{"id":16022,"date":"2024-01-19T22:49:33","date_gmt":"2024-01-19T19:19:33","guid":{"rendered":"https:\/\/rasanegar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/"},"modified":"2024-01-19T22:49:33","modified_gmt":"2024-01-19T19:19:33","slug":"%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c","status":"publish","type":"post","link":"https:\/\/rasanegaar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/","title":{"rendered":"\u0645\u0642\u062f\u0645\u0647 \u0627\u06cc \u0628\u0631 PyTorch \u0628\u0631\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc"},"content":{"rendered":"<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_85 counter-hierarchy ez-toc-counter ez-toc-custom ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\"><p class=\"ez-toc-title\" style=\"cursor:inherit\">\u0633\u0631\u0641\u0635\u0644\u0647\u0627\u06cc \u0645\u0637\u0644\u0628<\/p>\n<\/div><nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/rasanegaar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/#%d9%85%d8%ac%d9%85%d9%88%d8%b9%d9%87_%d8%af%d8%a7%d8%af%d9%87\" >\u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/rasanegaar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/#%d8%aa%d8%ac%d8%b2%db%8c%d9%87_%d9%88_%d8%aa%d8%ad%d9%84%db%8c%d9%84_%d8%af%d8%a7%d8%af%d9%87_%d9%87%d8%a7%db%8c_%d8%a7%da%a9%d8%aa%d8%b4%d8%a7%d9%81%db%8c\" >\u062a\u062c\u0632\u06cc\u0647 \u0648 \u062a\u062d\u0644\u06cc\u0644 \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u0627\u06a9\u062a\u0634\u0627\u0641\u06cc<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/rasanegaar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/#%d9%be%db%8c%d8%b4_%d9%be%d8%b1%d8%af%d8%a7%d8%b2%d8%b4_%d8%af%d8%a7%d8%af%d9%87_%d9%87%d8%a7\" >\u067e\u06cc\u0634 \u067e\u0631\u062f\u0627\u0632\u0634 \u062f\u0627\u062f\u0647 \u0647\u0627<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/rasanegaar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/#%d8%a7%db%8c%d8%ac%d8%a7%d8%af_%db%8c%da%a9_%d9%85%d8%af%d9%84_%d8%a8%d8%b1%d8%a7%db%8c_%d9%be%db%8c%d8%b4_%d8%a8%db%8c%d9%86%db%8c\" >\u0627\u06cc\u062c\u0627\u062f \u06cc\u06a9 \u0645\u062f\u0644 \u0628\u0631\u0627\u06cc \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/rasanegaar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/#%d8%a2%d9%85%d9%88%d8%b2%d8%b4_%d9%85%d8%af%d9%84\" >\u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/rasanegaar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/#%d9%be%db%8c%d8%b4%da%af%d9%88%db%8c%db%8c\" >\u067e\u06cc\u0634\u06af\u0648\u06cc\u06cc<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"https:\/\/rasanegaar.com\/blog\/%d9%85%d9%82%d8%af%d9%85%d9%87-%d8%a7%db%8c-%d8%a8%d8%b1-pytorch-%d8%a8%d8%b1%d8%a7%db%8c-%d8%b7%d8%a8%d9%82%d9%87-%d8%a8%d9%86%d8%af%db%8c\/#%d9%86%d8%aa%db%8c%d8%ac%d9%87\" >\u0646\u062a\u06cc\u062c\u0647<\/a><\/li><\/ul><\/nav><\/div>\n<span class=\"span-reading-time rt-reading-time\" style=\"display: block;\"><span class=\"rt-label rt-prefix\">\u0632\u0645\u0627\u0646 \u0644\u0627\u0632\u0645 \u0628\u0631\u0627\u06cc \u0645\u0637\u0627\u0644\u0639\u0647: <\/span> <span class=\"rt-time\"> 13<\/span> <span class=\"rt-label rt-postfix\">\u062f\u0642\u06cc\u0642\u0647<\/span><\/span><p> <br \/>\n<\/p>\n<div><noscript><\/noscript><\/p>\n<p><a rel=\"nofollow noopener\" target=\"_blank\" href=\"https:\/\/pytorch.org\/\">PyTorch<\/a> \u0648 <a rel=\"nofollow noopener\" target=\"_blank\" href=\"https:\/\/www.tensorflow.org\/\" class=\"broken_link\">TensorFlow<\/a> \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 \u0647\u0627 \u062f\u0648 \u0645\u0648\u0631\u062f \u0627\u0632 \u0631\u0627\u06cc\u062c \u062a\u0631\u06cc\u0646 \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 \u0647\u0627\u06cc \u067e\u0627\u06cc\u062a\u0648\u0646 \u0628\u0631\u0627\u06cc \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u0639\u0645\u06cc\u0642 \u0647\u0633\u062a\u0646\u062f.  PyTorch \u062a\u0648\u0633\u0637 \u0641\u06cc\u0633 \u0628\u0648\u06a9 \u062a\u0648\u0633\u0639\u0647 \u062f\u0627\u062f\u0647 \u0634\u062f\u0647 \u0627\u0633\u062a\u060c \u062f\u0631 \u062d\u0627\u0644\u06cc \u06a9\u0647 TensorFlow \u06cc\u06a9 \u067e\u0631\u0648\u0698\u0647 \u06af\u0648\u06af\u0644 \u0627\u0633\u062a.  \u062f\u0631 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u062e\u0648\u0627\u0647\u06cc\u062f \u062f\u06cc\u062f \u06a9\u0647 \u0686\u06af\u0648\u0646\u0647 \u0645\u06cc \u062a\u0648\u0627\u0646 \u0627\u0632 \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 PyTorch \u0628\u0631\u0627\u06cc \u062d\u0644 \u0645\u0634\u06a9\u0644\u0627\u062a \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u06a9\u0631\u062f.<\/p>\n<p>\u0645\u0633\u0627\u0626\u0644 \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0628\u0647 \u062f\u0633\u062a\u0647 \u0645\u0634\u06a9\u0644\u0627\u062a \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u0645\u0627\u0634\u06cc\u0646 \u062a\u0639\u0644\u0642 \u062f\u0627\u0631\u0646\u062f \u06a9\u0647 \u0628\u0627 \u062a\u0648\u062c\u0647 \u0628\u0647 \u0645\u062c\u0645\u0648\u0639\u0647\u200c\u0627\u06cc \u0627\u0632 \u0648\u06cc\u0698\u06af\u06cc\u200c\u0647\u0627\u060c \u0648\u0638\u06cc\u0641\u0647 \u067e\u06cc\u0634\u200c\u0628\u06cc\u0646\u06cc \u06cc\u06a9 \u0645\u0642\u062f\u0627\u0631 \u06af\u0633\u0633\u062a\u0647 \u0627\u0633\u062a.  \u067e\u06cc\u0634\u200c\u0628\u06cc\u0646\u06cc \u0633\u0631\u0637\u0627\u0646\u06cc \u0628\u0648\u062f\u0646 \u06cc\u0627 \u0646\u0628\u0648\u062f\u0646 \u062a\u0648\u0645\u0648\u0631\u060c \u06cc\u0627 \u0627\u06cc\u0646\u06a9\u0647 \u062f\u0627\u0646\u0634\u200c\u0622\u0645\u0648\u0632 \u0627\u062d\u062a\u0645\u0627\u0644\u0627\u064b \u062f\u0631 \u0627\u0645\u062a\u062d\u0627\u0646 \u0645\u0648\u0641\u0642 \u0645\u06cc\u200c\u0634\u0648\u062f \u06cc\u0627 \u0631\u062f \u0645\u06cc\u200c\u0634\u0648\u062f\u060c \u0627\u0632 \u0646\u0645\u0648\u0646\u0647\u200c\u0647\u0627\u06cc \u0631\u0627\u06cc\u062c \u0645\u0634\u06a9\u0644\u0627\u062a \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0647\u0633\u062a\u0646\u062f.<\/p>\n<p>\u062f\u0631 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u0628\u0627 \u062a\u0648\u062c\u0647 \u0628\u0647 \u0648\u06cc\u0698\u06af\u06cc \u0647\u0627\u06cc \u062e\u0627\u0635 \u06cc\u06a9 \u0645\u0634\u062a\u0631\u06cc \u0628\u0627\u0646\u06a9\u060c \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc \u0645\u06cc \u06a9\u0646\u06cc\u0645 \u06a9\u0647 \u0622\u06cc\u0627 \u0645\u0634\u062a\u0631\u06cc \u067e\u0633 \u0627\u0632 6 \u0645\u0627\u0647 \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u0645\u06cc \u06a9\u0646\u062f \u06cc\u0627 \u062e\u06cc\u0631.  \u067e\u062f\u06cc\u062f\u0647 \u0627\u06cc \u06a9\u0647 \u062f\u0631 \u0622\u0646 \u0645\u0634\u062a\u0631\u06cc \u0633\u0627\u0632\u0645\u0627\u0646 \u0631\u0627 \u062a\u0631\u06a9 \u0645\u06cc \u06a9\u0646\u062f\u060c \u0631\u06cc\u0632\u0634 \u0645\u0634\u062a\u0631\u06cc \u0646\u06cc\u0632 \u0646\u0627\u0645\u06cc\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f.  \u0628\u0646\u0627\u0628\u0631\u0627\u06cc\u0646\u060c \u0648\u0638\u06cc\u0641\u0647 \u0645\u0627 \u0627\u06cc\u0646 \u0627\u0633\u062a \u06a9\u0647 \u0628\u0631 \u0627\u0633\u0627\u0633 \u0631\u06cc\u0632\u0634 \u0645\u0634\u062a\u0631\u06cc \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc \u06a9\u0646\u06cc\u0645 \u0631\u0648\u06cc \u0648\u06cc\u0698\u06af\u06cc \u0647\u0627\u06cc \u0645\u062e\u062a\u0644\u0641 \u0645\u0634\u062a\u0631\u06cc<\/p>\n<p>\u0642\u0628\u0644 \u0627\u0632 \u0627\u062f\u0627\u0645\u0647\u060c \u0641\u0631\u0636 \u0628\u0631 \u0627\u06cc\u0646 \u0627\u0633\u062a \u06a9\u0647 \u0633\u0637\u062d \u0645\u062a\u0648\u0633\u0637\u06cc \u0627\u0632 \u0632\u0628\u0627\u0646 \u0628\u0631\u0646\u0627\u0645\u0647 \u0646\u0648\u06cc\u0633\u06cc \u067e\u0627\u06cc\u062a\u0648\u0646 \u062f\u0627\u0631\u06cc\u062f \u0648 \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 PyTorch \u0631\u0627 \u0646\u0635\u0628 \u06a9\u0631\u062f\u0647 \u0627\u06cc\u062f.  \u0647\u0645\u0686\u0646\u06cc\u0646\u060c \u062f\u0627\u0646\u0634 \u0645\u0641\u0627\u0647\u06cc\u0645 \u0627\u0648\u0644\u06cc\u0647 \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u0645\u0627\u0634\u06cc\u0646 \u0645\u0645\u06a9\u0646 \u0627\u0633\u062a \u06a9\u0645\u06a9 \u06a9\u0646\u062f.  \u0627\u06af\u0631 PyTorch \u0631\u0627 \u0646\u0635\u0628 \u0646\u06a9\u0631\u062f\u0647 \u0627\u06cc\u062f\u060c \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f \u0628\u0627 \u0645\u0648\u0627\u0631\u062f \u0632\u06cc\u0631 \u0627\u06cc\u0646 \u06a9\u0627\u0631 \u0631\u0627 \u0627\u0646\u062c\u0627\u0645 \u062f\u0647\u06cc\u062f pip \u062f\u0633\u062a\u0648\u0631:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-meta\">$<\/span><span class=\"bash\"> pip install pytorch<\/span>\n<\/code><\/pre>\n<h2 id=\"thedataset\"><span class=\"ez-toc-section\" id=\"%d9%85%d8%ac%d9%85%d9%88%d8%b9%d9%87_%d8%af%d8%a7%d8%af%d9%87\"><\/span>\u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>\u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0627\u06cc \u06a9\u0647 \u0645\u06cc \u062e\u0648\u0627\u0647\u06cc\u0645 \u062f\u0631 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u06a9\u0646\u06cc\u0645 \u0628\u0647 \u0635\u0648\u0631\u062a \u0631\u0627\u06cc\u06af\u0627\u0646 \u062f\u0631 \u0627\u06cc\u0646\u062c\u0627 \u0645\u0648\u062c\u0648\u062f \u0627\u0633\u062a <a rel=\"nofollow noopener\" target=\"_blank\" href=\"https:\/\/www.kaggle.com\/kmalit\/bank-customer-churn-prediction\/data\">\u0644\u06cc\u0646\u06a9 \u06a9\u0627\u06af\u0644<\/a>.  \u0627\u062c\u0627\u0632\u0647 \u062f\u0647\u06cc\u062f import \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 \u0647\u0627\u06cc \u0645\u0648\u0631\u062f \u0646\u06cc\u0627\u0632 \u0648 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u062f\u0631 \u0628\u0631\u0646\u0627\u0645\u0647 \u067e\u0627\u06cc\u062a\u0648\u0646 \u0645\u0627:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-keyword\">import<\/span> torch\n<span class=\"hljs-keyword\">import<\/span> torch.nn <span class=\"hljs-keyword\">as<\/span> nn\n<span class=\"hljs-keyword\">import<\/span> numpy <span class=\"hljs-keyword\">as<\/span> np\n<span class=\"hljs-keyword\">import<\/span> pandas <span class=\"hljs-keyword\">as<\/span> pd\n<span class=\"hljs-keyword\">import<\/span> matplotlib.pyplot <span class=\"hljs-keyword\">as<\/span> plt\n<span class=\"hljs-keyword\">import<\/span> seaborn <span class=\"hljs-keyword\">as<\/span> sns\n%matplotlib inline\n<\/code><\/pre>\n<p>\u0645\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u0645 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u06a9\u0646\u06cc\u0645 <code>read_csv()<\/code> \u0631\u0648\u0634 \u0627\u0632 <code>pandas<\/code> \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 \u0628\u0647 import \u0641\u0627\u06cc\u0644 CSV \u06a9\u0647 \u0634\u0627\u0645\u0644 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627 \u0627\u0633\u062a.<\/p>\n<pre><code class=\"hljs\">dataset = pd.read_csv(<span class=\"hljs-string\">r'E:Datasets\\customer_data.csv'<\/span>)\n<\/code><\/pre>\n<p>\u0627\u062c\u0627\u0632\u0647 \u062f\u0647\u06cc\u062f print \u0634\u06a9\u0644 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627:<\/p>\n<pre><code class=\"hljs\">dataset.shape\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">(10000, 14)\n<\/code><\/pre>\n<p>\u062e\u0631\u0648\u062c\u06cc \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u06a9\u0647 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u062f\u0627\u0631\u0627\u06cc 10 \u0647\u0632\u0627\u0631 \u0631\u06a9\u0648\u0631\u062f \u0648 14 \u0633\u062a\u0648\u0646 \u0627\u0633\u062a.<\/p>\n<p>\u0645\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u0645 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u06a9\u0646\u06cc\u0645 <code>head()<\/code> \u0631\u0648\u0634 \u0642\u0627\u0644\u0628 \u062f\u0627\u062f\u0647 \u067e\u0627\u0646\u062f\u0627\u0647\u0627 \u0628\u0647 print \u067e\u0646\u062c \u0631\u062f\u06cc\u0641 \u0627\u0648\u0644 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627.<\/p>\n<pre><code class=\"hljs\">dataset.head()\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<p><img decoding=\"async\" src=\"https:\/\/rasanegar.com\/blog\/wp-content\/uploads\/2024\/01\/Introduction-to-Pytorch-for-Classification-1.PNG\" alt=\"\u067e\u0646\u062c \u0631\u062f\u06cc\u0641 \u0627\u0648\u0644 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647\" title=\"\"><\/p>\n<p>\u0634\u0645\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f 14 \u0633\u062a\u0648\u0646 \u0631\u0627 \u062f\u0631 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627 \u0628\u0628\u06cc\u0646\u06cc\u062f.  \u0645\u0633\u062a\u0642\u0631 \u0631\u0648\u06cc \u062f\u0631 13 \u0633\u062a\u0648\u0646 \u0627\u0648\u0644\u060c \u0648\u0638\u06cc\u0641\u0647 \u0645\u0627 \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc \u0645\u0642\u062f\u0627\u0631 \u0628\u0631\u0627\u06cc \u0633\u062a\u0648\u0646 14 \u0627\u0633\u062a <code>Exited<\/code>.  \u0644\u0627\u0632\u0645 \u0628\u0647 \u0630\u06a9\u0631 \u0627\u0633\u062a \u06a9\u0647 \u0645\u0642\u0627\u062f\u06cc\u0631 13 \u0633\u062a\u0648\u0646 \u0627\u0648\u0644 6 \u0645\u0627\u0647 \u0642\u0628\u0644 \u0627\u0632 \u0645\u0642\u062f\u0627\u0631 \u0628\u0631\u0627\u06cc <code>Exited<\/code> \u0633\u062a\u0648\u0646 \u0628\u0647 \u062f\u0633\u062a \u0622\u0645\u062f \u0632\u06cc\u0631\u0627 \u0648\u0638\u06cc\u0641\u0647 \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc \u0631\u06cc\u0632\u0634 \u0645\u0634\u062a\u0631\u06cc \u067e\u0633 \u0627\u0632 6 \u0645\u0627\u0647 \u0627\u0632 \u0632\u0645\u0627\u0646 \u062b\u0628\u062a \u0627\u0637\u0644\u0627\u0639\u0627\u062a \u0645\u0634\u062a\u0631\u06cc \u0627\u0633\u062a.<\/p>\n<h2 id=\"exploratorydataanalysis\"><span class=\"ez-toc-section\" id=\"%d8%aa%d8%ac%d8%b2%db%8c%d9%87_%d9%88_%d8%aa%d8%ad%d9%84%db%8c%d9%84_%d8%af%d8%a7%d8%af%d9%87_%d9%87%d8%a7%db%8c_%d8%a7%da%a9%d8%aa%d8%b4%d8%a7%d9%81%db%8c\"><\/span>\u062a\u062c\u0632\u06cc\u0647 \u0648 \u062a\u062d\u0644\u06cc\u0644 \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u0627\u06a9\u062a\u0634\u0627\u0641\u06cc<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>\u0628\u06cc\u0627\u06cc\u06cc\u062f \u062a\u062c\u0632\u06cc\u0647 \u0648 \u062a\u062d\u0644\u06cc\u0644 \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u0627\u06a9\u062a\u0634\u0627\u0641\u06cc \u0631\u0627 \u0627\u0646\u062c\u0627\u0645 \u062f\u0647\u06cc\u0645 \u0631\u0648\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627  \u0645\u0627 \u0627\u0628\u062a\u062f\u0627 \u0646\u0633\u0628\u062a \u0645\u0634\u062a\u0631\u06cc \u0631\u0627 \u06a9\u0647 \u0648\u0627\u0642\u0639\u0627\u064b \u067e\u0633 \u0627\u0632 6 \u0645\u0627\u0647 \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u06a9\u0631\u062f\u0647 \u0627\u0633\u062a\u060c \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc \u0645\u06cc \u06a9\u0646\u06cc\u0645 \u0648 \u0627\u0632 \u0646\u0645\u0648\u062f\u0627\u0631 \u067e\u0627\u06cc \u0628\u0631\u0627\u06cc \u062a\u062c\u0633\u0645 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc \u06a9\u0646\u06cc\u0645.<\/p>\n<p>\u0628\u06cc\u0627\u06cc\u06cc\u062f \u0627\u0628\u062a\u062f\u0627 \u0627\u0646\u062f\u0627\u0632\u0647 \u0646\u0645\u0648\u062f\u0627\u0631\u0647\u0627\u06cc \u067e\u06cc\u0634 \u0641\u0631\u0636 \u0631\u0627 \u0627\u0641\u0632\u0627\u06cc\u0634 \u062f\u0647\u06cc\u0645:<\/p>\n<pre><code class=\"hljs\">fig_size = plt.rcParams(<span class=\"hljs-string\">\"figure.figsize\"<\/span>)\nfig_size(<span class=\"hljs-number\">0<\/span>) = <span class=\"hljs-number\">10<\/span>\nfig_size(<span class=\"hljs-number\">1<\/span>) = <span class=\"hljs-number\">8<\/span>\nplt.rcParams(<span class=\"hljs-string\">\"figure.figsize\"<\/span>) = fig_size\n<\/code><\/pre>\n<p>\u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0637\u0631\u062d \u062f\u0627\u06cc\u0631\u0647 \u0627\u06cc \u0631\u0627 \u062a\u0631\u0633\u06cc\u0645 \u0645\u06cc \u06a9\u0646\u062f <code>Exited<\/code> \u0633\u062a\u0648\u0646<\/p>\n<pre><code class=\"hljs\">dataset.Exited.value_counts().plot(kind=<span class=\"hljs-string\">'pie'<\/span>, autopct=<span class=\"hljs-string\">'%1.0f%%'<\/span>, colors=(<span class=\"hljs-string\">'skyblue'<\/span>, <span class=\"hljs-string\">'orange'<\/span>), explode=(<span class=\"hljs-number\">0.05<\/span>, <span class=\"hljs-number\">0.05<\/span>))\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<p><img decoding=\"async\" src=\"https:\/\/rasanegar.com\/blog\/wp-content\/uploads\/2024\/01\/Introduction-to-Pytorch-for-Classification-2.PNG\" alt=\"\u0631\u0633\u0645 \u0646\u0645\u0648\u062f\u0627\u0631 \u067e\u0627\u06cc \u0628\u0631\u0627\u06cc \u0633\u062a\u0648\u0646 \u062e\u0631\u0648\u062c\u06cc\" title=\"\"><\/p>\n<p>\u062e\u0631\u0648\u062c\u06cc \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u06a9\u0647 \u062f\u0631 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627\u060c 20\u066a \u0627\u0632 \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u06a9\u0631\u062f\u0646\u062f.  \u062f\u0631 \u0627\u06cc\u0646\u062c\u0627 1 \u0645\u062a\u0639\u0644\u0642 \u0628\u0647 \u062d\u0627\u0644\u062a\u06cc \u0627\u0633\u062a \u06a9\u0647 \u0645\u0634\u062a\u0631\u06cc \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u06a9\u0631\u062f\u0647 \u0627\u0633\u062a\u060c \u062c\u0627\u06cc\u06cc \u06a9\u0647 0 \u0628\u0647 \u0633\u0646\u0627\u0631\u06cc\u0648\u06cc\u06cc \u0627\u0634\u0627\u0631\u0647 \u062f\u0627\u0631\u062f \u06a9\u0647 \u0645\u0634\u062a\u0631\u06cc \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u0646\u06a9\u0631\u062f\u0647 \u0627\u0633\u062a.<\/p>\n<p>\u0628\u06cc\u0627\u06cc\u06cc\u062f \u062a\u0639\u062f\u0627\u062f \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0631\u0627 \u0627\u0632 \u0647\u0645\u0647 \u0645\u06a9\u0627\u0646\u200c\u0647\u0627\u06cc \u062c\u063a\u0631\u0627\u0641\u06cc\u0627\u06cc\u06cc \u062f\u0631 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u062a\u0631\u0633\u06cc\u0645 \u06a9\u0646\u06cc\u0645:<\/p>\n<pre><code class=\"hljs\">sns.countplot(x=<span class=\"hljs-string\">'Geography'<\/span>, data=dataset)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<p><img decoding=\"async\" src=\"https:\/\/rasanegar.com\/blog\/wp-content\/uploads\/2024\/01\/Introduction-to-Pytorch-for-Classification-4.PNG\" alt=\"\u062a\u0631\u0633\u06cc\u0645 \u062a\u0639\u062f\u0627\u062f \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0627\u0632 \u0647\u0645\u0647 \u0645\u06a9\u0627\u0646 \u0647\u0627\" title=\"\"><\/p>\n<p>\u062e\u0631\u0648\u062c\u06cc \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u06a9\u0647 \u062a\u0642\u0631\u06cc\u0628\u0627 \u0646\u06cc\u0645\u06cc \u0627\u0632 \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0645\u062a\u0639\u0644\u0642 \u0628\u0647 \u0641\u0631\u0627\u0646\u0633\u0647 \u0647\u0633\u062a\u0646\u062f\u060c \u062f\u0631 \u062d\u0627\u0644\u06cc \u06a9\u0647 \u0646\u0633\u0628\u062a \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0645\u062a\u0639\u0644\u0642 \u0628\u0647 \u0627\u0633\u067e\u0627\u0646\u06cc\u0627 \u0648 \u0622\u0644\u0645\u0627\u0646 \u0647\u0631 \u06a9\u062f\u0627\u0645 25 \u062f\u0631\u0635\u062f \u0627\u0633\u062a.<\/p>\n<p>\u0627\u06a9\u0646\u0648\u0646 \u0628\u06cc\u0627\u06cc\u06cc\u062f \u062a\u0639\u062f\u0627\u062f \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0627\u0632 \u0647\u0631 \u0645\u06a9\u0627\u0646 \u062c\u063a\u0631\u0627\u0641\u06cc\u0627\u06cc\u06cc \u0645\u0646\u062d\u0635\u0631 \u0628\u0647 \u0641\u0631\u062f \u0631\u0627 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 \u0627\u0637\u0644\u0627\u0639\u0627\u062a \u0631\u06cc\u0632\u0634 \u0645\u0634\u062a\u0631\u06cc \u062a\u0631\u0633\u06cc\u0645 \u06a9\u0646\u06cc\u0645.  \u0645\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u0645 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u06a9\u0646\u06cc\u0645 <code>countplot()<\/code> \u062a\u0627\u0628\u0639 \u0627\u0632 <code>seaborn<\/code> \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 \u0628\u0631\u0627\u06cc \u0627\u0646\u062c\u0627\u0645 \u0627\u06cc\u0646 \u06a9\u0627\u0631<\/p>\n<pre><code class=\"hljs\">sns.countplot(x=<span class=\"hljs-string\">'Exited'<\/span>, hue=<span class=\"hljs-string\">'Geography'<\/span>, data=dataset)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<p><img decoding=\"async\" src=\"https:\/\/rasanegar.com\/blog\/wp-content\/uploads\/2024\/01\/Introduction-to-Pytorch-for-Classification-3.PNG\" alt=\"\u062a\u0631\u0633\u06cc\u0645 \u062a\u0639\u062f\u0627\u062f \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0627\u0632 \u0645\u06a9\u0627\u0646 \u0647\u0627\u06cc \u0645\u0646\u062d\u0635\u0631 \u0628\u0647 \u0641\u0631\u062f\" title=\"\"><\/p>\n<p>\u062e\u0631\u0648\u062c\u06cc \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u06a9\u0647 \u0627\u06af\u0631\u0686\u0647 \u062a\u0639\u062f\u0627\u062f \u06a9\u0644\u06cc \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0641\u0631\u0627\u0646\u0633\u0648\u06cc \u062f\u0648 \u0628\u0631\u0627\u0628\u0631 \u062a\u0639\u062f\u0627\u062f \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0627\u0633\u067e\u0627\u0646\u06cc\u0627\u06cc\u06cc \u0648 \u0622\u0644\u0645\u0627\u0646\u06cc \u0627\u0633\u062a\u060c \u0646\u0633\u0628\u062a \u0645\u0634\u062a\u0631\u06cc\u0627\u0646\u06cc \u06a9\u0647 \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u06a9\u0631\u062f\u0647 \u0627\u0646\u062f \u0628\u0631\u0627\u06cc \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0641\u0631\u0627\u0646\u0633\u0648\u06cc \u0648 \u0622\u0644\u0645\u0627\u0646\u06cc \u06cc\u06a9\u0633\u0627\u0646 \u0627\u0633\u062a.  \u0628\u0647 \u0647\u0645\u06cc\u0646 \u062a\u0631\u062a\u06cc\u0628\u060c \u062a\u0639\u062f\u0627\u062f \u06a9\u0644\u06cc \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0622\u0644\u0645\u0627\u0646\u06cc \u0648 \u0627\u0633\u067e\u0627\u0646\u06cc\u0627\u06cc\u06cc \u06cc\u06a9\u0633\u0627\u0646 \u0627\u0633\u062a\u060c \u0627\u0645\u0627 \u062a\u0639\u062f\u0627\u062f \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0622\u0644\u0645\u0627\u0646\u06cc \u06a9\u0647 \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u06a9\u0631\u062f\u0647 \u0627\u0646\u062f\u060c \u062f\u0648 \u0628\u0631\u0627\u0628\u0631 \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0627\u0633\u067e\u0627\u0646\u06cc\u0627\u06cc\u06cc \u0627\u0633\u062a \u06a9\u0647 \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0622\u0644\u0645\u0627\u0646\u06cc \u067e\u0633 \u0627\u0632 6 \u0645\u0627\u0647 \u0628\u06cc\u0634\u062a\u0631 \u0627\u062d\u062a\u0645\u0627\u0644 \u062f\u0627\u0631\u062f \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u06a9\u0646\u0646\u062f.<\/p>\n<p>\u062f\u0631 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647\u060c \u0645\u0627 \u0627\u0637\u0644\u0627\u0639\u0627\u062a \u0645\u0631\u0628\u0648\u0637 \u0628\u0647 \u0628\u0642\u06cc\u0647 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u062e\u0648\u062f \u0631\u0627 \u0628\u0647 \u0635\u0648\u0631\u062a \u0628\u0635\u0631\u06cc \u0631\u0633\u0645 \u0646\u0645\u06cc \u06a9\u0646\u06cc\u0645\u060c \u0627\u0645\u0627 \u0627\u06af\u0631 \u0645\u06cc \u062e\u0648\u0627\u0647\u06cc\u062f \u0627\u06cc\u0646 \u06a9\u0627\u0631 \u0631\u0627 \u0627\u0646\u062c\u0627\u0645 \u062f\u0647\u06cc\u062f\u060c \u0645\u0642\u0627\u0644\u0647 \u0645\u0646 \u0631\u0627 \u0628\u0631\u0631\u0633\u06cc \u06a9\u0646\u06cc\u062f. \u0631\u0648\u06cc \u0631\u0648\u0634 \u0627\u0646\u062c\u0627\u0645 \u062a\u062c\u0632\u06cc\u0647 \u0648 \u062a\u062d\u0644\u06cc\u0644 \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u0627\u06a9\u062a\u0634\u0627\u0641\u06cc \u0628\u0627 Python Seaborn Library.<\/p>\n<h2 id=\"datapreprocessing\"><span class=\"ez-toc-section\" id=\"%d9%be%db%8c%d8%b4_%d9%be%d8%b1%d8%af%d8%a7%d8%b2%d8%b4_%d8%af%d8%a7%d8%af%d9%87_%d9%87%d8%a7\"><\/span>\u067e\u06cc\u0634 \u067e\u0631\u062f\u0627\u0632\u0634 \u062f\u0627\u062f\u0647 \u0647\u0627<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>\u0642\u0628\u0644 \u0627\u0632 \u0627\u06cc\u0646\u06a9\u0647 \u0645\u062f\u0644 PyTorch \u062e\u0648\u062f \u0631\u0627 \u0622\u0645\u0648\u0632\u0634 \u062f\u0647\u06cc\u0645\u060c \u0628\u0627\u06cc\u062f \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u062e\u0648\u062f \u0631\u0627 \u0627\u0632 \u0642\u0628\u0644 \u067e\u0631\u062f\u0627\u0632\u0634 \u06a9\u0646\u06cc\u0645.  \u0627\u06af\u0631 \u0628\u0647 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0646\u06af\u0627\u0647 \u06a9\u0646\u06cc\u062f\u060c \u0645\u06cc \u0628\u06cc\u0646\u06cc\u062f \u06a9\u0647 \u062f\u0648 \u0646\u0648\u0639 \u0633\u062a\u0648\u0646 \u062f\u0627\u0631\u062f: \u0639\u062f\u062f\u06cc \u0648 \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc.  \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc \u062d\u0627\u0648\u06cc \u0627\u0637\u0644\u0627\u0639\u0627\u062a \u0639\u062f\u062f\u06cc \u0647\u0633\u062a\u0646\u062f. <code>CreditScore<\/code>\u060c <code>Balance<\/code>\u060c <code>Age<\/code>\u0648 \u063a\u06cc\u0631\u0647 \u0628\u0647 \u0647\u0645\u06cc\u0646 \u062a\u0631\u062a\u06cc\u0628\u060c <code>Geography<\/code> \u0648 <code>Gender<\/code> \u0633\u062a\u0648\u0646\u200c\u0647\u0627\u06cc \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0647\u0633\u062a\u0646\u062f \u0632\u06cc\u0631\u0627 \u062d\u0627\u0648\u06cc \u0627\u0637\u0644\u0627\u0639\u0627\u062a \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0645\u0627\u0646\u0646\u062f \u0645\u06a9\u0627\u0646 \u0648 \u062c\u0646\u0633\u06cc\u062a \u0645\u0634\u062a\u0631\u06cc\u0627\u0646 \u0647\u0633\u062a\u0646\u062f.  \u0686\u0646\u062f \u0633\u062a\u0648\u0646 \u0648\u062c\u0648\u062f \u062f\u0627\u0631\u062f \u06a9\u0647 \u0645\u06cc \u062a\u0648\u0627\u0646 \u0622\u0646\u0647\u0627 \u0631\u0627 \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0639\u062f\u062f\u06cc \u0648 \u0647\u0645\u0686\u0646\u06cc\u0646 \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u06a9\u0631\u062f.  \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0645\u062b\u0627\u0644\u060c <code>HasCrCard<\/code> \u0633\u062a\u0648\u0646 \u0645\u06cc \u062a\u0648\u0627\u0646\u062f 1 \u06cc\u0627 0 \u0631\u0627 \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0645\u0642\u0627\u062f\u06cc\u0631 \u062e\u0648\u062f \u062f\u0627\u0634\u062a\u0647 \u0628\u0627\u0634\u062f.  \u0628\u0627 \u0627\u06cc\u0646 \u062d\u0627\u0644 <code>HasCrCard<\/code> \u0633\u062a\u0648\u0646 \u062d\u0627\u0648\u06cc \u0627\u0637\u0644\u0627\u0639\u0627\u062a\u06cc \u062f\u0631\u0628\u0627\u0631\u0647 \u062f\u0627\u0634\u062a\u0646 \u06cc\u0627 \u0646\u062f\u0627\u0634\u062a\u0646 \u06a9\u0627\u0631\u062a \u0627\u0639\u062a\u0628\u0627\u0631\u06cc \u0645\u0634\u062a\u0631\u06cc \u0627\u0633\u062a.  \u062a\u0648\u0635\u06cc\u0647 \u0645\u06cc\u200c\u0634\u0648\u062f \u06a9\u0647 \u0633\u062a\u0648\u0646\u200c\u0647\u0627\u06cc\u06cc \u0631\u0627 \u06a9\u0647 \u0645\u06cc\u200c\u062a\u0648\u0627\u0646 \u0647\u0645 \u0628\u0647\u200c\u0639\u0646\u0648\u0627\u0646 \u0645\u0642\u0648\u0644\u0647\u200c\u0627\u06cc \u0648 \u0647\u0645 \u0639\u062f\u062f\u06cc \u062f\u0631 \u0646\u0638\u0631 \u06af\u0631\u0641\u062a\u060c \u0628\u0647\u200c\u0639\u0646\u0648\u0627\u0646 \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u062f\u0631 \u0646\u0638\u0631 \u06af\u0631\u0641\u062a.  \u0628\u0627 \u0627\u06cc\u0646 \u062d\u0627\u0644\u060c \u06a9\u0627\u0645\u0644\u0627\u064b \u0628\u0647 \u062f\u0627\u0646\u0634 \u062f\u0627\u0645\u0646\u0647 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0628\u0633\u062a\u06af\u06cc \u062f\u0627\u0631\u062f.<\/p>\n<p>\u0628\u06cc\u0627\u06cc\u06cc\u062f \u062f\u0648\u0628\u0627\u0631\u0647 print \u062a\u0645\u0627\u0645 \u0633\u062a\u0648\u0646\u200c\u0647\u0627\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627 \u0631\u0627 \u062f\u0631\u06cc\u0627\u0628\u06cc\u0645 \u0648 \u0628\u0641\u0647\u0645\u06cc\u0645 \u06a9\u0647 \u06a9\u062f\u0627\u0645 \u06cc\u06a9 \u0627\u0632 \u0633\u062a\u0648\u0646\u200c\u0647\u0627 \u0631\u0627 \u0645\u06cc\u200c\u062a\u0648\u0627\u0646 \u0639\u062f\u062f\u06cc \u0648 \u06a9\u062f\u0627\u0645 \u0633\u062a\u0648\u0646\u200c\u0647\u0627 \u0631\u0627 \u0628\u0647\u200c\u0639\u0646\u0648\u0627\u0646 \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u06a9\u0631\u062f.  \u0627\u06cc\u0646 <code>columns<\/code> \u0648\u06cc\u0698\u06af\u06cc \u06cc\u06a9 Dataframe \u062a\u0645\u0627\u0645 \u0646\u0627\u0645 \u0633\u062a\u0648\u0646 \u0647\u0627 \u0631\u0627 \u0686\u0627\u067e \u0645\u06cc \u06a9\u0646\u062f:<\/p>\n<pre><code class=\"hljs\">dataset.columns\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">Index(('RowNumber', 'CustomerId', 'Surname', 'CreditScore', 'Geography',\n       'Gender', 'Age', 'Tenure', 'Balance', 'NumOfProducts', 'HasCrCard',\n       'IsActiveMember', 'EstimatedSalary', 'Exited'),\n      dtype='object')\n<\/code><\/pre>\n<p>\u0627\u0632 \u0633\u062a\u0648\u0646\u200c\u0647\u0627\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627\u060c \u0645\u0627 \u0627\u0632 \u0622\u0646 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0646\u062e\u0648\u0627\u0647\u06cc\u0645 \u06a9\u0631\u062f <code>RowNumber<\/code>\u060c <code>CustomerId<\/code>\u060c \u0648 <code>Surname<\/code> \u0633\u062a\u0648\u0646\u200c\u0647\u0627 \u0632\u06cc\u0631\u0627 \u0645\u0642\u0627\u062f\u06cc\u0631 \u0627\u06cc\u0646 \u0633\u062a\u0648\u0646\u200c\u0647\u0627 \u06a9\u0627\u0645\u0644\u0627\u064b \u062a\u0635\u0627\u062f\u0641\u06cc \u0647\u0633\u062a\u0646\u062f \u0648 \u0647\u06cc\u0686 \u0627\u0631\u062a\u0628\u0627\u0637\u06cc \u0628\u0627 \u062e\u0631\u0648\u062c\u06cc \u0646\u062f\u0627\u0631\u0646\u062f.  \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0645\u062b\u0627\u0644\u060c \u0646\u0627\u0645 \u062e\u0627\u0646\u0648\u0627\u062f\u06af\u06cc \u0645\u0634\u062a\u0631\u06cc \u0647\u06cc\u0686 \u062a\u0627\u062b\u06cc\u0631\u06cc \u0646\u062f\u0627\u0631\u062f \u0631\u0648\u06cc \u0622\u06cc\u0627 \u0645\u0634\u062a\u0631\u06cc \u0628\u0627\u0646\u06a9 \u0631\u0627 \u062a\u0631\u06a9 \u062e\u0648\u0627\u0647\u062f \u06a9\u0631\u062f \u06cc\u0627 \u062e\u06cc\u0631.  \u062f\u0631 \u0628\u06cc\u0646 \u0628\u0642\u06cc\u0647 \u0633\u062a\u0648\u0646 \u0647\u0627\u060c <code>Geography<\/code>\u060c <code>Gender<\/code>\u060c <code>HasCrCard<\/code>\u060c \u0648 <code>IsActiveMember<\/code> \u0633\u062a\u0648\u0646 \u0647\u0627 \u0631\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646 \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u06a9\u0631\u062f.  \u0628\u06cc\u0627\u06cc\u06cc\u062f \u0644\u06cc\u0633\u062a\u06cc \u0627\u0632 \u0627\u06cc\u0646 \u0633\u062a\u0648\u0646 \u0647\u0627 \u0627\u06cc\u062c\u0627\u062f \u06a9\u0646\u06cc\u0645:<\/p>\n<pre><code class=\"hljs\">categorical_columns = (<span class=\"hljs-string\">'Geography'<\/span>, <span class=\"hljs-string\">'Gender'<\/span>, <span class=\"hljs-string\">'HasCrCard'<\/span>, <span class=\"hljs-string\">'IsActiveMember'<\/span>)\n<\/code><\/pre>\n<p>\u062a\u0645\u0627\u0645 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0628\u0627\u0642\u06cc \u0645\u0627\u0646\u062f\u0647 \u0628\u0647 \u062c\u0632 <code>Exited<\/code> \u0633\u062a\u0648\u0646 \u0631\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646 \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc \u062f\u0631 \u0646\u0638\u0631 \u06af\u0631\u0641\u062a.<\/p>\n<pre><code class=\"hljs\">numerical_columns = (<span class=\"hljs-string\">'CreditScore'<\/span>, <span class=\"hljs-string\">'Age'<\/span>, <span class=\"hljs-string\">'Tenure'<\/span>, <span class=\"hljs-string\">'Balance'<\/span>, <span class=\"hljs-string\">'NumOfProducts'<\/span>, <span class=\"hljs-string\">'EstimatedSalary'<\/span>)\n<\/code><\/pre>\n<p>\u062f\u0631 \u0646\u0647\u0627\u06cc\u062a\u060c \u062e\u0631\u0648\u062c\u06cc (\u0645\u0642\u0627\u062f\u06cc\u0631 \u0627\u0632 <code>Exited<\/code> \u0633\u062a\u0648\u0646) \u062f\u0631 <code>outputs<\/code> \u0645\u062a\u063a\u06cc\u0631.<\/p>\n<pre><code class=\"hljs\">outputs = (<span class=\"hljs-string\">'Exited'<\/span>)\n<\/code><\/pre>\n<p>\u0645\u0627 \u0644\u06cc\u0633\u062a\u06cc \u0627\u0632 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc\u060c \u0639\u062f\u062f\u06cc \u0648 \u062e\u0631\u0648\u062c\u06cc \u0627\u06cc\u062c\u0627\u062f \u06a9\u0631\u062f\u0647 \u0627\u06cc\u0645.  \u0628\u0627 \u0627\u06cc\u0646 \u062d\u0627\u0644\u060c \u062f\u0631 \u062d\u0627\u0644 \u062d\u0627\u0636\u0631 \u0646\u0648\u0639 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0646\u0634\u062f\u0647 \u0627\u0633\u062a.  \u0628\u0627 \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f \u0646\u0648\u0639 \u062a\u0645\u0627\u0645 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0631\u0627 \u0628\u0631\u0631\u0633\u06cc \u06a9\u0646\u06cc\u062f:<\/p>\n<pre><code class=\"hljs\">dataset.dtypes\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">RowNumber            int64\nCustomerId           int64\nSurname             object\nCreditScore          int64\nGeography           object\nGender              object\nAge                  int64\nTenure               int64\nBalance            float64\nNumOfProducts        int64\nHasCrCard            int64\nIsActiveMember       int64\nEstimatedSalary    float64\nExited               int64\ndtype: object\n<\/code><\/pre>\n<p>\u0634\u0645\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f \u0628\u0628\u06cc\u0646\u06cc\u062f \u06a9\u0647 \u0646\u0648\u0639 \u0628\u0631\u0627\u06cc <code>Geography<\/code> \u0648 <code>Gender<\/code> \u0633\u062a\u0648\u0646 \u0647\u0627 \u0634\u06cc \u0648 \u0646\u0648\u0639 \u0628\u0631\u0627\u06cc \u0627\u0633\u062a <code>HasCrCard<\/code> \u0648 <code>IsActive<\/code> \u0633\u062a\u0648\u0646 int64 \u0627\u0633\u062a.  \u0645\u0627 \u0628\u0627\u06cc\u062f \u0627\u0646\u0648\u0627\u0639 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u0631\u0627 \u0628\u0647 \u062a\u0628\u062f\u06cc\u0644 \u06a9\u0646\u06cc\u0645 <code>category<\/code>.  \u0645\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u0645 \u0627\u06cc\u0646 \u06a9\u0627\u0631 \u0631\u0627 \u0628\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 <code>astype()<\/code> \u0639\u0645\u0644\u06a9\u0631\u062f\u060c \u0647\u0645\u0627\u0646\u0637\u0648\u0631 \u06a9\u0647 \u062f\u0631 \u0632\u06cc\u0631 \u0646\u0634\u0627\u0646 \u062f\u0627\u062f\u0647 \u0634\u062f\u0647 \u0627\u0633\u062a:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-keyword\">for<\/span> category <span class=\"hljs-keyword\">in<\/span> categorical_columns:\n    dataset(category) = dataset(category).astype(<span class=\"hljs-string\">'category'<\/span>)\n<\/code><\/pre>\n<p>\u062d\u0627\u0644 \u0627\u06af\u0631 \u062f\u0648\u0628\u0627\u0631\u0647 \u0627\u0646\u0648\u0627\u0639 \u0633\u062a\u0648\u0646\u200c\u0647\u0627 \u0631\u0627 \u062f\u0631 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627 \u0631\u0633\u0645 \u06a9\u0646\u06cc\u062f\u060c \u0628\u0627\u06cc\u062f \u0646\u062a\u0627\u06cc\u062c \u0632\u06cc\u0631 \u0631\u0627 \u0628\u0628\u06cc\u0646\u06cc\u062f:<\/p>\n<pre><code class=\"hljs\">dataset.dtypes\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc<\/strong><\/p>\n<pre><code class=\"hljs\">RowNumber             int64\nCustomerId            int64\nSurname              object\nCreditScore           int64\nGeography          category\nGender             category\nAge                   int64\nTenure                int64\nBalance             float64\nNumOfProducts         int64\nHasCrCard          category\nIsActiveMember     category\nEstimatedSalary     float64\nExited                int64\ndtype: object\n<\/code><\/pre>\n<p>\u0628\u06cc\u0627\u06cc\u06cc\u062f \u0627\u06a9\u0646\u0648\u0646 \u062a\u0645\u0627\u0645 \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u0647\u0627 \u0631\u0627 \u062f\u0631 \u0642\u0633\u0645\u062a \u0645\u0634\u0627\u0647\u062f\u0647 \u06a9\u0646\u06cc\u0645 <code>Geography<\/code> \u0633\u062a\u0648\u0646:<\/p>\n<pre><code class=\"hljs\">dataset(<span class=\"hljs-string\">'Geography'<\/span>).cat.categories\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">Index(('France', 'Germany', 'Spain'), dtype='object')\n<\/code><\/pre>\n<p>\u0647\u0646\u06af\u0627\u0645\u06cc \u06a9\u0647 \u0646\u0648\u0639 \u062f\u0627\u062f\u0647 \u0633\u062a\u0648\u0646 \u0631\u0627 \u0628\u0647 \u062f\u0633\u062a\u0647 \u062a\u063a\u06cc\u06cc\u0631 \u0645\u06cc \u062f\u0647\u06cc\u062f\u060c \u0628\u0647 \u0647\u0631 \u062f\u0633\u062a\u0647 \u062f\u0631 \u0633\u062a\u0648\u0646 \u06cc\u06a9 \u06a9\u062f \u0645\u0646\u062d\u0635\u0631 \u0628\u0647 \u0641\u0631\u062f \u0627\u062e\u062a\u0635\u0627\u0635 \u0645\u06cc \u06cc\u0627\u0628\u062f.  \u0628\u0631\u0627\u06cc \u0645\u062b\u0627\u0644\u060c \u0628\u06cc\u0627\u06cc\u06cc\u062f \u067e\u0646\u062c \u0631\u062f\u06cc\u0641 \u0627\u0648\u0644 \u0631\u0627 \u0631\u0633\u0645 \u06a9\u0646\u06cc\u0645 <code>Geography<\/code> \u0633\u062a\u0648\u0646 \u0648 print \u0645\u0642\u0627\u062f\u06cc\u0631 \u06a9\u062f \u0628\u0631\u0627\u06cc \u067e\u0646\u062c \u0631\u062f\u06cc\u0641 \u0627\u0648\u0644:<\/p>\n<pre><code class=\"hljs\">dataset(<span class=\"hljs-string\">'Geography'<\/span>).head()\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">0    France\n1     Spain\n2    France\n3    France\n4     Spain\nName: Geography, dtype: category\nCategories (3, object): (France, Germany, Spain)\n<\/code><\/pre>\n<p>\u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u06a9\u062f\u0647\u0627\u06cc \u0645\u0631\u0628\u0648\u0637 \u0628\u0647 \u0645\u0642\u0627\u062f\u06cc\u0631 \u067e\u0646\u062c \u0631\u062f\u06cc\u0641 \u0627\u0648\u0644 \u0631\u0627 \u062a\u0631\u0633\u06cc\u0645 \u0645\u06cc \u06a9\u0646\u062f <code>Geography<\/code> \u0633\u062a\u0648\u0646:<\/p>\n<pre><code class=\"hljs\">dataset(<span class=\"hljs-string\">'Geography'<\/span>).head().cat.codes\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">0    0\n1    2\n2    0\n3    0\n4    2\ndtype: int8\n<\/code><\/pre>\n<p>\u062e\u0631\u0648\u062c\u06cc \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u06a9\u0647 \u0641\u0631\u0627\u0646\u0633\u0647 \u0628\u0627 \u06a9\u062f 0 \u0648 \u0627\u0633\u067e\u0627\u0646\u06cc\u0627 \u06a9\u062f 2 \u0634\u062f\u0647 \u0627\u0633\u062a.<\/p>\n<p>\u0647\u062f\u0641 \u0627\u0635\u0644\u06cc \u0627\u0632 \u062c\u062f\u0627\u0633\u0627\u0632\u06cc \u0633\u062a\u0648\u0646\u200c\u0647\u0627\u06cc \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0627\u0632 \u0633\u062a\u0648\u0646\u200c\u0647\u0627\u06cc \u0639\u062f\u062f\u06cc \u0627\u06cc\u0646 \u0627\u0633\u062a \u06a9\u0647 \u0645\u0642\u0627\u062f\u06cc\u0631 \u0645\u0648\u062c\u0648\u062f \u062f\u0631 \u0633\u062a\u0648\u0646 \u0639\u062f\u062f\u06cc \u0631\u0627 \u0645\u06cc\u200c\u062a\u0648\u0627\u0646 \u0645\u0633\u062a\u0642\u06cc\u0645\u0627\u064b \u0628\u0647 \u0634\u0628\u06a9\u0647\u200c\u0647\u0627\u06cc \u0639\u0635\u0628\u06cc \u062a\u063a\u0630\u06cc\u0647 \u06a9\u0631\u062f.  \u0628\u0627 \u0627\u06cc\u0646 \u062d\u0627\u0644\u060c \u0645\u0642\u0627\u062f\u06cc\u0631 \u0633\u062a\u0648\u0646\u200c\u0647\u0627\u06cc \u062f\u0633\u062a\u0647\u200c\u0628\u0646\u062f\u06cc \u0627\u0628\u062a\u062f\u0627 \u0628\u0627\u06cc\u062f \u0628\u0647 \u0627\u0646\u0648\u0627\u0639 \u0639\u062f\u062f\u06cc \u062a\u0628\u062f\u06cc\u0644 \u0634\u0648\u0646\u062f.  \u06a9\u062f\u06af\u0630\u0627\u0631\u06cc \u0645\u0642\u0627\u062f\u06cc\u0631 \u062f\u0631 \u0633\u062a\u0648\u0646 \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u062a\u0627 \u062d\u062f\u06cc \u0648\u0638\u06cc\u0641\u0647 \u062a\u0628\u062f\u06cc\u0644 \u0639\u062f\u062f\u06cc \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0631\u0627 \u062d\u0644 \u0645\u06cc \u06a9\u0646\u062f.<\/p>\n<p>\u0627\u0632 \u0622\u0646\u062c\u0627\u06cc\u06cc \u06a9\u0647 \u0645\u0627 \u0627\u0632 PyTorch \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u062e\u0648\u0627\u0647\u06cc\u0645 \u06a9\u0631\u062f\u060c \u0628\u0627\u06cc\u062f \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u0648 \u0639\u062f\u062f\u06cc \u062e\u0648\u062f \u0631\u0627 \u0628\u0647 \u062a\u0627\u0646\u0633\u0648\u0631 \u062a\u0628\u062f\u06cc\u0644 \u06a9\u0646\u06cc\u0645.<\/p>\n<p>\u0627\u0628\u062a\u062f\u0627 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u0631\u0627 \u0628\u0647 \u062a\u0627\u0646\u0633\u0648\u0631 \u062a\u0628\u062f\u06cc\u0644 \u0645\u06cc \u06a9\u0646\u06cc\u0645.  \u062f\u0631 PyTorch\u060c \u062a\u0627\u0646\u0633\u0648\u0631\u0647\u0627 \u0631\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646 \u0627\u0632 \u0637\u0631\u06cc\u0642 \u0622\u0631\u0627\u06cc\u0647 \u0647\u0627\u06cc numpy \u0627\u06cc\u062c\u0627\u062f \u06a9\u0631\u062f.  \u0627\u0628\u062a\u062f\u0627 \u062f\u0627\u062f\u0647\u200c\u0647\u0627\u06cc \u0686\u0647\u0627\u0631 \u0633\u062a\u0648\u0646 \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0634\u062f\u0647 \u0631\u0627 \u0628\u0647 \u0622\u0631\u0627\u06cc\u0647\u200c\u0647\u0627\u06cc numpy \u062a\u0628\u062f\u06cc\u0644 \u0645\u06cc\u200c\u06a9\u0646\u06cc\u0645 \u0648 \u0633\u067e\u0633 \u062a\u0645\u0627\u0645 \u0633\u062a\u0648\u0646\u200c\u0647\u0627 \u0631\u0627 \u0628\u0647 \u0635\u0648\u0631\u062a \u0627\u0641\u0642\u06cc \u0631\u0648\u06cc \u0647\u0645 \u0642\u0631\u0627\u0631 \u0645\u06cc\u200c\u062f\u0647\u06cc\u0645\u060c \u0647\u0645\u0627\u0646\u0637\u0648\u0631 \u06a9\u0647 \u062f\u0631 \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0646\u0634\u0627\u0646 \u062f\u0627\u062f\u0647 \u0634\u062f\u0647 \u0627\u0633\u062a:<\/p>\n<pre><code class=\"hljs\">geo = dataset(<span class=\"hljs-string\">'Geography'<\/span>).cat.codes.values\ngen = dataset(<span class=\"hljs-string\">'Gender'<\/span>).cat.codes.values\nhcc = dataset(<span class=\"hljs-string\">'HasCrCard'<\/span>).cat.codes.values\niam = dataset(<span class=\"hljs-string\">'IsActiveMember'<\/span>).cat.codes.values\n\ncategorical_data = np.stack((geo, gen, hcc, iam), <span class=\"hljs-number\">1<\/span>)\n\ncategorical_data(:<span class=\"hljs-number\">10<\/span>)\n<\/code><\/pre>\n<p>\u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0628\u0627\u0644\u0627 \u062f\u0647 \u0631\u06a9\u0648\u0631\u062f \u0627\u0648\u0644 \u0631\u0627 \u0627\u0632 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0634\u062f\u0647\u060c \u0628\u0647 \u0635\u0648\u0631\u062a \u0627\u0641\u0642\u06cc \u0686\u0627\u067e \u0645\u06cc \u06a9\u0646\u062f.  \u062e\u0631\u0648\u062c\u06cc \u0628\u0647 \u0635\u0648\u0631\u062a \u0632\u06cc\u0631 \u0627\u0633\u062a:<\/p>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">array(((0, 0, 1, 1),\n       (2, 0, 0, 1),\n       (0, 0, 1, 0),\n       (0, 0, 0, 0),\n       (2, 0, 1, 1),\n       (2, 1, 1, 0),\n       (0, 1, 1, 1),\n       (1, 0, 1, 0),\n       (0, 1, 0, 1),\n       (0, 1, 1, 1)), dtype=int8)\n<\/code><\/pre>\n<p>\u062d\u0627\u0644 \u0628\u0631\u0627\u06cc \u0627\u06cc\u062c\u0627\u062f \u06cc\u06a9 \u062a\u0627\u0646\u0633\u0648\u0631 \u0627\u0632 \u0622\u0631\u0627\u06cc\u0647 numpy \u0641\u0648\u0642 \u0627\u0644\u0630\u06a9\u0631\u060c \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f \u0628\u0647 \u0633\u0627\u062f\u06af\u06cc \u0622\u0631\u0627\u06cc\u0647 \u0631\u0627 \u0628\u0647 <code>tensor<\/code> \u06a9\u0644\u0627\u0633 \u0627\u0632 <code>torch<\/code> \u0645\u062f\u0648\u0644.  \u0628\u0647 \u06cc\u0627\u062f \u062f\u0627\u0634\u062a\u0647 \u0628\u0627\u0634\u06cc\u062f\u060c \u0628\u0631\u0627\u06cc \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0634\u062f\u0647\u060c \u0646\u0648\u0639 \u062f\u0627\u062f\u0647 \u0628\u0627\u06cc\u062f \u0628\u0627\u0634\u062f <code>torch.int64<\/code>.<\/p>\n<pre><code class=\"hljs\">categorical_data = torch.tensor(categorical_data, dtype=torch.int64)\ncategorical_data(:<span class=\"hljs-number\">10<\/span>)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">tensor(((0, 0, 1, 1),\n        (2, 0, 0, 1),\n        (0, 0, 1, 0),\n        (0, 0, 0, 0),\n        (2, 0, 1, 1),\n        (2, 1, 1, 0),\n        (0, 1, 1, 1),\n        (1, 0, 1, 0),\n        (0, 1, 0, 1),\n        (0, 1, 1, 1)))\n<\/code><\/pre>\n<p>\u062f\u0631 \u062e\u0631\u0648\u062c\u06cc\u060c \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f \u0628\u0628\u06cc\u0646\u06cc\u062f \u06a9\u0647 \u0622\u0631\u0627\u06cc\u0647 numpy \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0634\u062f\u0647 \u0627\u06a9\u0646\u0648\u0646 \u0628\u0647 \u06cc\u06a9 \u062a\u0628\u062f\u06cc\u0644 \u0634\u062f\u0647 \u0627\u0633\u062a. <code>tensor<\/code> \u0647\u062f\u0641 &#8211; \u0634\u06cc.<\/p>\n<p>\u0628\u0647 \u0647\u0645\u06cc\u0646 \u062a\u0631\u062a\u06cc\u0628\u060c \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u0645 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc \u062e\u0648\u062f \u0631\u0627 \u0628\u0647 \u062a\u0627\u0646\u0633\u0648\u0631 \u062a\u0628\u062f\u06cc\u0644 \u06a9\u0646\u06cc\u0645:<\/p>\n<pre><code class=\"hljs\">numerical_data = np.stack((dataset(col).values <span class=\"hljs-keyword\">for<\/span> col <span class=\"hljs-keyword\">in<\/span> numerical_columns), <span class=\"hljs-number\">1<\/span>)\nnumerical_data = torch.tensor(numerical_data, dtype=torch.<span class=\"hljs-built_in\">float<\/span>)\nnumerical_data(:<span class=\"hljs-number\">5<\/span>)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">tensor(((6.1900e+02, 4.2000e+01, 2.0000e+00, 0.0000e+00, 1.0000e+00, 1.0135e+05),\n        (6.0800e+02, 4.1000e+01, 1.0000e+00, 8.3808e+04, 1.0000e+00, 1.1254e+05),\n        (5.0200e+02, 4.2000e+01, 8.0000e+00, 1.5966e+05, 3.0000e+00, 1.1393e+05),\n        (6.9900e+02, 3.9000e+01, 1.0000e+00, 0.0000e+00, 2.0000e+00, 9.3827e+04),\n        (8.5000e+02, 4.3000e+01, 2.0000e+00, 1.2551e+05, 1.0000e+00, 7.9084e+04)))\n<\/code><\/pre>\n<p>\u062f\u0631 \u062e\u0631\u0648\u062c\u06cc\u060c \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f \u067e\u0646\u062c \u0631\u062f\u06cc\u0641 \u0627\u0648\u0644 \u062d\u0627\u0648\u06cc \u0645\u0642\u0627\u062f\u06cc\u0631 \u0634\u0634 \u0633\u062a\u0648\u0646 \u0639\u062f\u062f\u06cc \u062f\u0631 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0645\u0627 \u0631\u0627 \u0628\u0628\u06cc\u0646\u06cc\u062f.<\/p>\n<p>\u0645\u0631\u062d\u0644\u0647 \u0622\u062e\u0631 \u062a\u0628\u062f\u06cc\u0644 \u0622\u0631\u0627\u06cc\u0647 numpy \u062e\u0631\u0648\u062c\u06cc \u0628\u0647 a \u0627\u0633\u062a <code>tensor<\/code> \u0647\u062f\u0641 &#8211; \u0634\u06cc.<\/p>\n<pre><code class=\"hljs\">outputs = torch.tensor(dataset(outputs).values).flatten()\noutputs(:<span class=\"hljs-number\">5<\/span>)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">tensor((1, 0, 1, 0, 0))\n<\/code><\/pre>\n<p>\u0627\u062c\u0627\u0632\u0647 \u062f\u0647\u06cc\u062f \u0634\u06a9\u0644 \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc\u060c \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc \u0648 \u062e\u0631\u0648\u062c\u06cc \u0645\u0631\u0628\u0648\u0637\u0647 \u0631\u0627 \u0631\u0633\u0645 \u06a9\u0646\u06cc\u0645:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-built_in\">print<\/span>(categorical_data.shape)\n<span class=\"hljs-built_in\">print<\/span>(numerical_data.shape)\n<span class=\"hljs-built_in\">print<\/span>(outputs.shape)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">torch.Size((10000, 4))\ntorch.Size((10000, 6))\ntorch.Size((10000))\n<\/code><\/pre>\n<p>\u0642\u0628\u0644 \u0627\u0632 \u0627\u06cc\u0646\u06a9\u0647 \u0628\u062a\u0648\u0627\u0646\u06cc\u0645 \u0645\u062f\u0644 \u062e\u0648\u062f \u0631\u0627 \u0622\u0645\u0648\u0632\u0634 \u062f\u0647\u06cc\u0645 \u06cc\u06a9 \u0645\u0631\u062d\u0644\u0647 \u0628\u0633\u06cc\u0627\u0631 \u0645\u0647\u0645 \u0648\u062c\u0648\u062f \u062f\u0627\u0631\u062f.  \u0645\u0627 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u062e\u0648\u062f \u0631\u0627 \u0628\u0647 \u0639\u062f\u062f\u06cc \u062a\u0628\u062f\u06cc\u0644 \u06a9\u0631\u062f\u06cc\u0645 \u06a9\u0647 \u062f\u0631 \u0622\u0646 \u06cc\u06a9 \u0645\u0642\u062f\u0627\u0631 \u0645\u0646\u062d\u0635\u0631 \u0628\u0647 \u0641\u0631\u062f \u0628\u0627 \u06cc\u06a9 \u0639\u062f\u062f \u0635\u062d\u06cc\u062d \u0646\u0634\u0627\u0646 \u062f\u0627\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f.  \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0645\u062b\u0627\u0644\u060c \u062f\u0631 <code>Geography<\/code> \u062f\u0631 \u0633\u062a\u0648\u0646\u060c \u062f\u06cc\u062f\u06cc\u0645 \u06a9\u0647 \u0641\u0631\u0627\u0646\u0633\u0647 \u0628\u0627 0 \u0648 \u0622\u0644\u0645\u0627\u0646 \u0628\u0627 1 \u0646\u0634\u0627\u0646 \u062f\u0627\u062f\u0647 \u0634\u062f\u0647 \u0627\u0633\u062a. \u0645\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u0645 \u0627\u0632 \u0627\u06cc\u0646 \u0645\u0642\u0627\u062f\u06cc\u0631 \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644 \u062e\u0648\u062f \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u06a9\u0646\u06cc\u0645.  \u0628\u0627 \u0627\u06cc\u0646 \u062d\u0627\u0644\u060c \u0631\u0627\u0647 \u0628\u0647\u062a\u0631 \u0627\u06cc\u0646 \u0627\u0633\u062a \u06a9\u0647 \u0628\u0647 \u062c\u0627\u06cc \u06cc\u06a9 \u0639\u062f\u062f \u0635\u062d\u06cc\u062d\u060c \u0645\u0642\u0627\u062f\u06cc\u0631 \u0631\u0627 \u062f\u0631 \u06cc\u06a9 \u0633\u062a\u0648\u0646 \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0628\u0647 \u0634\u06a9\u0644 \u06cc\u06a9 \u0628\u0631\u062f\u0627\u0631 N \u0628\u0639\u062f\u06cc \u0646\u0634\u0627\u0646 \u062f\u0647\u06cc\u0645.  \u06cc\u06a9 \u0628\u0631\u062f\u0627\u0631 \u0642\u0627\u062f\u0631 \u0628\u0647 \u06af\u0631\u0641\u062a\u0646 \u0627\u0637\u0644\u0627\u0639\u0627\u062a \u0628\u06cc\u0634\u062a\u0631 \u0627\u0633\u062a \u0648 \u0645\u06cc \u062a\u0648\u0627\u0646\u062f \u0631\u0648\u0627\u0628\u0637 \u0628\u06cc\u0646 \u0645\u0642\u0627\u062f\u06cc\u0631 \u0645\u0642\u0648\u0644\u0647 \u0627\u06cc \u0645\u062e\u062a\u0644\u0641 \u0631\u0627 \u0628\u0647 \u0631\u0648\u0634 \u0645\u0646\u0627\u0633\u0628 \u062a\u0631\u06cc \u067e\u06cc\u062f\u0627 \u06a9\u0646\u062f.  \u0628\u0646\u0627\u0628\u0631\u0627\u06cc\u0646\u060c \u0645\u0642\u0627\u062f\u06cc\u0631 \u0631\u0627 \u062f\u0631 \u0633\u062a\u0648\u0646\u200c\u0647\u0627\u06cc \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0628\u0647 \u0634\u06a9\u0644 \u0628\u0631\u062f\u0627\u0631\u0647\u0627\u06cc N \u0628\u0639\u062f\u06cc \u0646\u0634\u0627\u0646 \u062e\u0648\u0627\u0647\u06cc\u0645 \u062f\u0627\u062f.  \u0627\u06cc\u0646 process \u062a\u0639\u0628\u06cc\u0647 \u0646\u0627\u0645\u06cc\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f.<\/p>\n<p>\u0645\u0627 \u0628\u0627\u06cc\u062f \u0627\u0646\u062f\u0627\u0632\u0647 \u062a\u0639\u0628\u06cc\u0647 (\u0627\u0628\u0639\u0627\u062f \u0628\u0631\u062f\u0627\u0631\u06cc) \u0631\u0627 \u0628\u0631\u0627\u06cc \u062a\u0645\u0627\u0645 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u062a\u0639\u0631\u06cc\u0641 \u06a9\u0646\u06cc\u0645.  \u0647\u06cc\u0686 \u0642\u0627\u0646\u0648\u0646 \u0633\u062e\u062a \u0648 \u0633\u0631\u06cc\u0639\u06cc \u062f\u0631 \u0645\u0648\u0631\u062f \u062a\u0639\u062f\u0627\u062f \u0627\u0628\u0639\u0627\u062f \u0648\u062c\u0648\u062f \u0646\u062f\u0627\u0631\u062f.  \u06cc\u06a9 \u0642\u0627\u0646\u0648\u0646 \u0633\u0631\u0627\u0646\u06af\u0634\u062a\u06cc \u062e\u0648\u0628 \u0628\u0631\u0627\u06cc \u062a\u0639\u0631\u06cc\u0641 \u0627\u0646\u062f\u0627\u0632\u0647 \u062c\u0627\u0633\u0627\u0632\u06cc \u0628\u0631\u0627\u06cc \u06cc\u06a9 \u0633\u062a\u0648\u0646\u060c \u062a\u0642\u0633\u06cc\u0645 \u062a\u0639\u062f\u0627\u062f \u0645\u0642\u0627\u062f\u06cc\u0631 \u0645\u0646\u062d\u0635\u0631 \u0628\u0647 \u0641\u0631\u062f \u062f\u0631 \u0633\u062a\u0648\u0646 \u0628\u0631 2 (\u0627\u0645\u0627 \u0646\u0647 \u0628\u06cc\u0634\u062a\u0631 \u0627\u0632 50) \u0627\u0633\u062a.  \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0645\u062b\u0627\u0644\u060c \u0628\u0631\u0627\u06cc <code>Geography<\/code> \u0633\u062a\u0648\u0646\u060c \u062a\u0639\u062f\u0627\u062f \u0645\u0642\u0627\u062f\u06cc\u0631 \u0645\u0646\u062d\u0635\u0631 \u0628\u0647 \u0641\u0631\u062f 3 \u0627\u0633\u062a. \u0627\u0646\u062f\u0627\u0632\u0647 \u062a\u0639\u0628\u06cc\u0647 \u0645\u0631\u0628\u0648\u0637\u0647 \u0628\u0631\u0627\u06cc <code>Geography<\/code> \u0633\u062a\u0648\u0646 3\/2 = 1.5 = 2 (\u062f\u0648\u0631 \u06a9\u0631\u062f\u0646) \u062e\u0648\u0627\u0647\u062f \u0628\u0648\u062f.<\/p>\n<p>\u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u06cc\u06a9 \u062a\u0627\u067e\u0644 \u0627\u06cc\u062c\u0627\u062f \u0645\u06cc \u06a9\u0646\u062f \u06a9\u0647 \u0634\u0627\u0645\u0644 \u062a\u0639\u062f\u0627\u062f \u0645\u0642\u0627\u062f\u06cc\u0631 \u0645\u0646\u062d\u0635\u0631 \u0628\u0647 \u0641\u0631\u062f \u0648 \u0627\u0646\u062f\u0627\u0632\u0647 \u0627\u0628\u0639\u0627\u062f \u0628\u0631\u0627\u06cc \u062a\u0645\u0627\u0645 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0645\u06cc \u0634\u0648\u062f:<\/p>\n<pre><code class=\"hljs\">categorical_column_sizes = (<span class=\"hljs-built_in\">len<\/span>(dataset(column).cat.categories) <span class=\"hljs-keyword\">for<\/span> column <span class=\"hljs-keyword\">in<\/span> categorical_columns)\ncategorical_embedding_sizes = ((col_size, <span class=\"hljs-built_in\">min<\/span>(<span class=\"hljs-number\">50<\/span>, (col_size+<span class=\"hljs-number\">1<\/span>)\/\/<span class=\"hljs-number\">2<\/span>)) <span class=\"hljs-keyword\">for<\/span> col_size <span class=\"hljs-keyword\">in<\/span> categorical_column_sizes)\n<span class=\"hljs-built_in\">print<\/span>(categorical_embedding_sizes)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">((3, 2), (2, 1), (2, 1), (2, 1))\n<\/code><\/pre>\n<p>\u06cc\u06a9 \u0645\u062f\u0644 \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u0639\u0645\u06cc\u0642 \u062a\u062d\u062a \u0646\u0638\u0627\u0631\u062a\u060c \u0645\u0627\u0646\u0646\u062f \u0645\u062f\u0644\u06cc \u06a9\u0647 \u062f\u0631 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u062a\u0648\u0633\u0639\u0647 \u0645\u06cc\u200c\u062f\u0647\u06cc\u0645\u060c \u0628\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u062f\u0627\u062f\u0647\u200c\u0647\u0627\u06cc \u0622\u0645\u0648\u0632\u0634\u06cc \u0622\u0645\u0648\u0632\u0634 \u062f\u0627\u062f\u0647 \u0645\u06cc\u200c\u0634\u0648\u062f \u0648 \u0639\u0645\u0644\u06a9\u0631\u062f \u0645\u062f\u0644 \u0627\u0631\u0632\u06cc\u0627\u0628\u06cc \u0645\u06cc\u200c\u0634\u0648\u062f. \u0631\u0648\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0622\u0632\u0645\u0627\u06cc\u0634\u06cc  \u0628\u0646\u0627\u0628\u0631\u0627\u06cc\u0646\u060c \u0645\u0627 \u0628\u0627\u06cc\u062f \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u062e\u0648\u062f \u0631\u0627 \u0628\u0647 \u0645\u062c\u0645\u0648\u0639\u0647 \u0647\u0627\u06cc \u0622\u0645\u0648\u0632\u0634\u06cc \u0648 \u0622\u0632\u0645\u0627\u06cc\u0634\u06cc \u062a\u0642\u0633\u06cc\u0645 \u06a9\u0646\u06cc\u0645\u060c \u0647\u0645\u0627\u0646\u0637\u0648\u0631 \u06a9\u0647 \u062f\u0631 \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0646\u0634\u0627\u0646 \u062f\u0627\u062f\u0647 \u0634\u062f\u0647 \u0627\u0633\u062a:<\/p>\n<pre><code class=\"hljs\">total_records = <span class=\"hljs-number\">10000<\/span>\ntest_records = <span class=\"hljs-built_in\">int<\/span>(total_records * <span class=\"hljs-number\">.2<\/span>)\n\ncategorical_train_data = categorical_data(:total_records-test_records)\ncategorical_test_data = categorical_data(total_records-test_records:total_records)\nnumerical_train_data = numerical_data(:total_records-test_records)\nnumerical_test_data = numerical_data(total_records-test_records:total_records)\ntrain_outputs = outputs(:total_records-test_records)\ntest_outputs = outputs(total_records-test_records:total_records)\n<\/code><\/pre>\n<p>\u0645\u0627 10 \u0647\u0632\u0627\u0631 \u0631\u06a9\u0648\u0631\u062f \u062f\u0631 \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u062e\u0648\u062f \u062f\u0627\u0631\u06cc\u0645 \u06a9\u0647 80 \u062f\u0631\u0635\u062f \u0631\u06a9\u0648\u0631\u062f\u0647\u0627 \u06cc\u0639\u0646\u06cc 8000 \u0631\u06a9\u0648\u0631\u062f \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f \u062f\u0631 \u062d\u0627\u0644\u06cc \u06a9\u0647 20 \u062f\u0631\u0635\u062f \u0631\u06a9\u0648\u0631\u062f\u0647\u0627\u06cc \u0628\u0627\u0642\u06cc \u0645\u0627\u0646\u062f\u0647 \u0628\u0631\u0627\u06cc \u0627\u0631\u0632\u06cc\u0627\u0628\u06cc \u0639\u0645\u0644\u06a9\u0631\u062f \u0645\u062f\u0644 \u0645\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f.  \u062a\u0648\u062c\u0647 \u06a9\u0646\u06cc\u062f\u060c \u062f\u0631 \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0628\u0627\u0644\u0627\u060c \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u0648 \u0639\u062f\u062f\u06cc \u0648 \u0647\u0645\u0686\u0646\u06cc\u0646 \u062e\u0631\u0648\u062c\u06cc \u0647\u0627 \u0628\u0647 \u0645\u062c\u0645\u0648\u0639\u0647 \u0647\u0627\u06cc \u0622\u0645\u0648\u0632\u0634\u06cc \u0648 \u0622\u0632\u0645\u0627\u06cc\u0634\u06cc \u062a\u0642\u0633\u06cc\u0645 \u0634\u062f\u0647 \u0627\u0646\u062f.<\/p>\n<p>\u0628\u0631\u0627\u06cc \u062a\u0623\u06cc\u06cc\u062f \u0627\u06cc\u0646\u06a9\u0647 \u062f\u0627\u062f\u0647 \u0647\u0627 \u0631\u0627 \u0628\u0647 \u062f\u0631\u0633\u062a\u06cc \u0628\u0647 \u0645\u062c\u0645\u0648\u0639\u0647 \u0647\u0627\u06cc \u0622\u0645\u0648\u0632\u0634\u06cc \u0648 \u0622\u0632\u0645\u0627\u06cc\u0634\u06cc \u062a\u0642\u0633\u06cc\u0645 \u06a9\u0631\u062f\u0647 \u0627\u06cc\u0645\u060c \u0627\u062c\u0627\u0632\u0647 \u062f\u0647\u06cc\u062f print \u0645\u062f\u062a \u0632\u0645\u0627\u0646 \u0633\u0648\u0627\u0628\u0642 \u0622\u0645\u0648\u0632\u0634 \u0648 \u0622\u0632\u0645\u0648\u0646:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-built_in\">len<\/span>(categorical_train_data))\n<span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-built_in\">len<\/span>(numerical_train_data))\n<span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-built_in\">len<\/span>(train_outputs))\n\n<span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-built_in\">len<\/span>(categorical_test_data))\n<span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-built_in\">len<\/span>(numerical_test_data))\n<span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-built_in\">len<\/span>(test_outputs))\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">8000\n8000\n8000\n2000\n2000\n2000\n<\/code><\/pre>\n<h2 id=\"creatingamodelforprediction\"><span class=\"ez-toc-section\" id=\"%d8%a7%db%8c%d8%ac%d8%a7%d8%af_%db%8c%da%a9_%d9%85%d8%af%d9%84_%d8%a8%d8%b1%d8%a7%db%8c_%d9%be%db%8c%d8%b4_%d8%a8%db%8c%d9%86%db%8c\"><\/span>\u0627\u06cc\u062c\u0627\u062f \u06cc\u06a9 \u0645\u062f\u0644 \u0628\u0631\u0627\u06cc \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>\u0645\u0627 \u062f\u0627\u062f\u0647 \u0647\u0627 \u0631\u0627 \u0628\u0647 \u0645\u062c\u0645\u0648\u0639\u0647 \u0647\u0627\u06cc \u0622\u0645\u0648\u0632\u0634\u06cc \u0648 \u0622\u0632\u0645\u0627\u06cc\u0634\u06cc \u062a\u0642\u0633\u06cc\u0645 \u06a9\u0631\u062f\u0647 \u0627\u06cc\u0645\u060c \u0627\u06a9\u0646\u0648\u0646 \u0632\u0645\u0627\u0646 \u0622\u0646 \u0627\u0633\u062a \u06a9\u0647 \u0645\u062f\u0644 \u062e\u0648\u062f \u0631\u0627 \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u062a\u0639\u0631\u06cc\u0641 \u06a9\u0646\u06cc\u0645.  \u0628\u0631\u0627\u06cc \u0627\u0646\u062c\u0627\u0645 \u0627\u06cc\u0646 \u06a9\u0627\u0631\u060c \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u0645 \u06cc\u06a9 \u06a9\u0644\u0627\u0633 \u0628\u0647 \u0646\u0627\u0645 \u062a\u0639\u0631\u06cc\u0641 \u06a9\u0646\u06cc\u0645 <code>Model<\/code>\u060c \u06a9\u0647 \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u062e\u0648\u0627\u0647\u062f \u0634\u062f.  \u0628\u0647 \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0646\u06af\u0627\u0647 \u06a9\u0646\u06cc\u062f:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-class\"><span class=\"hljs-keyword\">class<\/span> <span class=\"hljs-title\">Model<\/span>(<span class=\"hljs-params\">nn.Module<\/span>):<\/span>\n\n    <span class=\"hljs-function\"><span class=\"hljs-keyword\">def<\/span> <span class=\"hljs-title\">__init__<\/span>(<span class=\"hljs-params\">self, embedding_size, num_numerical_cols, output_size, layers, p=<span class=\"hljs-number\">0.4<\/span><\/span>):<\/span>\n        <span class=\"hljs-built_in\">super<\/span>().__init__()\n        self.all_embeddings = nn.ModuleList((nn.Embedding(ni, nf) <span class=\"hljs-keyword\">for<\/span> ni, nf <span class=\"hljs-keyword\">in<\/span> embedding_size))\n        self.embedding_dropout = nn.Dropout(p)\n        self.batch_norm_num = nn.BatchNorm1d(num_numerical_cols)\n\n        all_layers = ()\n        num_categorical_cols = <span class=\"hljs-built_in\">sum<\/span>((nf <span class=\"hljs-keyword\">for<\/span> ni, nf <span class=\"hljs-keyword\">in<\/span> embedding_size))\n        input_size = num_categorical_cols + num_numerical_cols\n\n        <span class=\"hljs-keyword\">for<\/span> i <span class=\"hljs-keyword\">in<\/span> layers:\n            all_layers.append(nn.Linear(input_size, i))\n            all_layers.append(nn.ReLU(inplace=<span class=\"hljs-literal\">True<\/span>))\n            all_layers.append(nn.BatchNorm1d(i))\n            all_layers.append(nn.Dropout(p))\n            input_size = i\n\n        all_layers.append(nn.Linear(layers(-<span class=\"hljs-number\">1<\/span>), output_size))\n\n        self.layers = nn.Sequential(*all_layers)\n\n    <span class=\"hljs-function\"><span class=\"hljs-keyword\">def<\/span> <span class=\"hljs-title\">forward<\/span>(<span class=\"hljs-params\">self, x_categorical, x_numerical<\/span>):<\/span>\n        embeddings = ()\n        <span class=\"hljs-keyword\">for<\/span> i,e <span class=\"hljs-keyword\">in<\/span> <span class=\"hljs-built_in\">enumerate<\/span>(self.all_embeddings):\n            embeddings.append(e(x_categorical(:,i)))\n        x = torch.cat(embeddings, <span class=\"hljs-number\">1<\/span>)\n        x = self.embedding_dropout(x)\n\n        x_numerical = self.batch_norm_num(x_numerical)\n        x = torch.cat((x, x_numerical), <span class=\"hljs-number\">1<\/span>)\n        x = self.layers(x)\n        <span class=\"hljs-keyword\">return<\/span> x\n<\/code><\/pre>\n<p>\u0627\u06af\u0631 \u062a\u0627 \u0628\u0647 \u062d\u0627\u0644 \u0628\u0627 PyTorch \u06a9\u0627\u0631 \u0646\u06a9\u0631\u062f\u0647 \u0627\u06cc\u062f\u060c \u06a9\u062f \u0628\u0627\u0644\u0627 \u0645\u0645\u06a9\u0646 \u0627\u0633\u062a \u062a\u0631\u0633\u0646\u0627\u06a9 \u0628\u0647 \u0646\u0638\u0631 \u0628\u0631\u0633\u062f\u060c \u0628\u0627 \u0627\u06cc\u0646 \u062d\u0627\u0644 \u0633\u0639\u06cc \u0645\u06cc \u06a9\u0646\u0645 \u0622\u0646 \u0631\u0627 \u0628\u0631\u0627\u06cc \u0634\u0645\u0627 \u062a\u062c\u0632\u06cc\u0647 \u06a9\u0646\u0645.<\/p>\n<p>\u062f\u0631 \u062e\u0637 \u0627\u0648\u0644\u060c a \u0631\u0627 \u0627\u0639\u0644\u0627\u0645 \u0645\u06cc \u06a9\u0646\u06cc\u0645 <code>Model<\/code> \u06a9\u0644\u0627\u0633\u06cc \u06a9\u0647 \u0627\u0632 the \u0628\u0647 \u0627\u0631\u062b \u0645\u06cc \u0628\u0631\u062f <code>Module<\/code> \u06a9\u0644\u0627\u0633 \u0627\u0632 PyTorch&#8217;s <code>nn<\/code> \u0645\u062f\u0648\u0644.  \u062f\u0631 \u0633\u0627\u0632\u0646\u062f\u0647 \u06a9\u0644\u0627\u0633 (the <code>__init__()<\/code> \u0631\u0648\u0634) \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u0647\u0627\u06cc \u0632\u06cc\u0631 \u0627\u0631\u0633\u0627\u0644 \u0645\u06cc \u0634\u0648\u062f:<\/p>\n<ol>\n<li><code>embedding_size<\/code>: \u0634\u0627\u0645\u0644 \u0627\u0646\u062f\u0627\u0632\u0647 \u062c\u0627\u0633\u0627\u0632\u06cc \u0628\u0631\u0627\u06cc \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u0634\u062f\u0647 \u0627\u0633\u062a<\/li>\n<li><code>num_numerical_cols<\/code>: \u062a\u0639\u062f\u0627\u062f \u06a9\u0644 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc \u0631\u0627 \u0630\u062e\u06cc\u0631\u0647 \u0645\u06cc \u06a9\u0646\u062f<\/li>\n<li><code>output_size<\/code>: \u0627\u0646\u062f\u0627\u0632\u0647 \u0644\u0627\u06cc\u0647 \u062e\u0631\u0648\u062c\u06cc \u06cc\u0627 \u062a\u0639\u062f\u0627\u062f \u062e\u0631\u0648\u062c\u06cc \u0647\u0627\u06cc \u0645\u0645\u06a9\u0646.<\/li>\n<li><code>layers<\/code>: \u0644\u06cc\u0633\u062a\u06cc \u06a9\u0647 \u0634\u0627\u0645\u0644 \u062a\u0639\u062f\u0627\u062f \u0646\u0648\u0631\u0648\u0646 \u0628\u0631\u0627\u06cc \u0647\u0645\u0647 \u0644\u0627\u06cc\u0647 \u0647\u0627 \u0627\u0633\u062a.<\/li>\n<li><code>p<\/code>: \u062e\u0631\u0648\u062c \u0628\u0627 \u0645\u0642\u062f\u0627\u0631 \u067e\u06cc\u0634 \u0641\u0631\u0636 0.5<\/li>\n<\/ol>\n<p>\u062f\u0631 \u062f\u0627\u062e\u0644 \u0633\u0627\u0632\u0646\u062f\u0647\u060c \u0686\u0646\u062f \u0645\u062a\u063a\u06cc\u0631 \u0645\u0642\u062f\u0627\u0631\u062f\u0647\u06cc \u0627\u0648\u0644\u06cc\u0647 \u0645\u06cc \u0634\u0648\u0646\u062f.  \u0627\u0648\u0644\u0627\u060c <code>all_embeddings<\/code> \u0645\u062a\u063a\u06cc\u0631 \u0634\u0627\u0645\u0644 \u0644\u06cc\u0633\u062a\u06cc \u0627\u0632 <code>ModuleList<\/code> \u0627\u0634\u06cc\u0627\u0621 \u0628\u0631\u0627\u06cc \u062a\u0645\u0627\u0645 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0634\u062f\u0647.  \u0627\u06cc\u0646 <code>embedding_dropout<\/code> \u0645\u0642\u062f\u0627\u0631 \u062d\u0630\u0641 \u0631\u0627 \u0628\u0631\u0627\u06cc \u0647\u0645\u0647 \u0644\u0627\u06cc\u0647 \u0647\u0627 \u0630\u062e\u06cc\u0631\u0647 \u0645\u06cc \u06a9\u0646\u062f.  \u062f\u0631 \u0646\u0647\u0627\u06cc\u062a\u060c <code>batch_norm_num<\/code> \u0644\u06cc\u0633\u062a\u06cc \u0627\u0632 <code>BatchNorm1d<\/code> \u0627\u0634\u06cc\u0627\u0621 \u0628\u0631\u0627\u06cc \u062a\u0645\u0627\u0645 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc.<\/p>\n<p>\u062f\u0631 \u0645\u0631\u062d\u0644\u0647 \u0628\u0639\u062f\u060c \u0628\u0631\u0627\u06cc \u06cc\u0627\u0641\u062a\u0646 \u0627\u0646\u062f\u0627\u0632\u0647 \u0644\u0627\u06cc\u0647 \u0648\u0631\u0648\u062f\u06cc\u060c \u062a\u0639\u062f\u0627\u062f \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u0648 \u0639\u062f\u062f\u06cc \u0628\u0627 \u0647\u0645 \u062c\u0645\u0639 \u0634\u062f\u0647 \u0648 \u062f\u0631 <code>input_size<\/code> \u0645\u062a\u063a\u06cc\u0631.  \u067e\u0633 \u0627\u0632 \u0622\u0646\u060c \u0627\u0644\u0641 <code>for<\/code> \u062d\u0644\u0642\u0647 \u062a\u06a9\u0631\u0627\u0631 \u0645\u06cc \u0634\u0648\u062f \u0648 \u0644\u0627\u06cc\u0647 \u0647\u0627\u06cc \u0645\u0631\u0628\u0648\u0637\u0647 \u0628\u0647 \u0622\u0646 \u0627\u0636\u0627\u0641\u0647 \u0645\u06cc \u0634\u0648\u0646\u062f <code>all_layers<\/code> \u0641\u0647\u0631\u0633\u062a  \u0644\u0627\u06cc\u0647 \u0647\u0627\u06cc \u0627\u0636\u0627\u0641\u0647 \u0634\u062f\u0647 \u0639\u0628\u0627\u0631\u062a\u0646\u062f \u0627\u0632:<\/p>\n<ul>\n<li><code>Linear<\/code>: \u0628\u0631\u0627\u06cc \u0645\u062d\u0627\u0633\u0628\u0647 \u062d\u0627\u0635\u0644 \u0636\u0631\u0628 \u0646\u0642\u0637\u0647 \u0627\u06cc \u0628\u06cc\u0646 \u0648\u0631\u0648\u062f\u06cc \u0647\u0627 \u0648 \u0645\u0627\u062a\u0631\u06cc\u0633 \u0647\u0627\u06cc \u0648\u0632\u0646\u06cc \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f<\/li>\n<li><code>ReLu<\/code>: \u06a9\u0647 \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u06cc\u06a9 \u062a\u0627\u0628\u0639 \u0641\u0639\u0627\u0644 \u0633\u0627\u0632\u06cc \u0627\u0639\u0645\u0627\u0644 \u0645\u06cc \u0634\u0648\u062f<\/li>\n<li><code>BatchNorm1d<\/code>: \u0628\u0631\u0627\u06cc \u0627\u0639\u0645\u0627\u0644 \u0646\u0631\u0645\u0627\u0644 \u0633\u0627\u0632\u06cc \u062f\u0633\u062a\u0647 \u0627\u06cc \u0628\u0647 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f<\/li>\n<li><code>Dropout<\/code>: \u0628\u0631\u0627\u06cc \u062c\u0644\u0648\u06af\u06cc\u0631\u06cc \u0627\u0632 \u0646\u0635\u0628 \u0628\u06cc\u0634 \u0627\u0632 \u062d\u062f \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f<\/li>\n<\/ul>\n<p>\u0628\u0639\u062f \u0627\u0632 <code>for<\/code> \u062d\u0644\u0642\u0647\u060c \u0644\u0627\u06cc\u0647 \u062e\u0631\u0648\u062c\u06cc \u0628\u0647 \u0644\u06cc\u0633\u062a \u0644\u0627\u06cc\u0647 \u0647\u0627 \u0627\u0636\u0627\u0641\u0647 \u0645\u06cc \u0634\u0648\u062f.  \u0627\u0632 \u0622\u0646\u062c\u0627\u06cc\u06cc \u06a9\u0647 \u0645\u06cc \u062e\u0648\u0627\u0647\u06cc\u0645 \u062a\u0645\u0627\u0645 \u0644\u0627\u06cc\u0647 \u0647\u0627\u06cc \u0634\u0628\u06a9\u0647 \u0639\u0635\u0628\u06cc \u0628\u0647 \u0635\u0648\u0631\u062a \u0645\u062a\u0648\u0627\u0644\u06cc \u0627\u062c\u0631\u0627 \u0634\u0648\u0646\u062f\u060c \u0644\u06cc\u0633\u062a \u0644\u0627\u06cc\u0647 \u0647\u0627 \u0628\u0647 <code>nn.Sequential<\/code> \u06a9\u0644\u0627\u0633<\/p>\n<p>\u0628\u0639\u062f\u060c \u062f\u0631 <code>forward<\/code> \u0631\u0648\u0634\u060c \u0647\u0631 \u062f\u0648 \u0633\u062a\u0648\u0646 \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc \u0648 \u0639\u062f\u062f\u06cc \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0648\u0631\u0648\u062f\u06cc \u0627\u0631\u0633\u0627\u0644 \u0645\u06cc \u0634\u0648\u0646\u062f.  \u062c\u0627\u0633\u0627\u0632\u06cc \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0634\u062f\u0647 \u062f\u0631 \u062e\u0637\u0648\u0637 \u0632\u06cc\u0631 \u0627\u0646\u062c\u0627\u0645 \u0645\u06cc \u0634\u0648\u062f.<\/p>\n<pre><code class=\"hljs\">embeddings = ()\n<span class=\"hljs-keyword\">for<\/span> i, e <span class=\"hljs-keyword\">in<\/span> <span class=\"hljs-built_in\">enumerate<\/span>(self.all_embeddings):\n    embeddings.append(e(x_categorical(:,i)))\nx = torch.cat(embeddings, <span class=\"hljs-number\">1<\/span>)\nx = self.embedding_dropout(x)\n<\/code><\/pre>\n<p>\u0646\u0631\u0645\u0627\u0644 \u0633\u0627\u0632\u06cc \u062f\u0633\u062a\u0647 \u0627\u06cc \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc \u0628\u0627 \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0627\u0639\u0645\u0627\u0644 \u0645\u06cc \u0634\u0648\u062f:<\/p>\n<pre><code class=\"hljs\">x_numerical = self.batch_norm_num(x_numerical)\n<\/code><\/pre>\n<p>\u062f\u0631 \u0646\u0647\u0627\u06cc\u062a\u060c \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0634\u062f\u0647 \u062a\u0639\u0628\u06cc\u0647 \u0634\u062f\u0647 \u0627\u0633\u062a <code>x<\/code> \u0648 \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc <code>x_numerical<\/code> \u0628\u0647 \u0647\u0645 \u0627\u0644\u062d\u0627\u0642 \u0645\u06cc \u0634\u0648\u0646\u062f \u0648 \u0628\u0647 \u062a\u0631\u062a\u06cc\u0628 \u0645\u0646\u062a\u0642\u0644 \u0645\u06cc \u0634\u0648\u0646\u062f <code>layers<\/code>.<\/p>\n<h2 id=\"trainingthemodel\"><span class=\"ez-toc-section\" id=\"%d8%a2%d9%85%d9%88%d8%b2%d8%b4_%d9%85%d8%af%d9%84\"><\/span>\u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>\u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644\u060c \u0627\u0628\u062a\u062f\u0627 \u0628\u0627\u06cc\u062f \u06cc\u06a9 \u0634\u06cc \u0627\u0632 the \u0627\u06cc\u062c\u0627\u062f \u06a9\u0646\u06cc\u0645 <code>Model<\/code> \u06a9\u0644\u0627\u0633\u06cc \u06a9\u0647 \u062f\u0631 \u0642\u0633\u0645\u062a \u0622\u062e\u0631 \u062a\u0639\u0631\u06cc\u0641 \u06a9\u0631\u062f\u06cc\u0645.<\/p>\n<pre><code class=\"hljs\">model = Model(categorical_embedding_sizes, numerical_data.shape(<span class=\"hljs-number\">1<\/span>), <span class=\"hljs-number\">2<\/span>, (<span class=\"hljs-number\">200<\/span>,<span class=\"hljs-number\">100<\/span>,<span class=\"hljs-number\">50<\/span>), p=<span class=\"hljs-number\">0.4<\/span>)\n<\/code><\/pre>\n<p>\u0645\u06cc \u0628\u06cc\u0646\u06cc\u062f \u06a9\u0647 \u0627\u0646\u062f\u0627\u0632\u0647 \u062c\u0627\u0633\u0627\u0632\u06cc \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc\u060c \u062a\u0639\u062f\u0627\u062f \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0639\u062f\u062f\u06cc\u060c \u0627\u0646\u062f\u0627\u0632\u0647 \u062e\u0631\u0648\u062c\u06cc (\u062f\u0631 \u0645\u0648\u0631\u062f \u0645\u0627 2) \u0648 \u0646\u0648\u0631\u0648\u0646 \u0647\u0627 \u062f\u0631 \u0644\u0627\u06cc\u0647 \u0647\u0627\u06cc \u067e\u0646\u0647\u0627\u0646 \u0631\u0627 \u067e\u0627\u0633 \u0645\u06cc \u06a9\u0646\u06cc\u0645.  \u0645\u06cc \u0628\u06cc\u0646\u06cc\u062f \u06a9\u0647 \u0645\u0627 \u0633\u0647 \u0644\u0627\u06cc\u0647 \u067e\u0646\u0647\u0627\u0646 \u062f\u0627\u0631\u06cc\u0645 \u06a9\u0647 \u0628\u0647 \u062a\u0631\u062a\u06cc\u0628 200\u060c 100 \u0648 50 \u0646\u0648\u0631\u0648\u0646 \u062f\u0627\u0631\u0646\u062f.  \u062f\u0631 \u0635\u0648\u0631\u062a \u062a\u0645\u0627\u06cc\u0644 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f \u0647\u0631 \u0633\u0627\u06cc\u0632 \u062f\u06cc\u06af\u0631\u06cc \u0631\u0627 \u0627\u0646\u062a\u062e\u0627\u0628 \u06a9\u0646\u06cc\u062f.<\/p>\n<p>\u0627\u062c\u0627\u0632\u0647 \u062f\u0647\u06cc\u062f print \u0645\u062f\u0644 \u0645\u0627 \u0648 \u0628\u0628\u06cc\u0646\u06cc\u062f \u06a9\u0647 \u0686\u06af\u0648\u0646\u0647 \u0628\u0647 \u0646\u0638\u0631 \u0645\u06cc \u0631\u0633\u062f:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-built_in\">print<\/span>(model)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">Model(\n  (all_embeddings): ModuleList(\n    (0): Embedding(3, 2)\n    (1): Embedding(2, 1)\n    (2): Embedding(2, 1)\n    (3): Embedding(2, 1)\n  )\n  (embedding_dropout): Dropout(p=0.4)\n  (batch_norm_num): BatchNorm1d(6, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n  (layers): Sequential(\n    (0): Linear(in_features=11, out_features=200, bias=True)\n    (1): ReLU(inplace)\n    (2): BatchNorm1d(200, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n    (3): Dropout(p=0.4)\n    (4): Linear(in_features=200, out_features=100, bias=True)\n    (5): ReLU(inplace)\n    (6): BatchNorm1d(100, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n    (7): Dropout(p=0.4)\n    (8): Linear(in_features=100, out_features=50, bias=True)\n    (9): ReLU(inplace)\n    (10): BatchNorm1d(50, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n    (11): Dropout(p=0.4)\n    (12): Linear(in_features=50, out_features=2, bias=True)\n  )\n)\n<\/code><\/pre>\n<p>\u0645\u06cc \u0628\u06cc\u0646\u06cc\u062f \u06a9\u0647 \u062f\u0631 \u0627\u0648\u0644\u06cc\u0646 \u0644\u0627\u06cc\u0647 \u062e\u0637\u06cc \u0645\u0642\u062f\u0627\u0631 the <code>in_features<\/code> \u0645\u062a\u063a\u06cc\u0631 11 \u0627\u0633\u062a \u0632\u06cc\u0631\u0627 \u0645\u0627 6 \u0633\u062a\u0648\u0646 \u0639\u062f\u062f\u06cc \u062f\u0627\u0631\u06cc\u0645 \u0648 \u0645\u062c\u0645\u0648\u0639 \u0627\u0628\u0639\u0627\u062f \u062a\u0639\u0628\u06cc\u0647 \u0634\u062f\u0647 \u0628\u0631\u0627\u06cc \u0633\u062a\u0648\u0646 \u0647\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0634\u062f\u0647 5 \u0627\u0633\u062a\u060c \u0628\u0646\u0627\u0628\u0631\u0627\u06cc\u0646 6+5 = 11. \u0628\u0647 \u0647\u0645\u06cc\u0646 \u062a\u0631\u062a\u06cc\u0628\u060c \u062f\u0631 \u0622\u062e\u0631\u06cc\u0646 \u0644\u0627\u06cc\u0647\u060c <code>out_features<\/code> \u062f\u0627\u0631\u0627\u06cc \u0645\u0642\u062f\u0627\u0631 2 \u0627\u0633\u062a \u0632\u06cc\u0631\u0627 \u0645\u0627 \u0641\u0642\u0637 2 \u062e\u0631\u0648\u062c\u06cc \u0645\u0645\u06a9\u0646 \u062f\u0627\u0631\u06cc\u0645.<\/p>\n<p>\u0642\u0628\u0644 \u0627\u0632 \u0627\u06cc\u0646\u06a9\u0647 \u0628\u062a\u0648\u0627\u0646\u06cc\u0645 \u0645\u062f\u0644 \u062e\u0648\u062f \u0631\u0627 \u0648\u0627\u0642\u0639\u0627\u064b \u0622\u0645\u0648\u0632\u0634 \u062f\u0647\u06cc\u0645\u060c \u0628\u0627\u06cc\u062f \u062a\u0627\u0628\u0639 \u0636\u0631\u0631 \u0648 \u0628\u0647\u06cc\u0646\u0647\u200c\u0633\u0627\u0632\u06cc \u0631\u0627 \u06a9\u0647 \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc\u200c\u0634\u0648\u062f\u060c \u062a\u0639\u0631\u06cc\u0641 \u06a9\u0646\u06cc\u0645.  \u0627\u0632 \u0622\u0646\u062c\u0627\u06cc\u06cc \u06a9\u0647 \u0645\u0627 \u062f\u0631 \u062d\u0627\u0644 \u062d\u0644 \u06cc\u06a9 \u0645\u0634\u06a9\u0644 \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0647\u0633\u062a\u06cc\u0645\u060c \u0627\u0632 \u0622\u0646 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u062e\u0648\u0627\u0647\u06cc\u0645 \u06a9\u0631\u062f <a rel=\"nofollow noopener\" target=\"_blank\" href=\"https:\/\/gombru.github.io\/2018\/05\/23\/cross_entropy_loss\/\">\u0627\u0632 \u062f\u0633\u062a \u062f\u0627\u062f\u0646 \u0622\u0646\u062a\u0631\u0648\u067e\u06cc \u0645\u062a\u0642\u0627\u0628\u0644<\/a>.  \u0628\u0631\u0627\u06cc \u062a\u0627\u0628\u0639 \u0628\u0647\u06cc\u0646\u0647 \u0633\u0627\u0632\u060c \u0627\u0632 <a rel=\"nofollow noopener\" target=\"_blank\" href=\"https:\/\/machinelearningmastery.com\/adam-optimization-algorithm-for-deep-learning\/\">\u0628\u0647\u06cc\u0646\u0647 \u0633\u0627\u0632 \u0622\u062f\u0627\u0645<\/a>.<\/p>\n<p>\u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u062a\u0627\u0628\u0639 \u0636\u0631\u0631 \u0648 \u0628\u0647\u06cc\u0646\u0647 \u0633\u0627\u0632 \u0631\u0627 \u062a\u0639\u0631\u06cc\u0641 \u0645\u06cc \u06a9\u0646\u062f:<\/p>\n<pre><code class=\"hljs\">loss_function = nn.CrossEntropyLoss()\noptimizer = torch.optim.Adam(model.parameters(), lr=<span class=\"hljs-number\">0.001<\/span>)\n<\/code><\/pre>\n<p>\u0627\u06a9\u0646\u0648\u0646 \u0645\u0627 \u0647\u0645\u0647 \u0686\u06cc\u0632 \u0645\u0648\u0631\u062f \u0646\u06cc\u0627\u0632 \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644 \u0631\u0627 \u062f\u0627\u0631\u06cc\u0645.  \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0645\u062f\u0644 \u0631\u0627 \u0622\u0645\u0648\u0632\u0634 \u0645\u06cc \u062f\u0647\u062f:<\/p>\n<pre><code class=\"hljs\">epochs = <span class=\"hljs-number\">300<\/span>\naggregated_losses = ()\n\n<span class=\"hljs-keyword\">for<\/span> i <span class=\"hljs-keyword\">in<\/span> <span class=\"hljs-built_in\">range<\/span>(epochs):\n    i += <span class=\"hljs-number\">1<\/span>\n    y_pred = model(categorical_train_data, numerical_train_data)\n    single_loss = loss_function(y_pred, train_outputs)\n    aggregated_losses.append(single_loss)\n\n    <span class=\"hljs-keyword\">if<\/span> i%<span class=\"hljs-number\">25<\/span> == <span class=\"hljs-number\">1<\/span>:\n        <span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-string\">f'epoch: <span class=\"hljs-subst\">{i:<span class=\"hljs-number\">3<\/span>}<\/span> loss: <span class=\"hljs-subst\">{single_loss.item():<span class=\"hljs-number\">10.8<\/span>f}<\/span>'<\/span>)\n\n    optimizer.zero_grad()\n    single_loss.backward()\n    optimizer.step()\n\n<span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-string\">f'epoch: <span class=\"hljs-subst\">{i:<span class=\"hljs-number\">3<\/span>}<\/span> loss: <span class=\"hljs-subst\">{single_loss.item():<span class=\"hljs-number\">10.10<\/span>f}<\/span>'<\/span>)\n<\/code><\/pre>\n<p>\u062a\u0639\u062f\u0627\u062f \u062f\u0648\u0631\u0647 \u0647\u0627 \u0631\u0648\u06cc 300 \u062a\u0646\u0638\u06cc\u0645 \u0634\u062f\u0647 \u0627\u0633\u062a\u060c \u0628\u0647 \u0627\u06cc\u0646 \u0645\u0639\u0646\u06cc \u06a9\u0647 \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634 \u0645\u062f\u0644\u060c \u0645\u062c\u0645\u0648\u0639\u0647 \u062f\u0627\u062f\u0647 \u06a9\u0627\u0645\u0644 300 \u0628\u0627\u0631 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f.  \u0622 <code>for<\/code> \u062d\u0644\u0642\u0647 300 \u0628\u0627\u0631 \u0627\u062c\u0631\u0627 \u0645\u06cc \u0634\u0648\u062f \u0648 \u062f\u0631 \u0637\u06cc \u0647\u0631 \u062a\u06a9\u0631\u0627\u0631\u060c \u0636\u0631\u0631 \u0628\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u062a\u0627\u0628\u0639 \u0636\u0631\u0631 \u0645\u062d\u0627\u0633\u0628\u0647 \u0645\u06cc \u0634\u0648\u062f.  \u0636\u0631\u0631 \u062f\u0631 \u0637\u0648\u0644 \u0647\u0631 \u062a\u06a9\u0631\u0627\u0631 \u0628\u0647 \u0636\u0645\u06cc\u0645\u0647 \u0645\u06cc \u0634\u0648\u062f <code>aggregated_loss<\/code> \u0641\u0647\u0631\u0633\u062a  \u0628\u0631\u0627\u06cc \u0628\u0647 \u0631\u0648\u0632 \u0631\u0633\u0627\u0646\u06cc \u0648\u0632\u0646\u0647 \u0647\u0627\u060c <code>backward()<\/code> \u0639\u0645\u0644\u06a9\u0631\u062f \u0627\u0632 <code>single_loss<\/code> \u0634\u06cc \u0646\u0627\u0645\u06cc\u062f\u0647 \u0645\u06cc \u0634\u0648\u062f.  \u062f\u0631 \u0646\u0647\u0627\u06cc\u062a\u060c <code>step()<\/code> \u0631\u0648\u0634 \u0627\u0632 <code>optimizer<\/code> \u062a\u0627\u0628\u0639 \u06af\u0631\u0627\u062f\u06cc\u0627\u0646 \u0631\u0627 \u0628\u0647 \u0631\u0648\u0632 \u0645\u06cc \u06a9\u0646\u062f.  \u0627\u0632 \u062f\u0633\u062a \u062f\u0627\u062f\u0646 \u067e\u0633 \u0627\u0632 \u0647\u0631 25 \u062f\u0648\u0631\u0647 \u0686\u0627\u067e \u0645\u06cc \u0634\u0648\u062f.<\/p>\n<p>\u062e\u0631\u0648\u062c\u06cc \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0628\u0627\u0644\u0627 \u0628\u0647 \u0634\u0631\u062d \u0632\u06cc\u0631 \u0627\u0633\u062a:<\/p>\n<pre><code class=\"hljs\">epoch:   1 loss: 0.71847951\nepoch:  26 loss: 0.57145703\nepoch:  51 loss: 0.48110831\nepoch:  76 loss: 0.42529839\nepoch: 101 loss: 0.39972275\nepoch: 126 loss: 0.37837571\nepoch: 151 loss: 0.37133673\nepoch: 176 loss: 0.36773482\nepoch: 201 loss: 0.36305946\nepoch: 226 loss: 0.36079505\nepoch: 251 loss: 0.35350436\nepoch: 276 loss: 0.35540250\nepoch: 300 loss: 0.3465710580\n<\/code><\/pre>\n<p>\u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0632\u06cc\u0627\u0646 \u0647\u0627 \u0631\u0627 \u062f\u0631 \u0628\u0631\u0627\u0628\u0631 \u062f\u0648\u0631\u0627\u0646 \u0647\u0627 \u062a\u0631\u0633\u06cc\u0645 \u0645\u06cc \u06a9\u0646\u062f:<\/p>\n<pre><code class=\"hljs\">plt.plot(<span class=\"hljs-built_in\">range<\/span>(epochs), aggregated_losses)\nplt.ylabel(<span class=\"hljs-string\">'Loss'<\/span>)\nplt.xlabel(<span class=\"hljs-string\">'epoch'<\/span>);\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<p><img decoding=\"async\" src=\"https:\/\/rasanegar.com\/blog\/wp-content\/uploads\/2024\/01\/Introduction-to-Pytorch-for-Classification-5.PNG\" alt=\"\u0646\u0642\u0634\u0647 \u06a9\u0634\u06cc\u062f\u0646 \u0632\u06cc\u0627\u0646 \u0647\u0627 \u062f\u0631 \u0628\u0631\u0627\u0628\u0631 \u062f\u0648\u0631\u0627\u0646 \u0647\u0627\" title=\"\"><\/p>\n<p>\u062e\u0631\u0648\u062c\u06cc \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u06a9\u0647 \u062f\u0631 \u0627\u0628\u062a\u062f\u0627 \u0636\u0631\u0631 \u0628\u0647 \u0633\u0631\u0639\u062a \u06a9\u0627\u0647\u0634 \u0645\u06cc \u06cc\u0627\u0628\u062f.  \u067e\u0633 \u0627\u0632 \u062d\u062f\u0648\u062f 250 \u062f\u0648\u0631\u0647\u060c \u06a9\u0627\u0647\u0634 \u0628\u0633\u06cc\u0627\u0631 \u06a9\u0645\u06cc \u062f\u0631 \u0636\u0631\u0631 \u0648\u062c\u0648\u062f \u062f\u0627\u0631\u062f.<\/p>\n<h2 id=\"makingpredictions\"><span class=\"ez-toc-section\" id=\"%d9%be%db%8c%d8%b4%da%af%d9%88%db%8c%db%8c\"><\/span>\u067e\u06cc\u0634\u06af\u0648\u06cc\u06cc<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>\u0622\u062e\u0631\u06cc\u0646 \u0645\u0631\u062d\u0644\u0647 \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc \u0627\u0633\u062a \u0631\u0648\u06cc \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u062a\u0633\u062a  \u0628\u0631\u0627\u06cc \u0627\u0646\u062c\u0627\u0645 \u0627\u06cc\u0646 \u06a9\u0627\u0631\u060c \u0645\u0627 \u0628\u0647 \u0633\u0627\u062f\u06af\u06cc \u0646\u06cc\u0627\u0632 \u0628\u0647 \u0639\u0628\u0648\u0631 \u0627\u0632 <code>categorical_test_data<\/code> \u0648 <code>numerical_test_data<\/code> \u0628\u0647 <code>model<\/code> \u06a9\u0644\u0627\u0633  \u0633\u067e\u0633 \u0645\u0642\u0627\u062f\u06cc\u0631 \u0628\u0627\u0632\u06af\u0634\u062a\u06cc \u0631\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646 \u0628\u0627 \u0645\u0642\u0627\u062f\u06cc\u0631 \u062e\u0631\u0648\u062c\u06cc \u0622\u0632\u0645\u0627\u06cc\u0634\u06cc \u0648\u0627\u0642\u0639\u06cc \u0645\u0642\u0627\u06cc\u0633\u0647 \u06a9\u0631\u062f.  \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc \u0645\u06cc \u06a9\u0646\u062f \u0631\u0648\u06cc \u06a9\u0644\u0627\u0633 \u062a\u0633\u062a \u0648 \u0627\u0641\u062a \u0622\u0646\u062a\u0631\u0648\u067e\u06cc \u0645\u062a\u0642\u0627\u0637\u0639 \u0631\u0627 \u0628\u0631\u0627\u06cc \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u062a\u0633\u062a \u0686\u0627\u067e \u0645\u06cc \u06a9\u0646\u062f.<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-keyword\">with<\/span> torch.no_grad():\n    y_val = model(categorical_test_data, numerical_test_data)\n    loss = loss_function(y_val, test_outputs)\n<span class=\"hljs-built_in\">print<\/span>(<span class=\"hljs-string\">f'Loss: <span class=\"hljs-subst\">{loss:<span class=\"hljs-number\">.8<\/span>f}<\/span>'<\/span>)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">Loss: 0.36855841\n<\/code><\/pre>\n<p>\u0627\u0632 \u062f\u0633\u062a \u062f\u0627\u062f\u0646 \u0631\u0648\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u062a\u0633\u062a 0.3685 \u0627\u0633\u062a \u06a9\u0647 \u06a9\u0645\u06cc \u0628\u06cc\u0634\u062a\u0631 \u0627\u0632 0.3465 \u0628\u0647 \u062f\u0633\u062a \u0622\u0645\u062f\u0647 \u0627\u0633\u062a \u0631\u0648\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u0622\u0645\u0648\u0632\u0634\u06cc \u06a9\u0647 \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u0645\u062f\u0644 \u0645\u0627 \u06a9\u0645\u06cc \u0628\u06cc\u0634 \u0627\u0632 \u062d\u062f \u0645\u0646\u0627\u0633\u0628 \u0627\u0633\u062a.<\/p>\n<p>\u062a\u0648\u062c\u0647 \u0628\u0647 \u0627\u06cc\u0646 \u0646\u06a9\u062a\u0647 \u0645\u0647\u0645 \u0627\u0633\u062a \u06a9\u0647 \u0627\u0632 \u0622\u0646\u062c\u0627\u06cc\u06cc \u06a9\u0647 \u0645\u0627 \u0645\u0634\u062e\u0635 \u06a9\u0631\u062f\u06cc\u0645 \u06a9\u0647 \u0644\u0627\u06cc\u0647 \u062e\u0631\u0648\u062c\u06cc \u0645\u0627 \u0634\u0627\u0645\u0644 2 \u0646\u0648\u0631\u0648\u0646 \u062e\u0648\u0627\u0647\u062f \u0628\u0648\u062f\u060c \u0647\u0631 \u067e\u06cc\u0634 \u0628\u06cc\u0646\u06cc \u062d\u0627\u0648\u06cc 2 \u0645\u0642\u062f\u0627\u0631 \u062e\u0648\u0627\u0647\u062f \u0628\u0648\u062f.  \u0628\u0631\u0627\u06cc \u0645\u062b\u0627\u0644\u060c 5 \u0645\u0642\u062f\u0627\u0631 \u067e\u06cc\u0634\u200c\u0628\u06cc\u0646\u06cc\u200c\u0634\u062f\u0647 \u0627\u0648\u0644 \u0628\u0647 \u0627\u06cc\u0646 \u0635\u0648\u0631\u062a \u0627\u0633\u062a:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-built_in\">print<\/span>(y_val(:<span class=\"hljs-number\">5<\/span>))\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">tensor((( 1.2045, -1.3857),\n        ( 1.3911, -1.5957),\n        ( 1.2781, -1.3598),\n        ( 0.6261, -0.5429),\n        ( 2.5430, -1.9991)))\n<\/code><\/pre>\n<p>\u0627\u06cc\u062f\u0647 \u067e\u0634\u062a \u0686\u0646\u06cc\u0646 \u067e\u06cc\u0634\u200c\u0628\u06cc\u0646\u06cc\u200c\u0647\u0627\u06cc\u06cc \u0627\u06cc\u0646 \u0627\u0633\u062a \u06a9\u0647 \u0627\u06af\u0631 \u062e\u0631\u0648\u062c\u06cc \u0648\u0627\u0642\u0639\u06cc 0 \u0628\u0627\u0634\u062f\u060c \u0645\u0642\u062f\u0627\u0631 \u0634\u0627\u062e\u0635 0 \u0628\u0627\u06cc\u062f \u0628\u06cc\u0634\u062a\u0631 \u0627\u0632 \u0645\u0642\u062f\u0627\u0631 \u0634\u0627\u062e\u0635 1 \u0628\u0627\u0634\u062f \u0648 \u0628\u0627\u0644\u0639\u06a9\u0633.  \u0628\u0627 \u0627\u0633\u06a9\u0631\u06cc\u067e\u062a \u0632\u06cc\u0631 \u0645\u06cc\u200c\u062a\u0648\u0627\u0646\u06cc\u0645 \u0627\u06cc\u0646\u062f\u06a9\u0633 \u0628\u0632\u0631\u06af\u062a\u0631\u06cc\u0646 \u0645\u0642\u062f\u0627\u0631 \u0645\u0648\u062c\u0648\u062f \u062f\u0631 \u0644\u06cc\u0633\u062a \u0631\u0627 \u0628\u0627\u0632\u06cc\u0627\u0628\u06cc \u06a9\u0646\u06cc\u0645:<\/p>\n<pre><code class=\"hljs\">y_val = np.argmax(y_val, axis=<span class=\"hljs-number\">1<\/span>)\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<p>\u062d\u0627\u0644\u0627 \u062f\u0648\u0628\u0627\u0631\u0647 print \u067e\u0646\u062c \u0645\u0642\u062f\u0627\u0631 \u0627\u0648\u0644 \u0628\u0631\u0627\u06cc <code>y_val<\/code> \u0644\u06cc\u0633\u062a:<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-built_in\">print<\/span>(y_val(:<span class=\"hljs-number\">5<\/span>))\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">tensor((0, 0, 0, 0, 0))\n<\/code><\/pre>\n<p>\u0627\u0632 \u0622\u0646\u062c\u0627\u06cc\u06cc \u06a9\u0647 \u062f\u0631 \u0641\u0647\u0631\u0633\u062a \u062e\u0631\u0648\u062c\u06cc\u200c\u0647\u0627\u06cc \u067e\u06cc\u0634\u200c\u0628\u06cc\u0646\u06cc\u200c\u0634\u062f\u0647 \u0627\u0648\u0644\u06cc\u0647\u060c \u0628\u0631\u0627\u06cc \u067e\u0646\u062c \u0631\u06a9\u0648\u0631\u062f \u0627\u0648\u0644\u060c \u0645\u0642\u0627\u062f\u06cc\u0631 \u062f\u0631 \u0634\u0627\u062e\u0635\u200c\u0647\u0627\u06cc \u0635\u0641\u0631 \u0628\u0632\u0631\u06af\u200c\u062a\u0631 \u0627\u0632 \u0645\u0642\u0627\u062f\u06cc\u0631 \u0634\u0627\u062e\u0635\u200c\u0647\u0627\u06cc \u0627\u0648\u0644 \u0647\u0633\u062a\u0646\u062f\u060c \u0645\u06cc\u200c\u062a\u0648\u0627\u0646\u06cc\u0645 \u06f0 \u0631\u0627 \u062f\u0631 \u067e\u0646\u062c \u0631\u062f\u06cc\u0641 \u0627\u0648\u0644 \u062e\u0631\u0648\u062c\u06cc\u200c\u0647\u0627\u06cc \u067e\u0631\u062f\u0627\u0632\u0634 \u0634\u062f\u0647 \u0628\u0628\u06cc\u0646\u06cc\u0645.<\/p>\n<p>\u062f\u0631 \u0646\u0647\u0627\u06cc\u062a \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u0645 \u0627\u0632 <code>confusion_matrix<\/code>\u060c <code>accuracy_score<\/code>\u060c \u0648 <code>classification_report<\/code> \u06a9\u0644\u0627\u0633 \u0647\u0627 \u0627\u0632 <code>sklearn.metrics<\/code> \u0645\u0627\u0698\u0648\u0644 \u0628\u0631\u0627\u06cc \u06cc\u0627\u0641\u062a\u0646 \u0645\u0642\u0627\u062f\u06cc\u0631 \u062f\u0642\u062a\u060c \u062f\u0642\u062a \u0648 \u0641\u0631\u0627\u062e\u0648\u0627\u0646\u06cc \u0645\u062c\u0645\u0648\u0639\u0647 \u062a\u0633\u062a\u060c \u0647\u0645\u0631\u0627\u0647 \u0628\u0627 \u0645\u0627\u062a\u0631\u06cc\u0633 \u0633\u0631\u062f\u0631\u06af\u0645\u06cc.<\/p>\n<pre><code class=\"hljs\"><span class=\"hljs-keyword\">from<\/span> sklearn.metrics <span class=\"hljs-keyword\">import<\/span> classification_report, confusion_matrix, accuracy_score\n\n<span class=\"hljs-built_in\">print<\/span>(confusion_matrix(test_outputs,y_val))\n<span class=\"hljs-built_in\">print<\/span>(classification_report(test_outputs,y_val))\n<span class=\"hljs-built_in\">print<\/span>(accuracy_score(test_outputs, y_val))\n<\/code><\/pre>\n<p><strong>\u062e\u0631\u0648\u062c\u06cc:<\/strong><\/p>\n<pre><code class=\"hljs\">((1527   83)\n ( 224  166))\n              precision    recall  f1-score   support\n\n           0       0.87      0.95      0.91      1610\n           1       0.67      0.43      0.52       390\n\n   micro avg       0.85      0.85      0.85      2000\n   macro avg       0.77      0.69      0.71      2000\nweighted avg       0.83      0.85      0.83      2000\n\n0.8465\n<\/code><\/pre>\n<p>\u062e\u0631\u0648\u062c\u06cc \u0646\u0634\u0627\u0646 \u0645\u06cc \u062f\u0647\u062f \u06a9\u0647 \u0645\u062f\u0644 \u0645\u0627 \u0628\u0647 \u062f\u0642\u062a 84.65% \u062f\u0633\u062a \u0645\u06cc \u06cc\u0627\u0628\u062f \u06a9\u0647 \u0628\u0627 \u062a\u0648\u062c\u0647 \u0628\u0647 \u0627\u06cc\u0646 \u0648\u0627\u0642\u0639\u06cc\u062a \u06a9\u0647 \u0645\u0627 \u0628\u0647 \u0637\u0648\u0631 \u062a\u0635\u0627\u062f\u0641\u06cc \u062a\u0645\u0627\u0645 \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u0647\u0627 \u0631\u0627 \u0628\u0631\u0627\u06cc \u0645\u062f\u0644 \u0634\u0628\u06a9\u0647 \u0639\u0635\u0628\u06cc \u062e\u0648\u062f \u0627\u0646\u062a\u062e\u0627\u0628 \u06a9\u0631\u062f\u06cc\u0645 \u0628\u0633\u06cc\u0627\u0631 \u0686\u0634\u0645\u06af\u06cc\u0631 \u0627\u0633\u062a.  \u0645\u0646 \u067e\u06cc\u0634\u0646\u0647\u0627\u062f \u0645\u06cc \u06a9\u0646\u0645 \u06a9\u0647 \u0633\u0639\u06cc \u06a9\u0646\u06cc\u062f \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u0647\u0627\u06cc \u0645\u062f\u0644 \u0645\u0627\u0646\u0646\u062f \u062a\u0642\u0633\u06cc\u0645 \u0642\u0637\u0627\u0631\/\u062a\u0633\u062a\u060c \u062a\u0639\u062f\u0627\u062f \u0648 \u0627\u0646\u062f\u0627\u0632\u0647 \u0644\u0627\u06cc\u0647 \u0647\u0627\u06cc \u067e\u0646\u0647\u0627\u0646 \u0648 \u063a\u06cc\u0631\u0647 \u0631\u0627 \u062a\u063a\u06cc\u06cc\u0631 \u062f\u0647\u06cc\u062f \u062a\u0627 \u0628\u0628\u06cc\u0646\u06cc\u062f \u0622\u06cc\u0627 \u0645\u06cc \u062a\u0648\u0627\u0646\u06cc\u062f \u0646\u062a\u0627\u06cc\u062c \u0628\u0647\u062a\u0631\u06cc \u0628\u06af\u06cc\u0631\u06cc\u062f \u06cc\u0627 \u062e\u06cc\u0631.<\/p>\n<h2 id=\"conclusion\"><span class=\"ez-toc-section\" id=\"%d9%86%d8%aa%db%8c%d8%ac%d9%87\"><\/span>\u0646\u062a\u06cc\u062c\u0647<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>PyTorch \u06cc\u06a9 \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u0639\u0645\u06cc\u0642 \u0627\u0633\u062a \u06a9\u0647 \u062a\u0648\u0633\u0637 \u0641\u06cc\u0633 \u0628\u0648\u06a9 \u062a\u0648\u0633\u0639\u0647 \u06cc\u0627\u0641\u062a\u0647 \u0648 \u0645\u06cc \u062a\u0648\u0627\u0646\u062f \u0628\u0631\u0627\u06cc \u06a9\u0627\u0631\u0647\u0627\u06cc \u0645\u062e\u062a\u0644\u0641\u06cc \u0645\u0627\u0646\u0646\u062f \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc\u060c \u0631\u06af\u0631\u0633\u06cc\u0648\u0646 \u0648 \u062e\u0648\u0634\u0647 \u0628\u0646\u062f\u06cc \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0634\u0648\u062f.  \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u0631\u0648\u0634 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 PyTorch \u0631\u0627 \u0628\u0631\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u062c\u062f\u0648\u0644\u06cc \u062a\u0648\u0636\u06cc\u062d \u0645\u06cc \u062f\u0647\u062f.<\/p>\n<\/div>\n<p><script>\n                        !function(f,b,e,v,n,t,s)\n                        {if(f.fbq)return;n=f.fbq=function(){n.callMethod?\n                        n.callMethod.apply(n,arguments):n.queue.push(arguments)};\n                        if(!f._fbq)f._fbq=n;n.push=n;n.loaded=!0;n.version='2.0';\n                        n.queue=();t=b.createElement(e);t.async=!0;\n                        t.src=v;s=b.getElementsByTagName(e)(0);\n                        s.parentNode.insertBefore(t,s)}(window, document,'script',\n                        'https:\/\/connect.facebook.net\/en_US\/fbevents.js');\n                        fbq('init', '525232124909042');\n                        fbq('track', 'PageView');\n                    <\/script>    (\u0628\u0631\u0686\u0633\u0628\u200c\u0647\u0627 \u0628\u0647 \u062a\u0631\u062c\u0645\u0647)# python<br \/>\n<br \/><br \/>\n<br \/>\u0645\u0646\u062a\u0634\u0631 \u0634\u062f\u0647 \u062f\u0631 1403-01-19 22:49:05<br \/>\n<\/p>\n\n\n<div class=\"kk-star-ratings kksr-auto kksr-align-center kksr-valign-bottom\"\n    data-payload='{&quot;align&quot;:&quot;center&quot;,&quot;id&quot;:&quot;16022&quot;,&quot;slug&quot;:&quot;default&quot;,&quot;valign&quot;:&quot;bottom&quot;,&quot;ignore&quot;:&quot;&quot;,&quot;reference&quot;:&quot;auto&quot;,&quot;class&quot;:&quot;&quot;,&quot;count&quot;:&quot;0&quot;,&quot;legendonly&quot;:&quot;&quot;,&quot;readonly&quot;:&quot;&quot;,&quot;score&quot;:&quot;0&quot;,&quot;starsonly&quot;:&quot;&quot;,&quot;best&quot;:&quot;5&quot;,&quot;gap&quot;:&quot;5&quot;,&quot;greet&quot;:&quot;\u0627\u0645\u062a\u06cc\u0627\u0632 \u0634\u0645\u0627 \u0628\u0647 \u0627\u06cc\u0646 \u0645\u0637\u0644\u0628&quot;,&quot;legend&quot;:&quot;0\\\/5 (0 \u0631\u0627\u06cc)&quot;,&quot;size&quot;:&quot;30&quot;,&quot;title&quot;:&quot;\u0645\u0642\u062f\u0645\u0647 \u0627\u06cc \u0628\u0631 PyTorch \u0628\u0631\u0627\u06cc \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc&quot;,&quot;width&quot;:&quot;0&quot;,&quot;_legend&quot;:&quot;{score}\\\/{best} ({count} \u0631\u0627\u06cc)&quot;,&quot;font_factor&quot;:&quot;1.25&quot;}'>\n            \n<div class=\"kksr-stars\">\n    \n<div class=\"kksr-stars-inactive\">\n            <div class=\"kksr-star\" data-star=\"1\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n            <div class=\"kksr-star\" data-star=\"2\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n            <div class=\"kksr-star\" data-star=\"3\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n            <div class=\"kksr-star\" data-star=\"4\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n            <div class=\"kksr-star\" data-star=\"5\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n    <\/div>\n    \n<div class=\"kksr-stars-active\" style=\"width: 0px;\">\n            <div class=\"kksr-star\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n            <div class=\"kksr-star\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n            <div class=\"kksr-star\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n            <div class=\"kksr-star\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n            <div class=\"kksr-star\" style=\"padding-left: 5px\">\n            \n\n<div class=\"kksr-icon\" style=\"width: 30px; height: 30px;\"><\/div>\n        <\/div>\n    <\/div>\n<\/div>\n                \n\n<div class=\"kksr-legend\" style=\"font-size: 24px;\">\n            <span class=\"kksr-muted\">\u0627\u0645\u062a\u06cc\u0627\u0632 \u0634\u0645\u0627 \u0628\u0647 \u0627\u06cc\u0646 \u0645\u0637\u0644\u0628<\/span>\n    <\/div>\n    <\/div>\n","protected":false},"excerpt":{"rendered":"<p><span class=\"span-reading-time rt-reading-time\" style=\"display: block;\"><span class=\"rt-label rt-prefix\">\u0632\u0645\u0627\u0646 \u0644\u0627\u0632\u0645 \u0628\u0631\u0627\u06cc \u0645\u0637\u0627\u0644\u0639\u0647: <\/span> <span class=\"rt-time\"> 13<\/span> <span class=\"rt-label rt-postfix\">\u062f\u0642\u06cc\u0642\u0647<\/span><\/span>PyTorch \u0648 TensorFlow \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 \u0647\u0627 \u062f\u0648 \u0645\u0648\u0631\u062f \u0627\u0632 \u0631\u0627\u06cc\u062c \u062a\u0631\u06cc\u0646 \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 \u0647\u0627\u06cc \u067e\u0627\u06cc\u062a\u0648\u0646 \u0628\u0631\u0627\u06cc \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u0639\u0645\u06cc\u0642 \u0647\u0633\u062a\u0646\u062f. PyTorch \u062a\u0648\u0633\u0637 \u0641\u06cc\u0633 \u0628\u0648\u06a9 \u062a\u0648\u0633\u0639\u0647 \u062f\u0627\u062f\u0647 \u0634\u062f\u0647 \u0627\u0633\u062a\u060c \u062f\u0631 \u062d\u0627\u0644\u06cc \u06a9\u0647 TensorFlow \u06cc\u06a9 \u067e\u0631\u0648\u0698\u0647 \u06af\u0648\u06af\u0644 \u0627\u0633\u062a. \u062f\u0631 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u062e\u0648\u0627\u0647\u06cc\u062f \u062f\u06cc\u062f \u06a9\u0647 \u0686\u06af\u0648\u0646\u0647 \u0645\u06cc \u062a\u0648\u0627\u0646 \u0627\u0632 \u06a9\u062a\u0627\u0628\u062e\u0627\u0646\u0647 PyTorch \u0628\u0631\u0627\u06cc \u062d\u0644 \u0645\u0634\u06a9\u0644\u0627\u062a \u0637\u0628\u0642\u0647 \u0628\u0646\u062f\u06cc \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u06a9\u0631\u062f. \u0645\u0633\u0627\u0626\u0644 \u0637\u0628\u0642\u0647\u200c\u0628\u0646\u062f\u06cc \u0628\u0647 [&hellip;]<\/p>\n","protected":false},"author":3,"featured_media":16023,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1743,620],"tags":[],"class_list":["post-16022","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-python","category-programming"],"acf":[],"_links":{"self":[{"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/posts\/16022","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/comments?post=16022"}],"version-history":[{"count":0,"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/posts\/16022\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/media\/16023"}],"wp:attachment":[{"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/media?parent=16022"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/categories?post=16022"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/rasanegaar.com\/blog\/wp-json\/wp\/v2\/tags?post=16022"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}