6570\u636e\u5e93\u4e2d\u7684\u5927\u89c4\u6a21\u6570\u636e\uff0cSQLalchemy\u63d0\u4f9b\u4e86\u4e00\u4e2aPythonic\u7684\u65b9\u5f0f\u6765\u8fdb\u884c\u6570\u636e\u5e93\u64cd\u4f5c\u3002\u901a\u8fc7\u5c06\u67e5\u8be2\u903b\u8f91\u79fb\u52a8\u5230\u6570\u636e\u5e93\u5c42\uff0c\u53ef\u4ee5\u51cf\u5c11\u7f51\u7edc\u4f20\u8f93\u548c\u6570\u636e\u52a0\u8f7d\u7684\u5f00\u9500\u3002<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">from sqlalchemy import create_engine, select, Table, MetaData<\/p>\n<h2><strong>\u521b\u5efa\u6570\u636e\u5e93\u8fde\u63a5<\/strong><\/h2>\n<p>engine = create_engine(&#39;sqlite:\/\/\/example.db&#39;)<\/p>\n<p>metadata = MetaData(bind=engine)<\/p>\n<h2><strong>\u53cd\u5c04\u8868\u7ed3\u6784<\/strong><\/h2>\n<p>table = Table(&#39;sales_data&#39;, metadata, autoload=True)<\/p>\n<h2><strong>\u6784\u5efa\u67e5\u8be2<\/strong><\/h2>\n<p>query = select([table]).where(table.c.sales &gt; 500)<\/p>\n<h2><strong>\u6267\u884c\u67e5\u8be2<\/strong><\/h2>\n<p>with engine.connect() as connection:<\/p>\n<p>    result = connection.execute(query)<\/p>\n<p>    filtered_data = result.fetchall()<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>2. \u4f7f\u7528PySpark\u8fdb\u884c\u5206\u5e03\u5f0f\u8ba1\u7b97<\/h3>\n<\/p>\n<p><p>\u5f53\u6570\u636e\u91cf\u8fbe\u5230TB\u7ea7\u522b\u65f6\uff0c\u5355\u53f0\u673a\u5668\u53ef\u80fd\u65e0\u6cd5\u5904\u7406\u6b64\u7c7b\u6570\u636e\u3002\u6b64\u65f6\uff0cPySpark\u53ef\u4ee5\u901a\u8fc7\u5206\u5e03\u5f0f\u8ba1\u7b97\u6765\u5904\u7406\u5927\u89c4\u6a21\u6570\u636e\u96c6\u3002<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">from pyspark.sql import SparkSession<\/p>\n<h2><strong>\u521b\u5efaSparkSession<\/strong><\/h2>\n<p>spark = SparkSession.builder.appName(&#39;DataFiltering&#39;).getOrCreate()<\/p>\n<h2><strong>\u8bfb\u53d6\u6570\u636e<\/strong><\/h2>\n<p>df = spark.read.csv(&#39;large_data.csv&#39;, header=True, inferSchema=True)<\/p>\n<h2><strong>\u7b5b\u9009\u6570\u636e<\/strong><\/h2>\n<p>filtered_df = df.filter(df[&#39;Sales&#39;] &gt; 500)<\/p>\n<h2><strong>\u663e\u793a\u7ed3\u679c<\/strong><\/h2>\n<p>filtered_df.show()<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><h3>3. \u4f7f\u7528Vaex\u8fdb\u884c\u5feb\u901f\u6570\u636e\u5206\u6790<\/h3>\n<\/p>\n<p><p>Vaex\u662f\u4e00\u4e2a\u5feb\u901f\u7684\u6570\u636e\u5206\u6790\u5e93\uff0c\u4e13\u4e3a\u5927\u89c4\u6a21\u6570\u636e\u96c6\u8bbe\u8ba1\u3002\u5b83\u5728\u5185\u5b58\u4e2d\u64cd\u4f5c\u6570\u636e\u96c6\u7684\u5ef6\u8fdf\u52a0\u8f7d\u548c\u5185\u5b58\u6620\u5c04\uff0c\u4f7f\u5176\u80fd\u591f\u5904\u7406\u6bd4\u5185\u5b58\u66f4\u5927\u7684\u6570\u636e\u96c6\u3002<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import vaex<\/p>\n<h2><strong>\u8bfb\u53d6\u6570\u636e<\/strong><\/h2>\n<p>df = vaex.open(&#39;large_data.csv&#39;)<\/p>\n<h2><strong>\u7b5b\u9009\u6570\u636e<\/strong><\/h2>\n<p>filtered_df = df[df[&#39;Sales&#39;] &gt; 500]<\/p>\n<h2><strong>\u663e\u793a\u7ed3\u679c<\/strong><\/h2>\n<p>filtered_df.head()<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u4e94\u3001\u603b\u7ed3<\/p>\n<\/p>\n<p><p>Python\u63d0\u4f9b\u4e86\u591a\u79cd\u5de5\u5177\u548c\u65b9\u6cd5\u6765\u7b5b\u9009\u5927\u91cf\u6570\u636e\uff0c\u6bcf\u79cd\u65b9\u6cd5\u90fd\u6709\u5176\u7279\u5b9a\u7684\u9002\u7528\u573a\u666f\u3002Pandas\u9002\u5408\u5904\u7406\u4e2d\u5c0f\u89c4\u6a21\u6570\u636e\uff0c\u5e76\u63d0\u4f9b\u4e86\u4e30\u5bcc\u7684\u529f\u80fd\u6765\u8fdb\u884c\u6570\u636e\u7b5b\u9009\u548c\u5206\u6790\u3002\u5bf9\u4e8e\u5927\u89c4\u6a21\u6570\u636e\uff0cDask\u3001SQLalchemy\u3001PySpark\u548cVaex\u7b49\u5de5\u5177\u63d0\u4f9b\u4e86\u66f4\u9ad8\u6548\u7684\u89e3\u51b3\u65b9\u6848\u3002\u901a\u8fc7\u7ed3\u5408\u4f7f\u7528\u8fd9\u4e9b\u5de5\u5177\uff0c\u6211\u4eec\u53ef\u4ee5\u6839\u636e\u5177\u4f53\u9700\u6c42\u9009\u62e9\u5408\u9002\u7684\u65b9\u6cd5\u6765\u8fdb\u884c\u6570\u636e\u7b5b\u9009\u548c\u5904\u7406\u3002\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\uff0c\u7406\u89e3\u6570\u636e\u7684\u7279\u70b9\u3001\u5408\u7406\u9009\u62e9\u5de5\u5177\u3001\u4f18\u5316\u5904\u7406\u6d41\u7a0b\uff0c\u80fd\u591f\u663e\u8457\u63d0\u9ad8\u6570\u636e\u5904\u7406\u7684\u6548\u7387\u548c\u6027\u80fd\u3002<\/p>\n<\/p>\n<h2><strong>\u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n<p> <strong>\u5982\u4f55\u5728Python\u4e2d\u9ad8\u6548\u7b5b\u9009\u5927\u578b\u6570\u636e\u96c6\uff1f<\/strong><br \/>\u5728Python\u4e2d\uff0c\u4f7f\u7528pandas\u5e93\u53ef\u4ee5\u975e\u5e38\u9ad8\u6548\u5730\u7b5b\u9009\u5927\u91cf\u6570\u636e\u3002\u9996\u5148\uff0c\u786e\u4fdd\u4f60\u7684\u6570\u636e\u5df2\u52a0\u8f7d\u4e3aDataFrame\u5bf9\u8c61\u3002\u53ef\u4ee5\u5229\u7528\u6761\u4ef6\u7d22\u5f15\u3001\u5e03\u5c14\u7d22\u5f15\u548cquery\u65b9\u6cd5\u6765\u8fdb\u884c\u7b5b\u9009\u3002\u4f8b\u5982\uff0c\u5982\u679c\u4f60\u60f3\u7b5b\u9009\u51fa\u67d0\u5217\u5927\u4e8e\u7279\u5b9a\u503c\u7684\u884c\uff0c\u53ef\u4ee5\u76f4\u63a5\u4f7f\u7528<code>df[df[&#39;column_name&#39;] &gt; value]<\/code>\u3002\u6b64\u5916\uff0c\u4f7f\u7528<code>pandas<\/code>\u7684apply\u51fd\u6570\u53ef\u4ee5\u5bf9\u590d\u6742\u6761\u4ef6\u8fdb\u884c\u7b5b\u9009\u3002<\/p>\n<p><strong>\u5bf9\u4e8e\u5927\u578b\u6570\u636e\u96c6\uff0c\u662f\u5426\u6709\u63a8\u8350\u7684\u7b5b\u9009\u65b9\u6cd5\uff1f<\/strong><br \/>\u9488\u5bf9\u5927\u578b\u6570\u636e\u96c6\uff0c\u4f7f\u7528pandas\u7684\u5185\u5b58\u4f18\u5316\u6280\u5de7\u975e\u5e38\u91cd\u8981\u3002\u53ef\u4ee5\u8003\u8651\u5c06\u6570\u636e\u5206\u5757\u8bfb\u53d6\uff0c\u4f7f\u7528<code>chunksize<\/code>\u53c2\u6570\u6765\u5904\u7406\u6570\u636e\u3002\u6b64\u5916\uff0c\u4f7f\u7528<code>dask<\/code>\u5e93\u4e5f\u662f\u4e00\u4e2a\u4e0d\u9519\u7684\u9009\u62e9\uff0c\u5b83\u53ef\u4ee5\u5904\u7406\u8d85\u51fa\u5185\u5b58\u7684\u6570\u636e\u96c6\uff0c\u901a\u8fc7\u5e76\u884c\u8ba1\u7b97\u6765\u52a0\u5feb\u7b5b\u9009\u8fc7\u7a0b\u3002\u53e6\u4e00\u79cd\u9009\u62e9\u662f\u4f7f\u7528SQLAlchemy\u4e0e\u6570\u636e\u5e93\u7ed3\u5408\uff0c\u8fd9\u6837\u53ef\u4ee5\u5728\u6570\u636e\u5e93\u7aef\u8fdb\u884c\u7b5b\u9009\uff0c\u4ece\u800c\u51cf\u5c11\u6570\u636e\u4f20\u8f93\u7684\u8d1f\u62c5\u3002<\/p>\n<p><strong>\u5728Python\u4e2d\u5982\u4f55\u5904\u7406\u7f3a\u5931\u6570\u636e\u4ee5\u4fbf\u4e8e\u7b5b\u9009\uff1f<\/strong><br \/>\u5904\u7406\u7f3a\u5931\u6570\u636e\u662f\u7b5b\u9009\u8fc7\u7a0b\u4e2d\u975e\u5e38\u91cd\u8981\u7684\u4e00\u6b65\u3002\u5728pandas\u4e2d\uff0c\u53ef\u4ee5\u4f7f\u7528<code>dropna()<\/code>\u65b9\u6cd5\u5220\u9664\u5305\u542b\u7f3a\u5931\u503c\u7684\u884c\uff0c\u6216\u8005\u4f7f\u7528<code>fillna()<\/code>\u65b9\u6cd5\u586b\u5145\u7f3a\u5931\u503c\u3002\u9009\u62e9\u5408\u9002\u7684\u586b\u5145\u65b9\u6cd5\uff08\u5982\u5747\u503c\u3001\u4e2d\u4f4d\u6570\u6216\u7279\u5b9a\u503c\uff09\u80fd\u591f\u4fdd\u8bc1\u6570\u636e\u7684\u5b8c\u6574\u6027\u548c\u51c6\u786e\u6027\uff0c\u4ece\u800c\u63d0\u9ad8\u540e\u7eed\u7b5b\u9009\u7684\u6548\u7387\u548c\u7ed3\u679c\u7684\u53ef\u9760\u6027\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"\u4e00\u3001Python\u7b5b\u9009\u5927\u91cf\u6570\u636e\u7684\u65b9\u6cd5\u6982\u8ff0 \u5728Python\u4e2d\uff0c\u7b5b\u9009\u5927\u91cf\u6570\u636e\u7684\u5e38\u7528\u65b9\u6cd5\u5305\u62ec\uff1a\u4f7f\u7528Pandas\u5e93\u5904\u7406\u6570 [&hellip;]","protected":false},"author":3,"featured_media":985996,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[37],"tags":[],"acf":[],"_links":{"self":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/985992"}],"collection":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/comments?post=985992"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/985992\/revisions"}],"predecessor-version":[{"id":985997,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/985992\/revisions\/985997"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/985996"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=985992"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=985992"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=985992"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}