{"id":21374,"date":"2020-11-25T09:37:20","date_gmt":"2020-11-25T09:37:20","guid":{"rendered":"https:\/\/www.artefact.com\/?post_type=news&#038;p=21374"},"modified":"2024-09-20T17:45:32","modified_gmt":"2024-09-20T16:45:32","slug":"how-to-train-a-language-model-from-scratch-without-any-linguistic-knowledge","status":"publish","type":"blog","link":"https:\/\/www.artefact.com\/zh\/blog\/how-to-train-a-language-model-from-scratch-without-any-linguistic-knowledge\/","title":{"rendered":"\u5982\u4f55\u5728\u6ca1\u6709\u4efb\u4f55\u8bed\u8a00\u77e5\u8bc6\u7684\u60c5\u51b5\u4e0b\u4ece\u5934\u5f00\u59cb\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b"},"content":{"rendered":"<p><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-1 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-padding-top:0px;--awb-padding-right:0px;--awb-padding-bottom:0px;--awb-padding-left:0px;--awb-margin-bottom:40px;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-0 fusion_builder_column_1_2 1_2 fusion-flex-column avada-news-bloc-image\" style=\"--awb-bg-size:cover;--awb-width-large:50%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:50%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-image-element\" style=\"text-align:left;--awb-caption-title-font-family:var(--h2_typography-font-family);--awb-caption-title-font-weight:var(--h2_typography-font-weight);--awb-caption-title-font-style:var(--h2_typography-font-style);--awb-caption-title-size:var(--h2_typography-font-size);--awb-caption-title-transform:var(--h2_typography-text-transform);--awb-caption-title-line-height:var(--h2_typography-line-height);--awb-caption-title-letter-spacing:var(--h2_typography-letter-spacing);\"><span class=\"fusion-imageframe imageframe-none imageframe-1 hover-type-none\"><img decoding=\"async\" width=\"300\" height=\"74\" title=\"\u4e2d\u578b Blog\" src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog.png\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-300x74.png\" alt class=\"lazyload img-responsive wp-image-60582\" srcset=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%274000%27%20height%3D%27992%27%20viewBox%3D%270%200%204000%20992%27%3E%3Crect%20width%3D%274000%27%20height%3D%27992%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-srcset=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-200x50.png 200w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-400x99.png 400w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-600x149.png 600w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-800x198.png 800w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-1200x298.png 1200w\" data-sizes=\"auto\" data-orig-sizes=\"(max-width: 640px) 100vw, 800px\" \/><\/span><\/div><\/div><\/div><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-1 fusion_builder_column_1_2 1_2 fusion-flex-column\" style=\"--awb-padding-top:10px;--awb-padding-right:10px;--awb-padding-bottom:10px;--awb-padding-left:10px;--awb-bg-size:cover;--awb-width-large:50%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:50%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-text fusion-text-1\"><p>\u65b0\u95fb \/ \u4eba\u5de5\u667a\u80fd\u6280\u672f<\/p>\n<\/div><div class=\"fusion-text fusion-text-2\"><p>2020 \u5e74 11 \u6708 25 \u65e5<br \/>\n\u5728\u672c\u6587\u4e2d\uff0c\u6cd5\u56fd Artefact \u9ad8\u7ea7 Data \u79d1\u5b66\u5bb6 Amale El Hamri \u4ecb\u7ecd\u4e86\u5982\u4f55\u5728\u4e0d\u7406\u89e3\u8bed\u8a00\u7684\u60c5\u51b5\u4e0b\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u3002\u6587\u7ae0\u5185\u5bb9\u5305\u62ec\u4ece\u54ea\u91cc\u83b7\u53d6\u8bad\u7ec3 data\u3001\u9700\u8981\u591a\u5c11 data\u3001\u5982\u4f55\u9884\u5904\u7406 data\uff0c\u4ee5\u53ca\u5982\u4f55\u627e\u5230\u6700\u9002\u5408\u6a21\u578b\u7684\u67b6\u6784\u548c\u8d85\u53c2\u6570\u96c6\u3002.<\/p>\n<\/div><\/div><\/div><\/div><\/div><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-2 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-background-color:#ffffff;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-2 fusion_builder_column_1_1 1_1 fusion-flex-column\" style=\"--awb-bg-color:#ffffff;--awb-bg-color-hover:#ffffff;--awb-bg-size:cover;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-text fusion-text-3\" style=\"--awb-text-color:#000000;\"><p data-selectable-paragraph=\"\"><strong><span style=\"color: #ff0066;\">TLDR<\/span><\/strong><\/p>\n<p id=\"d30c\" class=\"ii ij dn ik b il im hp in io ip ht iq ir is it iu iv iw ix iy iz ja jb jc jd df ek\" data-selectable-paragraph=\"\">\u672c\u6587\u4ecb\u7ecd\u4e86\u6211\u5982\u4f55\u7528\u97e9\u8bed\u521b\u5efa\u81ea\u5df1\u7684\u8bed\u8a00\u6a21\u578b\uff0c\u97e9\u8bed\u662f\u4e00\u79cd\u590d\u6742\u7684\u8bed\u8a00\uff0c\u8bad\u7ec3 data \u6709\u9650\u3002\u5728\u8fd9\u91cc\uff0c\u60a8\u53ef\u4ee5\u5b66\u4e60\u5982\u4f55\u5728\u81ea\u5df1\u4e0d\u61c2\u8fd9\u79cd\u8bed\u8a00\u7684\u60c5\u51b5\u4e0b\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u3002\u60a8\u5c06\u4e86\u89e3\u5230\u4ece\u54ea\u91cc\u83b7\u53d6\u8bad\u7ec3 data\u3001\u9700\u8981\u591a\u5c11 data\u3001\u5982\u4f55\u9884\u5904\u7406 data\uff0c\u4ee5\u53ca\u5982\u4f55\u627e\u5230\u6700\u9002\u5408\u60a8\u7684\u6a21\u578b\u7684\u67b6\u6784\u548c\u8d85\u53c2\u6570\u96c6\u3002.<\/p>\n<p><em>\u6211\u7684\u4e3b\u8981\u4f53\u4f1a\u662f<\/em><\/p>\n<p id=\"fbfb\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">Data \u7cfb\u5217\uff1a<\/p>\n<ul class=\"\">\n<li id=\"c08c\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u5982\u679c\u7ef4\u57fa\u767e\u79d1\u7684\u5bb9\u91cf\u4e0d\u591f\u5927\uff0c\u6216\u8005\u60a8\u60f3\u8981\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u7684\u8bed\u8a00\u7684\u6bcd\u8bed\u4f7f\u7528\u8005\u5bf9\u7ef4\u57fa\u767e\u79d1\u7684\u4f7f\u7528\u4e0d\u591f\u591a\uff0c\u90a3\u4e48\u60a8\u53ef\u4ee5\u5c06\u7ef4\u57fa\u767e\u79d1\u4e0e\u5176\u4ed6 data \u8d44\u6e90\uff08\u5982 CommonCrawl\uff09\u7ed3\u5408\u8d77\u6765\u3002.<\/li>\n<\/ul>\n<p id=\"f465\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">Data \u97f3\u91cf\uff1a<\/p>\n<ul class=\"\">\n<li id=\"9d15\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u9009\u62e9\u6700\u80fd\u4ee3\u8868\u97e9\u8bed\u7684\u6587\u6863\u3002\u8fc7\u591a\u7684\u6587\u6863\u5e76\u6ca1\u6709\u7528\u5904\uff0c\u56e0\u4e3a\u4e0e\u5de8\u5927\u7684\u8bad\u7ec3\u65f6\u95f4\u76f8\u6bd4\uff0c\u8fb9\u9645\u6027\u80fd\u63d0\u5347\u592a\u5c0f\u3002.<\/li>\n<li id=\"d32b\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u9009\u62e9\u5305\u542b\u97e9\u8bed\u5e38\u7528\u8bcd\u7684\u6587\u4ef6\u3002.<\/li>\n<li id=\"b6cc\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u627e\u5230\u4e00\u79cd\u80fd\u5c06 data \u57f9\u8bad\u7684\u590d\u6742\u6027\u6a21\u578b\u5316\u7684\u7ed3\u6784\u3002.<\/li>\n<li id=\"c290\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u627e\u5230\u6b63\u5219\u5316\u53c2\u6570\u7684\u6b63\u786e\u7ec4\u5408\uff0c\u907f\u514d\u8fc7\u62df\u5408\u3002.<\/li>\n<\/ul>\n<p id=\"2e04\" class=\"hk hl dn ce hm hn ho hp hq hr hs ht hu hv hw hx hy hz ia ib ic id ie if ig ih ek\"><strong><span style=\"color: #ff0066;\">\u5bfc\u8a00<\/span><\/strong><\/p>\n<p class=\"hk hl dn ce hm hn ho hp hq hr hs ht hu hv hw hx hy hz ia ib ic id ie if ig ih ek\">\u5982\u679c\u4f60\u8fd8\u4e0d\u77e5\u9053\uff0cNLP \u5728\u8fc7\u53bb\u4e24\u5e74\u91cc\u5927\u8086\u7092\u4f5c\u8fc1\u79fb\u5b66\u4e60\u3002\u5176\u4e3b\u8981\u601d\u60f3\u662f\u5c06\u9884\u5148\u8bad\u7ec3\u597d\u7684\u8bed\u8a00\u6a21\u578b\u91cd\u65b0\u7528\u4e8e\u53e6\u4e00\u9879 NLP \u4efb\u52a1\uff0c\u5982\u6587\u672c\u5206\u7c7b\u3002\u8bed\u8a00\u6a21\u578b\u662f\u4e00\u79cd\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u7ed9\u5b9a\u53e5\u5b50\u7684\u4e00\u90e8\u5206\u5c31\u80fd\u9884\u6d4b\u53e5\u5b50\u7684\u4e0b\u4e00\u4e2a\u5355\u8bcd\u3002\u4ece\u76f4\u89c2\u4e0a\u7406\u89e3\uff0c\u8fd9\u79cd\u6a21\u578b\u80fd\u591f\u5f88\u597d\u5730\u7406\u89e3\u8bed\u8a00\u7ed3\u6784\u3001\u8bed\u6cd5\u548c\u8bcd\u6c47\uff0c\u5176\u76ee\u6807\u662f\u5c06\u8fd9\u4e9b\u77e5\u8bc6 \u2018\u8f6c\u79fb \u2019\u7ed9\u5176\u4ed6\u4e0b\u6e38\u6a21\u578b\u3002.<\/p>\n<div class=\"gt gu gv gw aj gx\" tabindex=\"0\" role=\"button\">\n<div class=\"cx cy jr\">\n<div class=\"hd s gv he\">\n<div class=\"js hg s\">\n<div class=\"gy gz t u v ha aj bm hb hc\"><\/div>\n<p><img decoding=\"async\" class=\"lazyload tq tr t u v ha aj c aligncenter\" src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0maA5NXtXpMp8KerA.png\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0maA5NXtXpMp8KerA.png\" data-sizes=\"auto\" data-orig-sizes=\"700px\" srcset=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%271020%27%20height%3D%27433%27%20viewBox%3D%270%200%201020%20433%27%3E%3Crect%20width%3D%271020%27%20height%3D%27433%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-srcset=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0maA5NXtXpMp8KerA.png 276w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0maA5NXtXpMp8KerA.png 552w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0maA5NXtXpMp8KerA.png 640w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0maA5NXtXpMp8KerA.png 700w\" alt=\"Image for post\" width=\"1020\" height=\"433\" \/><\/p>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<p id=\"8777\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\" style=\"text-align: center;\"><em>\u793a\u4f8b\uff1a\u5982\u4f55\u5229\u7528\u5fae\u8c03\u6539\u8fdb\u6587\u672c\u5206\u7c7b\u5668\u7684\u7b80\u5355\u65b9\u6cd5<\/em><\/p>\n<p id=\"d4c8\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u8be5\u56fe\u603b\u7ed3\u4e86\u6211\u7528\u6765\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u7684 ULM Fit \u65b9\u6cd5\uff0c\u4ece\u800c\u5bf9\u5176\u8fdb\u884c\u5fae\u8c03\u5e76\u5c06\u5176\u8f6c\u6362\u4e3a\u6587\u672c\u5206\u7c7b\u5668\u3002.<\/p>\n<ul class=\"\">\n<li id=\"73f0\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u6b65\u9aa4 1\uff1a\u5728\u5927\u91cf\u76ee\u6807\u8bed\u8a00 data \u8bed\u6599\u5e93\u4e0a\u8bad\u7ec3\u901a\u7528\u8bed\u8a00\u6a21\u578b\u3002\u8be5\u6a21\u578b\u5c06\u80fd\u591f\u7406\u89e3\u8bed\u8a00\u7ed3\u6784\u3001\u8bed\u6cd5\u548c\u4e3b\u8981\u8bcd\u6c47\u3002.<\/li>\n<li id=\"cd1a\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u7b2c 2 \u6b65\uff1a\u6839\u636e\u5206\u7c7b\u8bad\u7ec3 data \u5fae\u8c03\u901a\u7528\u8bed\u8a00\u6a21\u578b\u3002\u8fd9\u6837\uff0c\u60a8\u7684\u6a21\u578b\u5c31\u80fd\u66f4\u597d\u5730\u5b66\u4e60\u5982\u4f55\u8868\u73b0\u8bad\u7ec3\u8bed\u6599\u4e2d\u4f7f\u7528\u7684\u8bcd\u6c47\u3002.<\/li>\n<li id=\"797c\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u7b2c 3 \u6b65\uff1a\u4f7f\u7528\u7ecf\u8fc7\u5fae\u8c03\u7684\u9884\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u8bad\u7ec3\u6587\u672c\u5206\u7c7b\u5668\u3002\u8fd9\u79cd\u65b9\u6cd5\u53ef\u4ee5\u8ba9\u6a21\u578b\u5728\u4e0a\u4e0b\u6587\u4e2d\u7406\u89e3\u5355\u8bcd\u3002\u6b64\u5916\uff0c\u4f7f\u7528\u9884\u8bad\u7ec3\u7684\u8bed\u8a00\u6a21\u578b\uff0c\u60a8\u53ef\u4ee5\u5728\u6781\u5c11\u7684\u8bad\u7ec3\u793a\u4f8b\u4e0a\u8bad\u7ec3\u5206\u7c7b\u5668\uff08\u6bcf\u4e2a\u6807\u7b7e\u53ea\u9700\u8bad\u7ec3 400 \u4e2a\u6587\u672c\u5373\u53ef\uff09\u3002.<\/li>\n<\/ul>\n<p id=\"819b\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u6211\u4eec\u5df2\u7ecf\u77e5\u9053\uff0c\u6587\u672c\u5206\u7c7b\u5728\u82f1\u8bed\u3001\u6cd5\u8bed\u3001\u5fb7\u8bed\u3001\u897f\u73ed\u7259\u8bed\u3001\u4e2d\u6587\u7b49\u8bed\u8a00\u4e2d\u90fd\u80fd\u5f88\u597d\u5730\u53d1\u6325\u4f5c\u7528\uff0c\u4f46\u5bf9\u4e8e\u90a3\u4e9b\u73b0\u6210\u8bed\u8a00\u6a21\u578b\u5f88\u5c11\u7684\u8bed\u8a00\uff0c\u6211\u4eec\u8be5\u600e\u4e48\u529e\u5462\uff1f<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u5728\u4e86\u89e3\u66f4\u591a\u7ec6\u8282\u4e4b\u524d\uff0c\u60a8\u53ef\u80fd\u60f3\u77e5\u9053\u4e3a\u4ec0\u4e48\u50cf\u6211\u8fd9\u6837\u7684\u6cd5\u56fd data \u79d1\u5b66\u5bb6\u4f1a\u60f3\u8981\u4e00\u4e2a\u97e9\u8bed\u6587\u672c\u5206\u7c7b\u5668\uff1f\u539f\u56e0\u662f\u6211\u53c2\u4e0e\u4e86\u4e00\u4e2a\u9879\u76ee\uff0c\u8be5\u9879\u76ee\u5f00\u53d1\u4e86\u4e00\u79cd\u5c06\u793e\u4ea4\u5a92\u4f53\u5e16\u5b50\u5206\u4e3a\u4e0d\u540c\u7c7b\u522b\u7684\u4ea7\u54c1\u3002\u5728\u5bf9\u82f1\u8bed\u548c\u6cd5\u8bed\u7684\u65b9\u6cd5\u8bba\u8fdb\u884c\u9a8c\u8bc1\u540e\uff0c\u6211\u4eec\u5f00\u59cb\u5c06\u5176\u6269\u5c55\u5230\u5176\u4ed6\u8bed\u8a00\uff08\u82f1\u8bed\u3001\u6cd5\u8bed\u3001\u65e5\u8bed\u3001\u4e2d\u6587\u548c\u97e9\u8bed\uff09\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u53ea\u6709\u97e9\u8bed\u9762\u4e34\u66f4\u5927\u7684\u6311\u6218\uff0c\u56e0\u4e3a\u5728\u5f00\u653e\u6e90\u4ee3\u7801\u4e2d\u627e\u4e0d\u5230\u9884\u8bad\u7ec3\u7684\u8bed\u8a00\u6a21\u578b\uff0c\u6240\u4ee5\u6211\u4e0d\u5f97\u4e0d\u5728\u97e9\u8bed\u8bed\u8a00\u8d44\u6e90\u6781\u5c11\u7684\u60c5\u51b5\u4e0b\u81ea\u5df1\u52a8\u624b\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u672c\u6587\u5c06\u91cd\u70b9\u4ecb\u7ecd\u4f7f\u7528\u591a\u91cd\u62df\u5408\u6cd5\u5bf9\u97e9\u6587\u6587\u672c\u8fdb\u884c\u5206\u7c7b\u7684\u65b9\u6cd5\uff0c\u5177\u4f53\u8bf4\u660e\u5982\u4e0b <a class=\"ck ju\" href=\"https:\/\/arxiv.org\/abs\/1909.04761\" target=\"_blank\" rel=\"noopener nofollow noreferrer\">\u7eb8\u5f20<\/a>.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u5f88\u591a\u8bed\u8a00\u5728\u7f51\u7edc\u4e2d\u90fd\u6709\u5f88\u5f3a\u7684\u4ee3\u8868\u6027\uff0c\u4f8b\u5982\uff1a\u82f1\u8bed\u3001\u4e2d\u6587\u3001\u897f\u73ed\u7259\u8bed\u3001\u8461\u8404\u7259\u8bed\u3001\u6cd5\u8bed......\u97e9\u8bed\u7684\u8bb0\u5f55\u4ecd\u7136\u975e\u5e38\u5c11\uff0c\u800c\u4e14\u6ca1\u6709\u5f88\u591a\u5185\u5bb9\u53ef\u4ee5\u91cd\u590d\u4f7f\u7528\uff1a\u97e9\u8bed\u7684\u8bb0\u5f55\u4ecd\u7136\u975e\u5e38\u5c11\uff0c\u4e5f\u6ca1\u6709\u5f88\u591a\u5185\u5bb9\u53ef\u4ee5\u91cd\u590d\u4f7f\u7528\u3002\u56e0\u6b64\uff0c\u6211\u60f3\u8d21\u732e\u81ea\u5df1\u7684\u529b\u91cf\uff0c\u4e0e\u5927\u5bb6\u5206\u4eab\u6211\u5728\u63a2\u7d22\u97e9\u8bed NLP \u8fc7\u7a0b\u4e2d\u7684\u4e3b\u8981\u5fc3\u5f97\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u5728\u672c\u6587\u4e2d\uff0c\u6211\u5c06\u5411\u5927\u5bb6\u8bb2\u8ff0\u6211\u5728\u4e0d\u61c2\u4e00\u4e2a\u97e9\u8bed\u5355\u8bcd\u7684\u60c5\u51b5\u4e0b\u8bad\u7ec3\u97e9\u8bed\u6a21\u578b\u7684\u5386\u7a0b\uff0c\u4ee5\u53ca\u5982\u4f55\u5c06\u5176\u7528\u4e8e\u6587\u672c\u5206\u7c7b\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\"><strong class=\"ik jv\">\u514d\u8d23\u58f0\u660e\u00a0<\/strong>\u901a\u5e38\uff0c\u5f53\u4e00\u4e2a\u8bed\u8a00\u6a21\u578b\u7684\u51c6\u786e\u7387\u8fbe\u5230 45-50% \u5de6\u53f3\u65f6\uff0c\u6211\u4eec\u5c31\u8ba4\u4e3a\u8fd9\u4e2a\u6a21\u578b\u5f88\u597d\u3002\u7531\u4e8e\u6211\u7684\u76ee\u6807\u4e0d\u662f\u751f\u6210\u97e9\u6587\u6587\u672c\uff0c\u6240\u4ee5\u6211\u4e0d\u9700\u8981\u8fbe\u5230\u8fd9\u6837\u7684\u6027\u80fd\uff1a\u6211\u53ea\u9700\u8981\u4e00\u4e2a\u80fd \u201c\u7406\u89e3 \u201d\u97e9\u8bed\u8bed\u6cd5\u548c\u7ed3\u6784\u7684\u6a21\u578b\uff0c\u8fd9\u6837\u6211\u5c31\u53ef\u4ee5\u7528\u5b83\u6765\u8bad\u7ec3\u97e9\u8bed\u6587\u672c\u5206\u7c7b\u5668\u3002.<\/p>\n<\/div><\/div><\/div><\/div><\/div><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-3 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-background-color:#ffffff;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-3 fusion_builder_column_1_1 1_1 fusion-flex-column\" style=\"--awb-bg-color:#ffffff;--awb-bg-color-hover:#ffffff;--awb-bg-size:cover;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-text fusion-text-4\" style=\"--awb-text-color:#000000;\"><p id=\"9349\" class=\"hk hl dn ce hm hn ho hp hq hr hs ht hu hv hw hx hy hz ia ib ic id ie if ig ih ek\"><span style=\"color: #ff0066;\"><strong>1 - \u7528\u4e8e\u8bed\u8a00\u6a21\u578b\u8bad\u7ec3\u7684 Data \u6570\u636e\u96c6<\/strong><\/span><\/p>\n<p class=\"hk hl dn ce hm hn ho hp hq hr hs ht hu hv hw hx hy hz ia ib ic id ie if ig ih ek\"><span style=\"color: #ff0066;\"><strong>1.1 - Data \u4fe1\u53f7\u6e90<\/strong><\/span><\/p>\n<p id=\"fb69\" class=\"ii ij dn ik b il im hp in io ip ht iq ir is it iu iv iw ix iy iz ja jb jc jd df ek\">\u901a\u5e38\uff0c\u5728\u4ece\u5934\u5f00\u59cb\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u65f6\uff0cULM FiT \u6559\u7a0b\u5efa\u8bae\u4e0b\u8f7d\u7ed9\u5b9a\u8bed\u8a00\u7684\u6240\u6709\u7ef4\u57fa\u767e\u79d1\u5185\u5bb9\u3002\u53ea\u6709\u5728\u8be5\u8bed\u8a00\u7684\u6bcd\u8bed\u4f7f\u7528\u8005\u4e60\u60ef\u4e8e\u5728\u8be5\u9891\u9053\u53d1\u5e03\u5927\u91cf\u5185\u5bb9\u7684\u60c5\u51b5\u4e0b\uff0c\u8fd9\u4e9b\u5efa\u8bae\u624d\u4f1a\u6709\u6548\u3002.<\/p>\n<p class=\"ii ij dn ik b il im hp in io ip ht iq ir is it iu iv iw ix iy iz ja jb jc jd df ek\">\u5728\u97e9\u8bed\u4e2d\uff0c\u4eba\u4eec\u4f3c\u4e4e\u5e76\u4e0d\u4e60\u60ef\uff1a\u4e0d\u4ec5\u7ef4\u57fa\u767e\u79d1\u97e9\u8bed\u8bed\u5883\u7684\u97f3\u91cf\u4e0d\u591f\uff0c\u800c\u4e14\u4e5f\u4e0d\u80fd\u4ee3\u8868\u4ee5\u97e9\u8bed\u4e3a\u6bcd\u8bed\u7684\u4eba\u3002.<\/p>\n<p class=\"ii ij dn ik b il im hp in io ip ht iq ir is it iu iv iw ix iy iz ja jb jc jd df ek\">\u4e0b\u9762\u662f\u82f1\u6587\u7ef4\u57fa\u767e\u79d1\u548c\u97e9\u6587\u7ef4\u57fa\u767e\u79d1\u7684\u6587\u7ae0\u6570\u91cf\u5bf9\u6bd4\uff0c\u4ee5\u63d0\u4f9b\u4e00\u4e9b\u63d0\u793a\uff1a<\/p>\n<div class=\"cx cy jw\">\n<div class=\"hd s gv he\">\n<div class=\"jx hg s\" style=\"text-align: center;\">\n<div class=\"gy gz t u v ha aj bm hb hc\"><\/div>\n<p><img decoding=\"async\" class=\"lazyload tq tr t u v ha aj c aligncenter\" src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/1-Lj8fQGUHUKcgH-CYxd8Uw.png\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/1-Lj8fQGUHUKcgH-CYxd8Uw.png\" data-sizes=\"auto\" data-orig-sizes=\"611px\" srcset=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%27611%27%20height%3D%2795%27%20viewBox%3D%270%200%20611%2095%27%3E%3Crect%20width%3D%27611%27%20height%3D%2795%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-srcset=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/1-Lj8fQGUHUKcgH-CYxd8Uw.png 276w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/1-Lj8fQGUHUKcgH-CYxd8Uw.png 552w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/1-Lj8fQGUHUKcgH-CYxd8Uw.png 611w\" alt=\"Image for post\" width=\"611\" height=\"95\" \/><\/p>\n<figure class=\"gn go gp gq gr gs cx cy paragraph-image\"><figcaption class=\"jy jz cz cx cy ka kb ce b eu cg fx\" data-selectable-paragraph=\"\"><em>\u4e0d\u540c\u8bed\u8a00\u7684\u7ef4\u57fa\u767e\u79d1\u5377<\/em><\/figcaption><\/figure>\n<\/div>\n<\/div>\n<div><\/div>\n<\/div>\n<p>\u6211\u7684\u5efa\u8bae\u6211\u5c06\u7ef4\u57fa\u767e\u79d1\u6587\u7ae0\u4e0e Common Crawl data \u7ed3\u5408\u5728\u4e00\u8d77\uff0c\u60a8\u53ef\u4ee5\u4ece\u4ee5\u4e0b\u7f51\u5740\u4e0b\u8f7d <a class=\"ck ju\" href=\"https:\/\/traces1.inria.fr\/oscar\/\" target=\"_blank\" rel=\"noopener nofollow noreferrer\">\u8fd9\u91cc<\/a>.<\/p>\n<p><strong style=\"color: #ff0066;\">1.2 - Data \u4f53\u79ef<\/strong><\/p>\n<p>\u8ba9\u6211\u4eec\u8bb0\u4f4f\uff0c\u8bed\u8a00\u6a21\u578b\u662f\u4e00\u4e2a\u53ef\u4ee5\u9884\u6d4b\u6587\u672c\u4e2d\u4e0b\u4e00\u4e2a\u5355\u8bcd\u7684\u6a21\u578b\u3002\u8981\u505a\u5230\u8fd9\u4e00\u70b9\uff0c\u6211\u4eec\u7684\u6a21\u578b\u5e94\u8be5\u770b\u8fc7\u5927\u91cf\u7684\u4f8b\u5b50\u6765\u5b66\u4e60\u8bed\u8a00\uff0c\u5e76\u64c5\u957f\u8bf4\u8fd9\u79cd\u8bed\u8a00\u3002\u4e5f\u5c31\u662f\u8bf4\uff0c\u8d85\u8fc7 1 \u4ebf\u4e2a\u8bcd\u5e93\u662f\u6ca1\u6709\u7528\u7684\u3002\u8fd9\u53ea\u4f1a\u589e\u52a0\u6a21\u578b\u7684\u590d\u6742\u6027\u548c\u5927\u91cf\u7684\u8bad\u7ec3\u65f6\u95f4\u3002.<\/p>\n<p>\u56e0\u6b64\uff0c\u4e4d\u4e00\u770b\uff0c\u5f53\u6211\u68c0\u7d22\u4e86\u6240\u6709\u7ef4\u57fa\u767e\u79d1\u548c Common Crawl data \u540e\uff0c\u6211\u53d1\u73b0\u81ea\u5df1\u62e5\u6709\u4e86\u8d85\u8fc7 1 \u4ebf\u4e2a\u6807\u8bb0\uff0c\u56e0\u6b64\u6211\u5fc5\u987b\u6311\u9009\u6700\u76f8\u5173\u7684\u6587\u6863\u6765\u8bad\u7ec3\u6211\u7684\u6a21\u578b\u3002\u6211\u7684\u65b9\u6cd5\u7684\u76ee\u6807\u662f\u4fdd\u7559\u6700\u80fd\u4ee3\u8868\u97e9\u8bed\u6bcd\u8bed\u7684\u6587\u6863\uff1a<\/p>\n<ol class=\"\">\n<li id=\"2e13\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd kc jk jl ek\" data-selectable-paragraph=\"\">\u6211\u9996\u5148\u5bf9\u8bed\u6599\u5e93\u8fdb\u884c\u4e86\u5f31\u6807\u8bb0\u5316\u5904\u7406\uff0c\u901a\u8fc7\u5206\u5272\u8bed\u6599\u5e93\u7684\u7a7a\u683c\u6765\u4f30\u7b97\u6807\u8bb0\u7684\u6570\u91cf\u3002.<\/li>\n<li id=\"f246\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd kc jk jl ek\" data-selectable-paragraph=\"\">\u6211\u4ece\u83b7\u5f97\u7684\u4ee3\u5e01\u4e2d\u5220\u9664\u4e86\u6240\u6709\u6570\u5b57\u3001\u8868\u60c5\u7b26\u53f7\u3001\u6807\u70b9\u7b26\u53f7\u548c\u5176\u4ed6\u975e\u97e9\u8bed\u7279\u6709\u7684\u7b26\u53f7\u3002.<\/li>\n<li id=\"7991\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd kc jk jl ek\" data-selectable-paragraph=\"\">\u6211\u8ba1\u7b97\u4e86\u8bed\u6599\u5e93\u4e2d\u6240\u6709\u6807\u8bb0\u7684\u8ba1\u6570\u5668\uff0c\u5e76\u68c0\u7d22\u4e86\u63d0\u53ca\u6b21\u6570\u6700\u591a\u7684 7 \u4e07\u4e2a\u6807\u8bb0\u3002.<\/li>\n<li id=\"bc31\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd kc jk jl ek\" data-selectable-paragraph=\"\">\u7136\u540e\uff0c\u6211\u68c0\u7d22\u4e86\u63d0\u5230\u5927\u591a\u6570\u5e38\u7528\u6807\u8bb0\u7684\u6587\u6863\uff0c\u8fd9\u6837\u6211\u7684\u8bed\u6599\u5e93\u5c31\u5305\u542b\u4e86 1 \u4ebf\u4e2a\u6807\u8bb0\uff0c\u8fd9\u5c31\u662f\u6211\u7684\u8bad\u7ec3\u8bed\u6599\u5e93\uff01<\/li>\n<\/ol>\n<p id=\"c754\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u73b0\u5728\uff0c\u6211\u4eec\u6709\u4e86\u57f9\u8bad\u7684\u539f\u59cb\u8d44\u6599\u5e93\uff0c\u5c31\u53ef\u4ee5\u5f00\u59cb\u771f\u6b63\u7684\u4e1a\u52a1\u4e86\uff01<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\"><span style=\"color: #ff0066;\"><strong>2 - Data \u6807\u8bb0\u5316<\/strong><\/span><\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u6211\u60f3\uff0c\u5f53\u6211\u4e4b\u524d\u544a\u8bc9\u4f60\u6211\u7528\u62c6\u5206\u51fd\u6570\u8fdb\u884c\u4e86\u6807\u8bb0\u5316\u5904\u7406\u65f6\uff0c\u4f60\u4f1a\u5f00\u59cb\u89c9\u5f97\u8fd9\u7bc7\u6587\u7ae0\u771f\u7684\u662f\u4e2a\u7b11\u8bdd\uff0c\u4f46\u8ba9\u6211\u4eec\u5411\u4f60\u4fdd\u8bc1\uff0c\u8fd9\u7edd\u4e0d\u662f\u6211\u7684\u6700\u7ec8\u76ee\u7684\uff01<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u9996\u5148\u8981\u63d0\u9192\u5927\u5bb6\u7684\u662f\uff0c\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u4e0d\u9700\u8981\u8fdb\u4e00\u6b65\u7684 data \u9884\u5904\u7406\u3002\u5f88\u591a NLP \u4efb\u52a1\u90fd\u4f1a\u5bf9\u6587\u672c\u8fdb\u884c\u4e00\u4e9b\u5265\u79bb\u5904\u7406\uff0c\u5982\u6570\u5b57\u3001\u5355\u8bcd\u3001\u5c0f\u5199\u3001\u8bcd\u5e72......\u6240\u6709\u8fd9\u4e9b\u90fd\u4f1a\u5c06\u6587\u672c\u4ece\u4e0a\u4e0b\u6587\u4e2d\u5265\u79bb\u51fa\u6765\uff0c\u800c\u6211\u4eec\u7684\u76ee\u6807\u662f\u5b66\u4e60\u8bf4\u97e9\u8bed\uff0c\u56e0\u6b64\u6211\u4eec\u5fc5\u987b\u4fdd\u6301\u6240\u6709\u6587\u672c\u7684\u539f\u8c8c\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u4e3a\u4e86\u6807\u8bb0\u97e9\u8bed\u6587\u672c\uff0c\u6211\u5c1d\u8bd5\u4e86\u4e24\u79cd\u6807\u8bb0\u6a21\u578b\uff1a<\/p>\n<ul class=\"\">\n<li id=\"d36a\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\"><a class=\"ck ju\" href=\"https:\/\/github.com\/explosion\/spaCy\/tree\/master\/spacy\/lang\/ko\" rel=\"noopener nofollow\" target=\"_blank\">\u97e9\u56fd\u7a7a\u95f4\u6a21\u578b<\/a> \u662f\u97e9\u56fd mecab \u4ee4\u724c\u751f\u6210\u5668\u7684\u4e00\u4e2a\u5c01\u88c5\u3002.<\/li>\n<li id=\"3ebc\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\"><a class=\"ck ju\" href=\"https:\/\/github.com\/google\/sentencepiece\" rel=\"noopener nofollow\" target=\"_blank\">\u53e5\u5b50<\/a>\u00a0\u5728\u6211\u7684\u8bed\u6599\u5e93\u4e2d\u8bad\u7ec3\u7684 subwords \u6807\u8bb0\u7b26\u53f7\u6a21\u578b\uff0c\u6700\u5927\u6807\u8bb0\u7b26\u53f7\u6570\u4e3a 28000 \u4e2a<\/li>\n<\/ul>\n<p id=\"510f\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u6b63\u5982\u00a0<a class=\"ck ju\" href=\"https:\/\/nlp.fast.ai\/classification\/2019\/09\/10\/multifit.html\" rel=\"noopener nofollow\" target=\"_blank\">\u591a\u529f\u80fd\u6587\u7ae0<\/a>, \u56e0\u6b64\uff0c\u6211\u9009\u62e9\u4e86\u7b2c\u4e8c\u79cd\u65b9\u6848\uff0c\u4ee5\u83b7\u5f97\u5b50\u5b57\u7c92\u5ea6\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\"><strong><span style=\"color: #ff0066;\">3 - \u57f9\u8bad\u6a21\u5f0f<\/span><\/strong><\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u5728\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u548c\u8bad\u7ec3\u4efb\u4f55\u6a21\u578b\u65f6\uff0c\u90fd\u8981\u907f\u514d\u4ee5\u4e0b\u4e24\u70b9\u00a0<strong class=\"ik jv\">\u6b20\u59a5<\/strong>\u00a0\u548c\u00a0<strong class=\"ik jv\">\u8fc7\u5ea6\u62df\u5408\u3002.<\/strong><\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u4e00\u4e2a\u6a21\u578b\u00a0<strong class=\"ik jv\">\u4e0d\u9002\u5408<\/strong> \u5f53\u5b83\u8bd5\u56fe\u6a21\u62df\u7684 data \u8fc7\u4e8e\u7b80\u5355\u65f6\u3002\u5982\u679c\u60a8\u53d1\u73b0\u6a21\u578b\u65e0\u6cd5\u5728\u8bad\u7ec3 data \u4e0a\u5b66\u4e60\uff0c\u800c\u4e14\u8bad\u7ec3\u635f\u5931\u5b8c\u5168\u6ca1\u6709\u6536\u655b\u5230 0\uff0c\u60a8\u5c31\u53ef\u4ee5\u53d1\u73b0\u8fd9\u4e00\u70b9\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u76f8\u53cd\uff0c\u4e00\u4e2a\u6a21\u578b\u00a0<strong class=\"ik jv\">\u8d85\u8fc7\u9002\u5408<\/strong>\u00a0\u5f53\u5b83\u5b66\u4e60\u5f97 \u201c\u592a\u597d\u201d\uff0c\u53ef\u4ee5\u5bf9\u8bad\u7ec3 data \u8fdb\u884c\u5efa\u6a21\uff0c\u4f46\u5728\u6d4b\u8bd5 data \u4e0a\u7684\u6027\u80fd\u4ecd\u7136\u5f88\u4f4e\u65f6\u3002\u8fd9\u8868\u660e\u60a8\u7684\u6a21\u578b\u4e0d\u53ef\u80fd\u5f88\u597d\u5730\u9884\u6d4b\u5b83\u672a\u89c1\u8fc7\u7684 data\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u5f53\u6211\u5f00\u59cb\u8bad\u7ec3\u6211\u7684\u8bed\u8a00\u6a21\u578b\u65f6\uff0c\u4e00\u5f00\u59cb\u6211\u771f\u7684\u5f88\u96be\u4ece\u6211\u7684 data \u4e0a\u5b66\u5230\u4efb\u4f55\u4e1c\u897f\u3002\u5982\u4e0b\u56fe\u6240\u793a\uff0c\u7ecf\u8fc7 10 \u4e2a\u5386\u5143\u7684\u8bad\u7ec3\u540e\uff0c\u6211\u7684\u8bad\u7ec3\u635f\u5931\u5bf8\u6b65\u96be\u884c\u3002.<\/p>\n<div class=\"gt gu gv gw aj gx\" tabindex=\"0\" role=\"button\">\n<div class=\"cx cy kd\">\n<div class=\"hd s gv he\">\n<div class=\"ke hg s\">\n<p><img decoding=\"async\" class=\"lazyload tq tr t u v ha aj c aligncenter\" src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0IU-CQKAbqWW_heiB.png\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0IU-CQKAbqWW_heiB.png\" data-sizes=\"auto\" data-orig-sizes=\"700px\" srcset=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%27716%27%20height%3D%27432%27%20viewBox%3D%270%200%20716%20432%27%3E%3Crect%20width%3D%27716%27%20height%3D%27432%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-srcset=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0IU-CQKAbqWW_heiB.png 276w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0IU-CQKAbqWW_heiB.png 552w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0IU-CQKAbqWW_heiB.png 640w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/09\/0IU-CQKAbqWW_heiB.png 700w\" alt=\"Image for post\" width=\"716\" height=\"432\" \/><\/p>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<p id=\"7409\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u8fd9\u8bf4\u660e\u6211\u7684\u6a21\u578b\u8fc7\u4e8e\u7b80\u5355\uff0c\u65e0\u6cd5\u8868\u73b0\u97e9\u8bed\u7684\u590d\u6742\u6027\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u6211\u662f\u8fd9\u6837\u89e3\u51b3\u8fd9\u4e2a\u95ee\u9898\u7684\uff1a<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u53ef\u4ee5\u60f3\u8c61\uff0c\u8c03\u8bd5\u4efb\u4f55\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\u90fd\u4e0d\u662f\u4e00\u4ef6\u5bb9\u6613\u7684\u4e8b\uff0c\u56e0\u4e3a\u5176\u4e2d\u6709\u592a\u591a\u7684\u81ea\u7531\u5ea6\u3002\u4f60\u5fc5\u987b\u627e\u5230\u6b63\u786e\u7684\u7f51\u7edc\u7ed3\u6784\u4ee5\u53ca\u6b63\u786e\u7684\u8d85\u53c2\u6570\u96c6\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u4e3a\u4e86\u5728\u4e00\u5f00\u59cb\u5c31\u7b80\u5316\u95ee\u9898\uff0c\u6b63\u786e\u7684\u65b9\u6cd5\u662f\u5c1d\u8bd5\u5bf9\u5355\u6279 data \u8fdb\u884c\u8d85\u62df\u5408\u3002\u8fd9\u6837\u505a\u7684\u76ee\u7684\u662f\u786e\u4fdd\u5728\u7ed9\u5b9a\u67d0\u4e2a data \u7684\u60c5\u51b5\u4e0b\uff0c\u60a8\u7684\u6a21\u578b\u80fd\u591f\u89e3\u91ca\u5176\u590d\u6742\u6027\uff0c\u5e76\u5728\u8bad\u7ec3\u96c6\u4e0a\u8868\u73b0\u826f\u597d\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u4ee5\u4e0b\u662f\u6211\u5c1d\u8bd5\u8fc7\u7684\u6240\u6709\u65b9\u6cd5\uff1a<\/p>\n<ul class=\"\">\n<li id=\"46bd\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u589e\u5927\u5d4c\u5165\u5c3a\u5bf8<\/li>\n<li id=\"4727\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u589e\u52a0\u9690\u85cf\u5c42\u6570<\/li>\n<li id=\"a0be\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u66f4\u6539\u4f18\u5316\u5668\u529f\u80fd<\/li>\n<li id=\"085d\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u6539\u53d8\u5b66\u4e60\u901f\u5ea6<\/li>\n<\/ul>\n<p id=\"9594\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u7ecf\u8fc7\u591a\u6b21\u5c1d\u8bd5\uff0c\u4ee5\u4e0b\u662f\u8ba9\u6211\u7684\u6a21\u578b\u5f00\u59cb\u5b66\u4e60\u7684\u7ed3\u6784\u548c\u8d85\u53c2\u6570\uff1a<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u795e\u7ecf\u7f51\u7edc\u67b6\u6784<\/p>\n<ul class=\"\">\n<li id=\"3703\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">QRNN \u7ed3\u6784<\/li>\n<li id=\"a89d\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u9690\u85cf\u5c42\u6570\uff1a 2500<\/li>\n<li id=\"69bf\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u5c42\u6570\uff1a4<\/li>\n<li id=\"3ea6\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u5d4c\u5165\u5c3a\u5bf8\uff1a768<\/li>\n<\/ul>\n<p id=\"9f79\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u4e00\u65e6\u6a21\u578b\u80fd\u591f\u5728\u8bad\u7ec3\u96c6\u4e0a\u6b63\u786e\u9884\u6d4b\uff0c\u63a5\u4e0b\u6765\u8981\u907f\u514d\u7684\u5c31\u662f\u8fc7\u5ea6\u62df\u5408\u3002.<\/p>\n<p class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u4ee5\u4e0b\u662f\u6211\u5c1d\u8bd5\u7684\u4e00\u4e9b\u6b63\u5219\u5316\u65b9\u6cd5\uff0c\u4ee5\u786e\u4fdd\u6211\u7684\u6a21\u578b\u4e0d\u4f1a\u8fc7\u62df\u5408\u3002.<\/p>\n<ul class=\"\">\n<li id=\"191d\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u6dfb\u52a0\u8f8d\u5b66<\/li>\n<li id=\"7c3e\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u589e\u52a0\u91cd\u91cf\u8870\u51cf<\/li>\n<li id=\"0000\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u6dfb\u52a0\u6e10\u53d8\u526a\u5207<\/li>\n<\/ul>\n<p id=\"e1c1\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd df ek\">\u4ee5\u4e0b\u662f\u6211\u5728\u8bad\u7ec3\u6a21\u578b\u65f6\u4f7f\u7528\u7684\u6b63\u5219\u8868\u8fbe\u5f0f\uff1a<\/p>\n<ul class=\"\">\n<li id=\"fe77\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u5b66\u4e60\u73870.0002<\/li>\n<li id=\"a072\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u91cd\u91cf\u8870\u51cf1e-8<\/li>\n<li id=\"bc79\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u6e10\u53d8\u526a\u52070.25<\/li>\n<\/ul>\n<\/div><\/div><\/div><\/div><\/div><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-4 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-background-color:#ffffff;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-4 fusion_builder_column_1_1 1_1 fusion-flex-column\" style=\"--awb-bg-color:#ffffff;--awb-bg-color-hover:#ffffff;--awb-bg-size:cover;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-text fusion-text-5\" style=\"--awb-text-color:#000000;\"><p><span style=\"color: #ff0066;\"><strong>\u6210\u679c<\/strong><\/span><\/p>\n<p>\u5728\u5bf9\u6a21\u578b\u8fdb\u884c\u4e86 15 \u6b21\u5386\u65f6\u8bad\u7ec3\u540e\uff0c\u6211\u6700\u7ec8\u83b7\u5f97\u4e86 25% \u7684\u51c6\u786e\u7387\u548c 100 \u7684\u56f0\u60d1\u5ea6\u3002\u6b63\u5982\u6211\u5728\u4e00\u5f00\u59cb\u6240\u8bf4\u7684\uff0c\u6211\u4ece\u672a\u6253\u7b97\u7528\u6211\u7684\u8bed\u8a00\u6a21\u578b\u6765\u751f\u6210\u6587\u672c\uff0c\u6240\u4ee5\u6211\u5df2\u7ecf\u5f88\u6ee1\u610f\u6211\u7684\u6a21\u578b\u80fd\u591f\u6b63\u786e\u9884\u6d4b 4 \u4e2a\u5355\u8bcd\u4e2d\u7684 1 \u4e2a\u3002.<\/p>\n<p>\u7136\u540e\uff0c\u6211\u5c06\u9884\u5148\u8bad\u7ec3\u597d\u7684\u6a21\u578b\u91cd\u65b0\u7528\u4e8e\u6587\u672c\u5206\u7c7b\u3002\u6211\u4f7f\u7528\u7684 dataset \u662f\u4e00\u4e2a\u5e73\u8861\u7684 dataset\uff0c\u7531\u6765\u81ea Instagram\u3001Facebook\u3001Youtube \u548c\u7f51\u7ad9\u7684 10k \u7bc7\u793e\u4ea4\u6587\u6863\u7ec4\u6210\uff0c\u8fd9\u4e9b\u6587\u6863\u88ab\u6807\u8bb0\u4e3a \u201clabel1 \u201d\u6216\u975e \u201cnot label1\u201d\u3002\u6211\u7684\u76ee\u6807\u662f\u9884\u6d4b\u65b0\u51fa\u7248\u7269\u662f\u5426\u4e0e \u201clabel1 \u201d\u6709\u5173\u3002.<\/p>\n<p>\u4ee5\u4e0b\u662f\u6211\u4eec\u5f00\u53d1\u7684\u6240\u6709\u8bed\u8a00\u7684\u6027\u80fd\uff1a<\/p>\n<p>\u5e16\u5b50\u56fe\u7247<\/p>\n<p>\u4e0d\u540c\u8bed\u8a00\u6587\u672c\u5206\u7c7b\u5668\u7684\u6027\u80fd<\/p>\n<p>\u56e0\u6b64\uff0c\u5373\u4f7f\u4e0d\u8bf4\u97e9\u8bed\uff0c\u4e0d\u4eb2\u81ea\u8bad\u7ec3\u9884\u8bad\u7ec3\u7684\u8bed\u8a00\u6a21\u578b\uff0c\u97e9\u8bed\u6587\u672c\u5206\u7c7b\u5668\u7684\u6027\u80fd\u4e5f\u80fd\u5f88\u597d\u5730\u8fbe\u5230\u5176\u4ed6\u8bed\u8a00\u7684\u6027\u80fd\u3002.<\/p>\n<p>\u867d\u7136\u6211\u8fd8\u6709\u5f88\u591a\u5730\u65b9\u9700\u8981\u6539\u8fdb\uff0c\u4f46\u5728\u5b8c\u5168\u4e0d\u61c2\u97e9\u8bed\u7684\u60c5\u51b5\u4e0b\uff0c\u5728\u7f51\u4e0a\u627e\u4e0d\u5230\u76f8\u5173\u4fe1\u606f\u548c\u5efa\u8bae\u7684\u60c5\u51b5\u4e0b\uff0c\u5b66\u4e60\u5904\u7406\u97e9\u8bed\u8fd9\u79cd\u590d\u6742\u8bed\u8a00\u7684\u6587\u4ef6\uff0c\u8fd8\u662f\u8ba9\u6211\u53d7\u5ba0\u82e5\u60ca\u3002.<\/p>\n<p><span style=\"color: #ff0066;\"><strong>\u4e0b\u4e00\u6b65\u5de5\u4f5c<\/strong><\/span><\/p>\n<p>\u6211\u521a\u624d\u4ecb\u7ecd\u4e86\u5982\u4f55\u5229\u7528\u4ece\u96f6\u5f00\u59cb\u5efa\u7acb\u7684\u7b80\u5355\u8bed\u8a00\u6a21\u578b\u6765\u6539\u8fdb\u97e9\u8bed\u6587\u672c\u5206\u7c7b\u6a21\u578b\u3002\u6700\u521d\u7684\u6027\u80fd\u5df2\u7ecf\u4e0d\u9519\uff0c\u4f46\u8fd8\u6709\u6539\u8fdb\u7684\u4f59\u5730\u3002\u6211\u8ba4\u4e3a\u77ed\u671f\u5185\u6211\u60f3\u505a\u7684\u5de5\u4f5c\u662f\uff1a<\/p>\n<ul class=\"\">\n<li id=\"11e5\" class=\"ii ij dn ik b il je hp in io jf ht iq ir jg it iu iv jh ix iy iz ji jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u6821\u5bf9\u6807\u8bb0\u5316\uff1a\u7531\u4e8e\u6211\u4e00\u53e5\u97e9\u8bed\u90fd\u4e0d\u4f1a\u8bf4\uff0c\u6240\u4ee5\u8bf7\u4e00\u4f4d\u4ee5\u97e9\u8bed\u4e3a\u6bcd\u8bed\u7684\u4eba\u770b\u4e00\u4e0b\u6807\u8bb0\u5316\uff0c\u786e\u8ba4\u662f\u5426\u5408\u7406\u3002.<\/li>\n<li id=\"182a\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u901a\u8fc7\u4ee5\u4e0b\u65b9\u6cd5\u589e\u5f3a\u6211\u7684\u8bed\u8a00\u6a21\u578b\u5e76\u6bd4\u8f83\u5206\u7c7b\u6027\u80fd\uff1a<\/li>\n<li id=\"30ab\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u8fc1\u79fb\u5b66\u4e60\u540e\u5411\u8bed\u8a00\u6a21\u578b\uff0c\u56e0\u4e3a\u5b83\u5728\u82f1\u8bed\u6216\u6cd5\u8bed\u4e2d\u4f3c\u4e4e\u8868\u73b0\u66f4\u4f73\u3002.<\/li>\n<li id=\"0ae5\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u53cc\u5411\u8bed\u8a00\u6a21\u578b\u7684\u8fc1\u79fb\u5b66\u4e60.<\/li>\n<li id=\"281a\" class=\"ii ij dn ik b il jm hp in io jn ht iq ir jo it iu iv jp ix iy iz jq jb jc jd jj jk jl ek\" data-selectable-paragraph=\"\">\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u91c7\u7528\u52a8\u6001\u5b66\u4e60\u7387\uff0c\u907f\u514d\u9677\u5165\u5c40\u90e8\u6700\u5c0f\u503c\u3002.<\/li>\n<\/ul>\n<p><a href=\"https:\/\/medium.com\/artefact-engineering-and-data-science\" target=\"_blank\" rel=\"noopener noreferrer\">\u672c\u6587\u9996\u53d1\u4e8e The Artefact Tech Blog - \u5de5\u7a0b\u548c data \u79d1\u5b66\u6587\u7ae0\u5e93\u3002.<\/a><\/p>\n<\/div><\/div><\/div><\/div><\/div><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-5 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--link_hover_color: #ff0066;--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-padding-top:0px;--awb-padding-right:0px;--awb-padding-bottom:0px;--awb-padding-left:0px;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-5 fusion_builder_column_1_2 1_2 fusion-flex-column\" style=\"--awb-bg-size:cover;--awb-width-large:50%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:50%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-image-element\" style=\"text-align:center;--awb-margin-top:0px;--awb-margin-right:0px;--awb-margin-bottom:0px;--awb-margin-left:0px;--awb-caption-title-font-family:var(--h2_typography-font-family);--awb-caption-title-font-weight:var(--h2_typography-font-weight);--awb-caption-title-font-style:var(--h2_typography-font-style);--awb-caption-title-size:var(--h2_typography-font-size);--awb-caption-title-transform:var(--h2_typography-text-transform);--awb-caption-title-line-height:var(--h2_typography-line-height);--awb-caption-title-letter-spacing:var(--h2_typography-letter-spacing);\"><span class=\"fusion-imageframe imageframe-none imageframe-2 hover-type-none\"><img decoding=\"async\" width=\"1057\" height=\"591\" title=\"data-driven\" src=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%271057%27%20height%3D%27591%27%20viewBox%3D%270%200%201057%20591%27%3E%3Crect%20width%3D%271057%27%20height%3D%27591%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-orig-src=\"https:\/\/www.artefact.com\/wp-content\/uploads\/2020\/06\/Global-Newsletter-Data-Digest_LinkedIn_600x315_april.jpg\" alt class=\"lazyload img-responsive wp-image-30246\"\/><\/span><\/div><\/div><\/div><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-6 fusion_builder_column_1_2 1_2 fusion-flex-column fusion-flex-align-self-center fusion-column-inner-bg-wrapper\" style=\"--awb-inner-bg-size:cover;--awb-width-large:50%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:50%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><span class=\"fusion-column-inner-bg hover-type-none\"><a class=\"fusion-column-anchor\" href=\"https:\/\/go.pardot.com\/l\/597421\/2020-05-06\/2k5zzj\" target=\"_blank\" rel=\"noopener\"><span class=\"fusion-column-inner-bg-image\"><\/span><\/a><\/span><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-center fusion-content-layout-column\"><div class=\"fusion-text fusion-text-6\"><p><b>\u5bf9\u6570\u5b57\u6280\u672f\u548c data marketing \u611f\u5174\u8da3\uff1f<\/b><\/p>\n<p>\u6ce8\u518c Data Digest\uff08Artefact \u65f6\u4e8b\u901a\u8baf\uff09\uff0c\u6bcf\u6708\u5728\u60a8\u7684\u6536\u4ef6\u7bb1\u4e2d\u6536\u5230\u53ef\u884c\u7684\u5efa\u8bae\u3001\u89c1\u89e3\u548c\u610f\u89c1\u3002.<\/p>\n<p><b>\u7ed9\u6211\u62a5\u540d<\/b><\/p>\n<\/div><\/div><\/div><\/div><\/div><\/p>","protected":false},"excerpt":{"rendered":"<p>2020 \u5e74 11 \u6708 25 \u65e5<br \/>\n\u5728\u672c\u6587\u4e2d\uff0c\u6cd5\u56fd Artefact \u9ad8\u7ea7 Data \u79d1\u5b66\u5bb6 Amale El Hamri \u4ecb\u7ecd\u4e86\u5982\u4f55\u5728\u4e0d\u7406\u89e3\u8bed\u8a00\u7684\u60c5\u51b5\u4e0b\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u3002\u6587\u7ae0\u5185\u5bb9\u5305\u62ec\u4ece\u54ea\u91cc\u83b7\u53d6\u8bad\u7ec3 data\u3001\u9700\u8981\u591a\u5c11 data\u3001\u5982\u4f55\u9884\u5904\u7406 data\uff0c\u4ee5\u53ca\u5982\u4f55\u627e\u5230\u6700\u9002\u5408\u6a21\u578b\u7684\u67b6\u6784\u548c\u8d85\u53c2\u6570\u96c6\u3002.<\/p>","protected":false},"featured_media":21375,"parent":0,"template":"","meta":{"_acf_changed":false,"ep_exclude_from_search":false},"blog-category":[2995,21930],"blog-language":[2991],"class_list":["post-21374","blog","type-blog","status-publish","has-post-thumbnail","hentry","blog-category-ai-technology","blog-category-finance","blog-language-en"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.artefact.com\/zh\/wp-json\/wp\/v2\/blog\/21374","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.artefact.com\/zh\/wp-json\/wp\/v2\/blog"}],"about":[{"href":"https:\/\/www.artefact.com\/zh\/wp-json\/wp\/v2\/types\/blog"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.artefact.com\/zh\/wp-json\/wp\/v2\/media\/21375"}],"wp:attachment":[{"href":"https:\/\/www.artefact.com\/zh\/wp-json\/wp\/v2\/media?parent=21374"}],"wp:term":[{"taxonomy":"blog-category","embeddable":true,"href":"https:\/\/www.artefact.com\/zh\/wp-json\/wp\/v2\/blog-category?post=21374"},{"taxonomy":"blog-language","embeddable":true,"href":"https:\/\/www.artefact.com\/zh\/wp-json\/wp\/v2\/blog-language?post=21374"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}