{"id":12273,"date":"2025-01-28T19:30:30","date_gmt":"2025-01-28T10:30:30","guid":{"rendered":"https:\/\/highreso.jp\/edgehub\/?p=12273"},"modified":"2025-01-28T19:30:31","modified_gmt":"2025-01-28T10:30:31","slug":"llamaindexnodeparser","status":"publish","type":"post","link":"https:\/\/highreso.jp\/edgehub\/machinelearning\/llamaindexnodeparser.html","title":{"rendered":"LlamaIndex\u306e\u30c1\u30e3\u30f3\u30af\u5206\u5272\u30fbNode Parser\u306e\u4f7f\u3044\u65b9\uff01"},"content":{"rendered":"\n<p><span class=\"swl-format-2\">\u3053\u306e\u8a18\u4e8b\u3067\u306f\u3001LlamaIndex\u3067\u30c1\u30e3\u30f3\u30af\u5206\u5272\u3092\u3059\u308bNode Parser\u306e\u4f7f\u3044\u65b9\u306b\u3064\u3044\u3066\u89e3\u8aac\u3057\u3066\u3044\u307e\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u69cb\u6587\u30d9\u30fc\u30b9\u3001\u9577\u3055\u30d9\u30fc\u30b9\u3001\u610f\u5473\u30d9\u30fc\u30b9\u3001\u30ea\u30ec\u30fc\u30b7\u30e7\u30f3\u30d9\u30fc\u30b9\u306e\u305d\u308c\u305e\u308c\u306b\u5bfe\u5fdc\u3059\u308b\u30d1\u30fc\u30b5\u30fc\u3092\u7d39\u4ecb\u3057\u3001\u30c6\u30ad\u30b9\u30c8\u3084\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u306e\u52b9\u679c\u7684\u306a\u5206\u5272\u65b9\u6cd5\u3092\u89e3\u8aac\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">LlamaIndex\u306eNode Parser<\/h2>\n\n\n\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" width=\"600\" height=\"300\" src=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/10\/27_anime_midashi.jpg\" alt=\"\u898b\u51fa\u3057\u753b\u50cf\" class=\"wp-image-12415\" srcset=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/10\/27_anime_midashi.jpg 600w, https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/10\/27_anime_midashi-300x150.jpg 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/><\/figure>\n\n\n\n<p>LlamaIndex\u3067\u306f\u3001\u30c6\u30ad\u30b9\u30c8\u3084\u30c7\u30fc\u30bf\u3092\u52b9\u679c\u7684\u306b\u30c1\u30e3\u30f3\u30af\u5316\u3059\u308b\u305f\u3081\u306b\u3001\u3055\u307e\u3056\u307e\u306a\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u304c\u7528\u610f\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u69cb\u6587\u30d9\u30fc\u30b9\u3001\u9577\u3055\u30d9\u30fc\u30b9\u3001\u610f\u5473\u30d9\u30fc\u30b9\u3001\u30ea\u30ec\u30fc\u30b7\u30e7\u30f3\u30d9\u30fc\u30b9\u3068\u3044\u3063\u305f\u306a\u30d1\u30fc\u30b5\u30fc\u3092\u6d3b\u7528\u3059\u308b\u3053\u3068\u3067\u3001\u30c7\u30fc\u30bf\u306e\u5f62\u5f0f\u306b\u5fdc\u3058\u305f\u67d4\u8edf\u306a\u30c1\u30e3\u30f3\u30af\u5206\u5272\u304c\u53ef\u80fd\u3067\u3059\u3002<\/p>\n\n\n\n<p>\u3053\u306e\u8a18\u4e8b\u3067\u306f\u5404\u7a2e\u30d1\u30fc\u30b5\u30fc\u306b\u3064\u3044\u3066\u8a73\u3057\u304f\u89e3\u8aac\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n<div class=\"wp-block-group is-style-big_icon_good\"><div class=\"wp-block-group__inner-container\">\n<ul class=\"wp-block-list\">\n<li>\u69cb\u6587\u30d9\u30fc\u30b9\u306eNode Parser<\/li>\n\n\n\n<li>\u9577\u3055\u30d9\u30fc\u30b9Node Parser<\/li>\n\n\n\n<li>\u610f\u5473\u30d9\u30fc\u30b9\u306eNode Parser<\/li>\n\n\n\n<li>\u30ea\u30ec\u30fc\u30b7\u30e7\u30f3\u30d9\u30fc\u30b9\u306eNode Parser<\/li>\n<\/ul>\n<\/div><\/div>\n\n\n\n<div class=\"p-blogParts post_content\" data-partsID=\"4485\"><p><div class=\"p-adBox -ranking -border-on\" data-id=\"4507\" data-ad=\"ranking\"><div class=\"p-adBox__title -rank1\">1\u6642\u959350\u5186\u304b\u3089\u4f7f\u3048\u308bGPU\u30af\u30e9\u30a6\u30c9\u300cGPUSOROBAN\u300d<\/div><div class=\"p-adBox__body\"><div class=\"p-adBox__img\"><div class=\"centerw250\"><a href=\"https:\/\/soroban.highreso.jp\/?utm_medium=edgehub&utm_source=edgehub&utm_campaign=highend-boxtop&adid=edgehub\" rel=\"nofollow\" referrerpolicy=\"no-referrer-when-downgrade\"><img decoding=\"async\" src=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/06\/gpusorobanlllm_banner.jpg\" alt=\"GPUSOROBAN\"><\/a><\/div><\/div><div class=\"p-adBox__details\"><div class=\"p-adBox__star c-reviewStars\"><i class=\"icon-star-full\"><\/i><i class=\"icon-star-full\"><\/i><i class=\"icon-star-full\"><\/i><i class=\"icon-star-full\"><\/i><i class=\"icon-star-full\"><\/i><\/div><div class=\"p-adBox__price u-thin u-fz-s\">GPU\u306e\u30b3\u30b9\u30c8\u3067\u304a\u60a9\u307f\u306e\u65b9\u306b\uff01<\/div><div class=\"p-adBox__desc\">\u751f\u6210AI\u30fbLLM\u5411\u3051\u306e\u9ad8\u901fGPU\u3092\u696d\u754c\u6700\u5b89\u7d1a\u306e1\u6642\u959350\u5186\u304b\u3089\u5229\u7528\u3067\u304d\u307e\u3059\u3002\r\n\u30a4\u30f3\u30bf\u30fc\u30cd\u30c3\u30c8\u74b0\u5883\u3055\u3048\u3042\u308c\u3070\u3001GPU\u30b5\u30fc\u30d0\u30fc\u3092\u3059\u3050\u306b\u4f7f\u3048\u307e\u3059\u3002\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u3092\u8d77\u52d5\u3057\u3066\u3044\u308b\u3068\u304d\u306e\u307f\u306e\u8ab2\u91d1\u3067\u3001\u30b3\u30b9\u30c8\u3092\u6700\u5c0f\u5316\u3067\u304d\u307e\u3059\u3002\r\n<\/div><\/div><\/div><div class=\"p-adBox__btns\"><a href=\"https:\/\/soroban.highreso.jp\/compute\/?utm_medium=edgehub&utm_source=edgehub&utm_campaign=boxtop&adid=edgehub\" class=\"p-adBox__btn -btn1\" target=\"_blank\" rel=\"noopener nofollow\">\u516c\u5f0f\u30b5\u30a4\u30c8\u3092\u898b\u308b<\/a><a href=\"https:\/\/soroban.highreso.jp\/download\/compute-catalog\/?utm_medium=edgehub&utm_source=edgehub&utm_campaign=boxtop&adid=edgehub\" class=\"p-adBox__btn -btn2\" target=\"_blank\" rel=\"noopener nofollow\">\u8cc7\u6599\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9<\/a><\/div><\/div><\/p>\n<\/div>\n\n\n\n<h2 class=\"wp-block-heading\">\u69cb\u6587\u30d9\u30fc\u30b9\u306eNode Parser<\/h2>\n\n\n\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" width=\"600\" height=\"300\" src=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/08\/koara8_midashi.jpg\" alt=\"\u898b\u51fa\u3057\u753b\u50cf\" class=\"wp-image-9903\" srcset=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/08\/koara8_midashi.jpg 600w, https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/08\/koara8_midashi-300x150.jpg 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/><\/figure>\n\n\n\n<p><span class=\"swl-format-2\">\u69cb\u6587\u30d9\u30fc\u30b9\u306e\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u306f\u3001\u30c6\u30ad\u30b9\u30c8\u306e\u69cb\u9020\u306b\u57fa\u3065\u3044\u3066\u30c7\u30fc\u30bf\u3092\u5206\u5272\u3059\u308b\u624b\u6cd5\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>HTML\u3084JSON\u3001Markdown\u306a\u3069\u7279\u5b9a\u306e\u30d5\u30a9\u30fc\u30de\u30c3\u30c8\u306b\u5f93\u3063\u3066\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3092\u89e3\u6790\u3057\u3001\u898b\u51fa\u3057\u3084\u6bb5\u843d\u3001\u30bf\u30b0\u306b\u57fa\u3065\u3044\u3066\u30ce\u30fc\u30c9\uff08\u30c1\u30e3\u30f3\u30af\uff09\u306b\u5206\u5272\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u3053\u308c\u306b\u3088\u308a\u6587\u66f8\u69cb\u9020\u3092\u4fdd\u6301\u3057\u306a\u304c\u3089\u3001\u30c7\u30fc\u30bf\u3092\u52b9\u7387\u7684\u306b\u51e6\u7406\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u69cb\u6587\u30d9\u30fc\u30b9\u306e\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u306f\u4ee5\u4e0b\u306e\u3088\u3046\u306a\u3082\u306e\u304c\u3042\u308a\u307e\u3059\u3002<\/p>\n\n\n\n<div class=\"wp-block-group is-style-big_icon_good\"><div class=\"wp-block-group__inner-container\">\n<ul class=\"wp-block-list\">\n<li>SimpleFileNodeParser<\/li>\n\n\n\n<li>HTMLNodeParser\u3001<\/li>\n\n\n\n<li>JSONNodeParser<\/li>\n\n\n\n<li>MarkdownNodeParser<\/li>\n<\/ul>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">\u4e8b\u524d\u6e96\u5099<\/h3>\n\n\n\n<p>\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>pip install llama_index llama-index-core<\/code><\/pre>\n\n\n\n<h3 class=\"wp-block-heading\">SimpleFile NodeParser<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">SimpleFileNodeParser \u306f\u3001\u3055\u307e\u3056\u307e\u306a\u5f62\u5f0f\u306e\u30d5\u30a1\u30a4\u30eb\u3092\u30b7\u30f3\u30d7\u30eb\u306b\u30d1\u30fc\u30b9\u3059\u308b\u305f\u3081\u306e\u30af\u30e9\u30b9\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u30d5\u30a1\u30a4\u30eb\u304b\u3089\u8aad\u307f\u8fbc\u307e\u308c\u305f\u30c6\u30ad\u30b9\u30c8\u30c7\u30fc\u30bf\u3092\u9069\u5207\u306a\u30b5\u30a4\u30ba\u306e\u30c1\u30e3\u30f3\u30af\u306b\u5206\u5272\u3057\u3001\u305d\u308c\u3092\u30ce\u30fc\u30c9\u306b\u5909\u63db\u3059\u308b\u5f79\u5272\u3092\u62c5\u3063\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u7279\u5b9a\u306e\u30d5\u30a9\u30fc\u30de\u30c3\u30c8\u306b\u4f9d\u5b58\u305b\u305a\u306b\u30c6\u30ad\u30b9\u30c8\u3092\u6271\u3048\u308b\u305f\u3081\u3001\u6c4e\u7528\u6027\u304c\u9ad8\u304f\u67d4\u8edf\u306b\u5bfe\u5fdc\u3067\u304d\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>#SimpleFileNodeParser\nfrom llama_index.core.node_parser import SimpleFileNodeParser\nfrom llama_index.readers.file import FlatReader\nfrom pathlib import Path\n\ndocs = FlatReader().load_data(Path(\"\u7b2c1\u7ae0_\u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069.txt\"))\n\nparser = SimpleFileNodeParser()\n\nnodes = parser.get_nodes_from_documents(docs)\nnodes&#091;0].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'\u7b2c1\u7ae0 \u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069\\n\\t \u7b2c1\u7bc0\\t \u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u5f85\u9047\u6539\u5584\u3001\u9577\u6642\u9593\u52b4\u50cd\u306e\u662f\u6b63\u7b49\\n\\t1\\t\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u5bfe\u7b56\\n\uff081\uff09\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u8ab2\u984c\\n\u8fd1\u5e74\u3001\u30d1\u30fc\u30c8\u30bf\u30a4\u30e0\u52b4\u50cd\u8005\u3001\u6709\u671f\u96c7\u7528\u52b4\u50cd\u8005\u3001\u6d3e\u9063\u52b4\u50cd\u8005\u3068\u3044\u3063\u305f\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306f\\n\u5168\u4f53\u3068\u3057\u3066\u5897\u52a0\u50be\u5411\u306b\u3042\u308a\u3001\u96c7\u7528\u8005\u306e\u7d044\u5272\u3092\u5360\u3081\u308b\u72b6\u6cc1\u306b\u3042\u308b\u3002\u3053\u308c\u306f\u3001\u9ad8\u9f62\u8005\u304c\u5897\u3048\\n\u308b\u4e2d\u3001\u9ad8\u9f62\u5c64\u3067\u306e\u7d99\u7d9a\u96c7\u7528\u306b\u3088\u308a\u975e\u6b63\u898f\u96c7\u7528\u304c\u5897\u52a0\u3057\u3066\u3044\u308b\u3053\u3068\u3084\u3001\u5973\u6027\u3092\u4e2d\u5fc3\u306b\u30d1\u30fc\u30c8\\n\u306a\u3069\u3067\u50cd\u304d\u59cb\u3081\u308b\u52b4\u50cd\u8005\u304c\u5897\u52a0\u3057\u3066\u3044\u308b\u3053\u3068\u306a\u3069\u306e\u8981\u56e0\u304c\u5927\u304d\u3044\u3002\u306a\u304a\u3001\u65b0\u578b\u30b3\u30ed\u30ca\u30a6\u30a4\\n\u30eb\u30b9\u611f\u67d3\u75c7\u306e\u611f\u67d3\u62e1\u5927\u306e\u5f71\u97ff\u3082\u3042\u308a\u30012020\uff08\u4ee4\u548c2\uff09\u5e74\u30012021\uff08\u4ee4\u548c3\uff09\u5e74\u306e\u975e\u6b63\u898f\u96c7\\n\u7528\u52b4\u50cd\u8005\u306f\u5bfe\u524d\u5e74\u6bd4\u3067\u6e1b\u5c11\u3057\u305f\u304c\u30012022\uff08\u4ee4\u548c4\uff09\u5e74\u4ee5\u964d\u306f\u5897\u52a0\u3057\u30012023\uff08\u4ee4\u548c5\uff09\u5e74\\n\u306f\u30012,124\u4e07\u4eba\u3068\u306a\u3063\u3066\u3044\u308b\u3002\\n\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306f\u3001\u96c7\u7528\u304c\u4e0d\u5b89\u5b9a\u3001\u8cc3\u91d1\u304c\u4f4e\u3044\u3001\u80fd\u529b\u958b\u767a\u6a5f\u4f1a\u304c\u4e4f\u3057\u3044\u306a\u3069\u306e\u8ab2\u984c\u304c\\n\u3042\u308a\u3001\u6b63\u898f\u96c7\u7528\u3092\u5e0c\u671b\u3057\u306a\u304c\u3089\u305d\u308c\u304c\u304b\u306a\u308f\u305a\u3001\u975e\u6b63\u898f\u96c7\u7528\u3067\u50cd\u304f\u8005\uff08\u4e0d\u672c\u610f\u975e\u6b63\u898f\u96c7\u7528\\n\u52b4\u50cd\u8005\uff09\u304c9.6\uff05\uff082023\u5e74\uff09\u5b58\u5728\u3057\u3001\u5e74\u9f62\u968e\u7d1a\u5225\u3067\u306f25\uff5e34\u6b73\u306e\u82e5\u5e74\u5c64\u306713.1\uff05\uff082023\\n\u5e74\uff09\u3068\u9ad8\u304f\u306a\u3063\u3066\u3044\u308b\u3002\u4e00\u65b9\u3001\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u4e2d\u306b\u306f\u300c\u81ea\u5206\u306e\u90fd\u5408\u306e\u3088\u3044\u6642\u9593\u306b\u50cd\u304d\\n\u305f\u3044\u304b\u3089\u300d\u7b49\u306e\u7406\u7531\u306b\u3088\u308a\u81ea\u3089\u975e\u6b63\u898f\u96c7\u7528\u3092\u9078\u3076\u65b9\u3082\u304a\u308a\u3001\u591a\u69d8\u306a\u50cd\u304d\u65b9\u304c\u9032\u3080\u4e2d\u3067\u3001\u3069\\n\u306e\u3088\u3046\u306a\u96c7\u7528\u5f62\u614b\u3092\u9078\u629e\u3057\u3066\u3082\u7d0d\u5f97\u304c\u5f97\u3089\u308c\u308b\u51e6\u9047\u3092\u53d7\u3051\u3089\u308c\u308b\u3053\u3068\u304c\u91cd\u8981\u3067\u3042\u308b\u3002\\n\uff082\uff09\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u3078\u306e\u7dcf\u5408\u7684\u306a\u5bfe\u7b56\u306e\u63a8\u9032\\n1 \u6b63\u793e\u54e1\u8ee2\u63db\u30fb\u5f85\u9047\u6539\u5584\u306e\u63a8\u9032\\n\u6b63\u793e\u54e1\u3092\u5e0c\u671b\u3059\u308b\u65b9\u306e\u6b63\u793e\u54e1\u8ee2\u63db\u3084\u975e\u6b63\u898f\u96c7\u7528\u3092\u9078\u629e\u3059\u308b\u65b9\u306e\u5f85\u9047\u6539\u5584\u3092\u63a8\u9032\u3059\u308b\u305f\\n\u3081\u3001\u30ad\u30e3\u30ea\u30a2\u30a2\u30c3\u30d7\u52a9\u6210\u91d1\u306b\u304a\u3044\u3066\u3001\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u6b63\u793e\u54e1\u8ee2\u63db\u3001\u51e6\u9047\u6539\u5584\u306e\u53d6\u7d44\u307f\\n\u3092\u56f3\u308b\u4e8b\u696d\u4e3b\u306b\u5bfe\u3057\u3066\u52a9\u6210\u3092\u884c\u3063\u3066\u3044\u308b\u3002\\n\u307e\u305f\u3001\u3069\u306e\u3088\u3046\u306a\u50cd\u304d\u65b9\u3092\u9078\u629e\u3057\u3066\u3082\u516c\u6b63\u306a\u5f85\u9047\u3092\u53d7\u3051\u3089\u308c\u308b\u3088\u3046\u306b\u3057\u3001\u4eba\u3005\u304c\u81ea\u5206\u306e\\n\u5e0c\u671b\u306b\u5408\u308f\u305b\u3066\u591a\u69d8\u306a\u50cd\u304d\u65b9\u3092\u81ea\u7531\u306b\u9078\u629e\u3067\u304d\u308b\u3088\u3046\u306b\u3059\u308b\u3053\u3068\u304c\u91cd\u8981\u3067\u3042\u308b\u3002\\n2020\uff08\u4ee4\u548c2\uff09\u5e744\u67081\u65e5\u306b\u65bd\u884c\u3055\u308c\u305f\u300c\u77ed\u6642\u9593\u52b4\u50cd\u8005\u53ca\u3073\u6709\u671f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u96c7\u7528\u7ba1\u7406\\n\u306e\u6539\u5584\u7b49\u306b\u95a2\u3059\u308b\u6cd5\u5f8b\u300d\uff08\u5e73\u62105\u5e74\u6cd5\u5f8b\u7b2c76\u53f7\u3002\u4ee5\u4e0b\u300c\u30d1\u30fc\u30c8\u30bf\u30a4\u30e0\u30fb\u6709\u671f\u96c7\u7528\u52b4\u50cd\u6cd5\u300d\\n\u3068\u3044\u3046\u3002\u540c\u6cd5\u306e\u4e2d\u5c0f\u4f01\u696d\u3078\u306e\u9069\u7528\u306f2021\uff08\u4ee4\u548c3\uff09\u5e744\u67081\u65e5\u3002\uff09\u53ca\u3073\u300c\u52b4\u50cd\u8005\u6d3e\u9063\u4e8b\u696d\\n\u306e\u9069\u6b63\u306a\u904b\u55b6\u306e\u78ba\u4fdd\u53ca\u3073\u6d3e\u9063\u52b4\u50cd\u8005\u306e\u4fdd\u8b77\u7b49\u306b\u95a2\u3059\u308b\u6cd5\u5f8b\u300d\uff08\u662d\u548c60\u5e74\u6cd5\u5f8b\u7b2c88\u53f7\uff09\u3067\\n\u306f\u3001\u96c7\u7528\u5f62\u614b\u306b\u304b\u304b\u308f\u3089\u306a\u3044\u516c\u6b63\u306a\u5f85\u9047\u306e\u78ba\u4fdd\u306b\u5411\u3051\u3001\u2460\u4e0d\u5408\u7406\u306a\u5f85\u9047\u5dee\u3092\u89e3\u6d88\u3059\u308b\u305f\u3081\\n\u306e\u898f\u5b9a\u306e\u6574\u5099\u3001\u2461\u52b4\u50cd\u8005\u306b\u5bfe\u3059\u308b\u5f85\u9047\u306b\u95a2\u3059\u308b\u8aac\u660e\u7fa9\u52d9\u306e\u5f37\u5316\u3001\u2462\u884c\u653f\u306b\u3088\u308b\u6cd5\u306e\u5c65\u884c\u78ba\\n\u4fdd\u63aa\u7f6e\u53ca\u3073\u88c1\u5224\u5916\u7d1b\u4e89\u89e3\u6c7a\u624b\u7d9a\uff08\u884c\u653fADR\uff09\u304c\u6574\u5099\u3055\u308c\u305f\u3002\\n\u30d5\u30ea\u30fc\u30bf\u30fc\u7b49\uff0a1\u306e\u6b63\u793e\u54e1\u5c31\u8077\u652f\u63f4\u306e\u305f\u3081\u3001\u300c\u308f\u304b\u3082\u306e\u30cf\u30ed\u30fc\u30ef\u30fc\u30af\u300d\uff082024\uff08\u4ee4\u548c6\uff09\u5e74\\n4\u67081\u65e5\u73fe\u572821\u304b\u6240\uff09\u7b49\u3092\u62e0\u70b9\u306b\u3001\u62c5\u5f53\u8005\u5236\u306b\u3088\u308b\u500b\u5225\u652f\u63f4\u3001\u6b63\u793e\u54e1\u5c31\u8077\u306b\u5411\u3051\u305f\u30bb\u30df\\n\uff0a1 \u304a\u304a\u3080\u306d35\u6b73\u672a\u6e80\u3067\u6b63\u793e\u54e1\u3067\u306e\u5c31\u8077\u3092\u5e0c\u671b\u3059\u308b\u6c42\u8077\u8005\n\uff1c\u4ee5\u4e0b\u7701\u7565\uff1e<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">FlatReader().load_data()<\/span><\/p>\n\n\n\n<p>\u6307\u5b9a\u3057\u305f\u30c6\u30ad\u30b9\u30c8\u30d5\u30a1\u30a4\u30eb\u3092\u8aad\u307f\u8fbc\u307f\u3001\u30c6\u30ad\u30b9\u30c8\u30c7\u30fc\u30bf\u3092\u8fd4\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">parser = SimpleFileNodeParser()<\/span><\/p>\n\n\n\n<p><code>SimpleFileNodeParser<\/code>\u306e\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u3092\u4f7f\u7528\u3057\u3066\u3001\u30d5\u30a1\u30a4\u30eb\u306e\u5185\u5bb9\u3092\u8907\u6570\u306e\u30ce\u30fc\u30c9\u306b\u5206\u5272\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">nodes = parser.get_nodes_from_documents(docs)<\/span><\/p>\n\n\n\n<p>\u8aad\u307f\u8fbc\u3093\u3060\u30c6\u30ad\u30b9\u30c8\u30c7\u30fc\u30bf<code>docs<\/code>\u3092\u30ce\u30fc\u30c9\u306b\u5206\u5272\u3057\u307e\u3059\u3002\u7d50\u679c\u306f <code>nodes<\/code> \u3068\u3044\u3046\u30ea\u30b9\u30c8\u306b\u4fdd\u5b58\u3055\u308c\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">HTML NodeParser<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">HTMLNodeParser \u306f\u3001HTML\u6587\u66f8\u3092\u89e3\u6790\u3057\u3001HTML\u30bf\u30b0\u306b\u57fa\u3065\u3044\u3066\u30c6\u30ad\u30b9\u30c8\u3092\u30c1\u30e3\u30f3\u30af\uff08\u30ce\u30fc\u30c9\uff09\u306b\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u30af\u30e9\u30b9\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u3053\u306e\u30d1\u30fc\u30b5\u30fc\u306f\u3001HTML\u69cb\u9020\u3092\u7406\u89e3\u3057\u3001\u6307\u5b9a\u3055\u308c\u305f\u30bf\u30b0\uff08\u4f8b\u3048\u3070<code>&lt;p&gt;<\/code> \u3084 <code>&lt;h1&gt;<\/code>\uff09\u3092\u57fa\u6e96\u306b\u30c6\u30ad\u30b9\u30c8\u3092\u5206\u5272\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>HTML\u306e\u30e1\u30bf\u30c7\u30fc\u30bf\u3082\u30ce\u30fc\u30c9\u306b\u542b\u3081\u308b\u3053\u3068\u304c\u3067\u304d\u3001\u69cb\u9020\u5316\u3055\u308c\u305f\u30c7\u30fc\u30bf\u306e\u6271\u3044\u306b\u4fbf\u5229\u3067\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>#HTMLNodeParser\nimport requests\nfrom llama_index.core import Document\nfrom llama_index.core.node_parser import HTMLNodeParser\n\nurl = \"https:\/\/ja.wikipedia.org\/wiki\/%E5%8E%9A%E7%94%9F%E5%8A%B4%E5%83%8D%E7%9C%81\"\n\nresponse = requests.get(url)\nprint(response)\n\nif response.status_code == 200:\n    html_doc = response.text\n    document = Document(id_=url, text=html_doc)\n\n    parser = HTMLNodeParser(tags=&#091;\"p\", \"h1\"])\n    nodes = parser.get_nodes_from_documents(&#091;document])\n    print(nodes)\nelse:\n    print(\"Failed to fetch HTML content:\", response.status_code)<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>Response &#091;200]&gt;\n&#091;TextNode(id_='7ea74821-03a3-435a-93d1-415152a092ea', embedding=None, metadata={'tag': 'h1'}, excluded_embed_metadata_keys=&#091;], excluded_llm_metadata_keys=&#091;], relationships={&lt;NodeRelationship.SOURCE: '1'&gt;: RelatedNodeInfo(node_id='https:\/\/ja.wikipedia.org\/wiki\/%E5%8E%9A%E7%94%9F%E5%8A%B4%E5%83%8D%E7%9C%81', node_type=&lt;ObjectType.DOCUMENT: '4'&gt;, metadata={}, hash='33ad0f2335430d2596f98c31eb90646051448d37ee2a267f386d735ea0c5b971'), &lt;NodeRelationship.NEXT: '3'&gt;: RelatedNodeInfo(node_id='55e8868f-5021-4b6e-9aa8-021fc58c078a', node_type=&lt;ObjectType.TEXT: '1'&gt;, metadata={'tag': 'p'}, hash='1c798cb79adb44866e6431e2a5b24dd97f30b587c101ec697755fe738807a411')}, text='\u539a\u751f\u52b4\u50cd\u7701', mimetype='text\/plain', start_char_idx=614, end_char_idx=619, text_template='{metadata_str}\\n\\n{content}', metadata_template='{key}: {value}', metadata_seperator='\\n'), TextNode(id_='55e8868f-5021-4b6e-9aa8-021fc58c078a', embedding=None, metadata={'tag': 'p'}, excluded_embed_metadata_keys=&#091;], excluded_llm_metadata_keys=&#091;], relationships={&lt;NodeRelationship.SOURCE: '1'&gt;: RelatedNodeInfo(node_id='https:\/\/ja.wikipedia.org\/wiki\/%E5%8E%9A%E7%94%9F%E5%8A%B4%E5%83%8D%E7%9C%81', node_type=&lt;ObjectType.DOCUMENT: '4'&gt;, metadata={}, hash='33ad0f2335430d2596f98c31eb90646051448d37ee2a267f386d735ea0c5b971'), &lt;NodeRelationship.PREVIOUS: '2'&gt;: RelatedNodeInfo(node_id='7ea74821-03a3-435a-93d1-415152a092ea', node_type=&lt;ObjectType.TEXT: '1'&gt;, metadata={'tag': 'h1'}, hash='826fdfac59e2ad455a222e85cdabff72e36e9084c8ae8cbfd69288d79d43be83')}, text='\u539a\u751f\u52b4\u50cd\u7701\\n\uff08\u3053\u3046\u305b\u3044\u308d\u3046\u3069\u3046\u3057\u3087\u3046\u3001\\n\u82f1\\n:\\nMinistry of Health, Labour and Welfare\\n\u3001\u7565\u79f0:\\nMHLW\\n\uff09\u306f\u3001\\n\u65e5\u672c\\n\u306e\\n\u884c\u653f\u6a5f\u95a2\\n\u306e\u3072\u3068\u3064\\n&#091;4]\\n\u3002\\n\u5065\u5eb7\\n\u3001\\n\u533b\u7642\\n\u3001\\n\u798f\u7949\\n\u3001\\n\u4ecb\u8b77\\n\u3001\\n\u96c7\u7528\\n\u3001\\n\u52b4\u50cd\\n\u3001\u304a\u3088\u3073\\n\u5e74\u91d1\\n\u306b\u95a2\u3059\u308b\\n\u884c\u653f\\n&#091;5]\\n\u306a\u3089\u3073\u306b\\n\u5fa9\u54e1\\n\u3001\\n\u6226\u6ca1\u8005\\n\u907a\u65cf\\n\u7b49\u306e\u63f4\u8b77\u3001\u65e7\\n\u9678\\n\u6d77\u8ecd\\n\u306e\u6b8b\u52d9\u6574\u7406\u3092\u6240\u7ba1\u3059\u308b\\n&#091;\u6ce8\u91c8 1]\\n\u3002\\n\u65e5\u672c\u8a9e\\n\u7565\u79f0\u30fb\u901a\u79f0\u306f\u3001\\n\u539a\u52b4\u7701\\n\uff08\u3053\u3046\u308d\u3046\u3057\u3087\u3046\uff09\u3002\\n2001\u5e74\\n\uff08\\n\u5e73\u6210\\n13\u5e74\uff091\u6708\u306e\\n\u4e2d\u592e\u7701\u5e81\u518d\u7de8\\n\u306b\u3088\u308a\u3001\\n\u539a\u751f\u7701\\n\u3068\\n\u52b4\u50cd\u7701\\n\u3092\u7d71\u5408\u3057\u3066\u8a95\u751f\u3057\u305f\u3002\u4e88\u7b97\u898f\u6a21\u306f\u4e2d\u592e\u7701\u5e81\u306e\u4e2d\u3067\u6700\u5927\u3067\u3042\u308b\u3002\\n\u539a\u751f\u52b4\u50cd\u7701\u8a2d\u7f6e\u6cd5\u7b2c4\u6761\u306f\u8a08118\u9805\u76ee\u306e\u6240\u638c\u3059\u308b\u4e8b\u52d9\u3092\u5217\u8a18\u3057\u3066\u3044\u308b\u3002\u5177\u4f53\u7684\u306b\u306f\u4ee5\u4e0b\u306e\u4e8b\u9805\u306b\u95a2\u3059\u308b\u4e8b\u52d9\u304c\u3042\u308b\u3002\\n\u5185\u90e8\u7d44\u7e54\u306f\u4e00\u822c\u7684\u306b\u3001\u6cd5\u5f8b\u306e\u539a\u751f\u52b4\u50cd\u7701\u8a2d\u7f6e\u6cd5\u3001\u653f\u4ee4\u306e\u539a\u751f\u52b4\u50cd\u7701\u7d44\u7e54\u4ee4\u53ca\u3073\u7701\u4ee4\u306e\u539a\u751f\u52b4\u50cd\u7701\u7d44\u7e54\u898f\u5247\u304c\u968e\u5c64\u7684\u306b\u898f\u5b9a\u3057\u3066\u3044\u308b\u3002\\n\\n\u5f53\u7701\u306e\\n\u65bd\u8a2d\u7b49\u6a5f\u95a2\\n\u306f\u4ee5\u4e0b\u306e7\u533a\u5206\u304c\u3042\u308b\u3002\u56fd\u7acb\u5150\u7ae5\u81ea\u7acb\u652f\u63f4\u65bd\u8a2d\uff08\u73fe\u5728\u306f\u3053\u3069\u3082\u5bb6\u5ead\u5e81\u306b\u79fb\u7ba1\uff09\u304a\u3088\u3073\u56fd\u7acb\u969c\u5bb3\u8005\u30ea\u30cf\u30d3\u30ea\u30c6\u30fc\u30b7\u30e7\u30f3\u30bb\u30f3\u30bf\u30fc\\n&#091;\u6ce8\u91c8 6]\\n\u306f\u6163\u4f8b\u4e0a\u3001\u300c\u56fd\u7acb\u66f4\u751f\u63f4\u8b77\u6a5f\u95a2\u300d\u3068\u7dcf\u79f0\u3055\u308c\u308b\u3002\\n\u539a\u751f\u52b4\u50cd\u7701\u691c\u75ab\u6240\\n\u306f\u4ee5\u4e0b\u306e13\u691c\u75ab\u6240\u306e\u4e0b\u306b14\u652f\u6240\u306880\u51fa\u5f35\u6240\u304c\u7f6e\u304b\u308c\u3066\u3044\u308b\u3002\\n\u5730\u65b9\u652f\u5206\u90e8\u5c40\\n\u306f\u5730\u65b9\u539a\u751f\u5c40\u3068\u90fd\u9053\u5e9c\u770c\u52b4\u50cd\u5c40\u306e2\u533a\u5206\u304c\u3042\u308b\u3002\u90fd\u9053\u5e9c\u770c\u52b4\u50cd\u5c40\u306f47\u5404\u90fd\u9053\u5e9c\u770c\u306b1\u3064\u8a2d\u7f6e\u3055\u308c\u3066\u3044\u308b\u3002\\n\u592a\u5b57\\n\u306f\u4eba\u4e8b\u30d6\u30ed\u30c3\u30af\u57fa\u5e79\u5c40\\n\uff08\u5317\u6d77\u9053\u30fb\u5bae\u57ce\u30fb\u57fc\u7389\u30fb\u6771\u4eac\u30fb\u65b0\u6f5f\u30fb\u611b\u77e5\u30fb\u5927\u962a\u30fb\u5e83\u5cf6\u30fb\u9999\u5ddd\u30fb\u798f\u5ca1\uff09\\n\u4e3b\u7ba1\u3059\u308b\\n\u72ec\u7acb\u884c\u653f\u6cd5\u4eba\\n\u306f2024\u5e744\u67081\u65e5\u73fe\u5728\u3001\u4ee5\u4e0b\u306b\u793a\u3059\u901a\u308a\u3001\\n\u4e2d\u671f\u76ee\u6a19\u7ba1\u7406\u6cd5\u4eba\\n10\u3001\\n\u56fd\u7acb\u7814\u7a76\u958b\u767a\u6cd5\u4eba\\n7\u306e\u8a0817\u6cd5\u4eba\u3067\u3042\u308b\\n&#091;9]\\n\uff09\u3002\\n\u884c\u653f\u57f7\u884c\u6cd5\u4eba\\n\u306f\u6240\u7ba1\u3057\u306a\u3044\u3002\\n\u56fd\u7acb\u75c5\u9662\u6a5f\u69cb\u306f\u3001\u4ee5\u524d\u306f\u5f79\u8077\u54e1\u304c\\n\u56fd\u5bb6\u516c\u52d9\u54e1\\n\u3067\u3042\u308b\u300c\u7279\u5b9a\u72ec\u7acb\u884c\u653f\u6cd5\u4eba\uff08\u73fe\u30fb\u884c\u653f\u57f7\u884c\u6cd5\u4eba\uff09\u300d\u3067\u3042\u3063\u305f\u304c\u3001\\n\u72ec\u7acb\u884c\u653f\u6cd5\u4eba\u901a\u5247\u6cd5\\n\u306e\u6539\u6b63\u6cd5\uff08\u5e73\u621026\u5e746\u670813\u65e5\u6cd5\u5f8b\u7b2c66\u53f7\uff09\u65bd\u884c\u306b\u4f34\u3044\u30012015\u5e74\uff08\u5e73\u621027\u5e74\uff094\u67081\u65e5\u304b\u3089\u4e2d\u671f\u76ee\u6a19\u7ba1\u7406\u6cd5\u4eba\u3068\u306a\u308a\u3001\u5f79\u8077\u54e1\u306f\u56fd\u5bb6\u516c\u52d9\u54e1\u3067\u306f\u306a\u304f\u306a\u3063\u305f\u3002\\n\u4e3b\u7ba1\u3059\u308b\u7279\u6b8a\u6cd5\u4eba\u306f2024\u5e74\uff08\u4ee4\u548c6\u5e74\uff094\u67081\u65e5\u73fe\u5728\u3001\\n\u65e5\u672c\u5e74\u91d1\u6a5f\u69cb\\n\uff08\u5e74\u91d1\u5c40\uff09\u306e\u307f\u3067\u3042\u308b\\n&#091;10]\\n\u3002\u65e7\\n\u793e\u4f1a\u4fdd\u967a\u5e81\\n\u306e\u5f8c\u8eab\u306b\u3042\u305f\u308b\u3002\u306a\u304a\u30012025\u5e744\u6708\u306b\u56fd\u7acb\u611f\u67d3\u75c7\u7814\u7a76\u6240\u3068\u56fd\u7acb\u56fd\u969b\u533b\u7642\u7814\u7a76\u30bb\u30f3\u30bf\u30fc\u3092\u7d71\u5408\u3057\u3066\u7279\u6b8a\u6cd5\u4eba\u3068\u3057\u3066\u56fd\u7acb\u5065\u5eb7\u5371\u6a5f\u7ba1\u7406\u7814\u7a76\u6a5f\u69cb\u304c\u8a2d\u7acb\u3055\u308c\u308b\\n&#091;11]\\n\u3002\\n\u7279\u5225\u306e\u6cd5\u5f8b\u306b\u3088\u308a\u8a2d\u7acb\u3055\u308c\u308b\u6c11\u9593\u6cd5\u4eba\\n\uff08\u7279\u5225\u6c11\u9593\u6cd5\u4eba\uff09\u306b\u306f2024\u5e74\uff08\u4ee4\u548c6\u5e74\uff094\u67081\u65e5\u73fe\u5728\u3001\u4ee5\u4e0b\u306e11\u6cd5\u4eba\u3067\u3042\u308b\\n&#091;12]\\n\u3002\u4ee5\u524d\u306f\u52b4\u50cd\u57fa\u6e96\u5c40\u6240\u7ba1\u306e\\n\u9271\u696d\u52b4\u50cd\u707d\u5bb3\u9632\u6b62\u5354\u4f1a\\n\uff08\u7565\u79f0\uff1a\u9271\u707d\u9632\uff09\u304c\u3042\u3063\u305f\u304c\u30012014\u5e74\uff08\u5e73\u621026\u5e74\uff093\u670831\u65e5\u306b\u89e3\u6563\u3057\u305f\u3002\u5168\u56fd\u5065\u5eb7\u4fdd\u967a\u5354\u4f1a\u306f\u3001\u73fe\u5728\u306e\u8cc7\u6599\u3067\u306f\u7279\u5225\u306e\u6cd5\u5f8b\u306b\u3088\u308a\u8a2d\u7acb\u3055\u308c\u308b\u6cd5\u4eba\u3067\u306f\u306a\u304f\u3001\u7279\u5225\u306e\u6cd5\u5f8b\u306b\u3088\u308a\u8a2d\u7acb\u3055\u308c\u308b\u6c11\u9593\u6cd5\u4eba\u3068\u3055\u308c\u3066\u3044\u308b\u3002\\n\u7279\u5225\u306e\u6cd5\u5f8b\u306b\u3088\u308a\u8a2d\u7acb\u3055\u308c\u308b\u6cd5\u4eba\\n\u306b\u306f\\n\u5065\u5eb7\u4fdd\u967a\n\uff1c\u4ee5\u4e0b\u7701\u7565\uff1e<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">response = requests.get(url)<\/span><\/p>\n\n\n\n<p>\u6307\u5b9a\u3055\u308c\u305fURL\u306eHTML\u30c7\u30fc\u30bf\u3092\u53d6\u5f97\u3057\u307e\u3059\u3002<code>response<\/code> \u306b\u306fHTML\u306e\u5fdc\u7b54\u304c\u683c\u7d0d\u3055\u308c\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">if response.status_code == 200&#8230;<\/span><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>if response.status_code == 200:<\/code>:HTTP\u30ea\u30af\u30a8\u30b9\u30c8\u304c\u6210\u529f\u3057\u305f\u304b\u3069\u3046\u304b\u3092\u78ba\u8a8d\u3057\u307e\u3059\u3002<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>html_doc = response.text<\/code>: \u53d6\u5f97\u3057\u305fHTML\u30c7\u30fc\u30bf\u3092\u30c6\u30ad\u30b9\u30c8\u5f62\u5f0f\u3067 <code>html_doc<\/code> \u306b\u4fdd\u5b58\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>Document(id_=url, text=html_doc)<\/code>: HTML\u30c7\u30fc\u30bf\u3092 <code>Document<\/code> \u30af\u30e9\u30b9\u3067\u30e9\u30c3\u30d7\u3057\u3001\u5f8c\u306b<code>HTMLNodeParser<\/code> \u3067\u51e6\u7406\u3067\u304d\u308b\u5f62\u5f0f\u306b\u5909\u63db\u3055\u308c\u307e\u3059\u3002<code>id_<\/code>\u306b\u306fURL\u304c\u3001<code>text<\/code>\u306b\u306fHTML\u30c6\u30ad\u30b9\u30c8\u304c\u683c\u7d0d\u3055\u308c\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>HTMLNodeParser(tags=[\"p\", \"h1\"])<\/code>: <code>&lt;p&gt;<\/code>\u30bf\u30b0\u3068<code>&lt;h1&gt;<\/code>\u30bf\u30b0\u306b\u57fa\u3065\u3044\u3066HTML\u3092\u5206\u5272\u3059\u308b\u30d1\u30fc\u30b5\u3092\u4f5c\u6210\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>parser.get_nodes_from_documents([document])<\/code>: <code>Document<\/code> \u5185\u306eHTML\u30c6\u30ad\u30b9\u30c8\u3092\u30bf\u30b0\u306b\u57fa\u3065\u3044\u3066\u5206\u5272\u3057\u307e\u3059\u3002<code>nodes<\/code> \u306b\u306f\u5404\u30ce\u30fc\u30c9\u304c\u30ea\u30b9\u30c8\u3068\u3057\u3066\u4fdd\u5b58\u3055\u308c\u307e\u3059\u3002<\/li>\n<\/ul>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">JSON NodeParser<\/h3>\n\n\n\n<p>JSONNodeParser \u306f\u3001JSON\u5f62\u5f0f\u306e\u30c7\u30fc\u30bf\u3092\u89e3\u6790\u3057\u3001\u30c1\u30e3\u30f3\u30af\uff08\u30ce\u30fc\u30c9\uff09\u306b\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u30af\u30e9\u30b9\u3067\u3059\u3002<\/p>\n\n\n\n<p>\u3053\u306e\u30d1\u30fc\u30b5\u30fc\u306f\u3001JSON\u30c7\u30fc\u30bf\u306e\u69cb\u9020\u3092\u7406\u89e3\u3057\u3001\u5fc5\u8981\u306b\u5fdc\u3058\u3066\u6df1\u3044\u30cd\u30b9\u30c8\u3055\u308c\u305f\u69cb\u9020\u3082\u89e3\u6790\u3057\u3066\u5206\u5272\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u5404\u30ce\u30fc\u30c9\u306b\u306f\u3001JSON\u30c7\u30fc\u30bf\u5185\u306e\u30ad\u30fc\u3084\u5024\u306b\u95a2\u3059\u308b\u30e1\u30bf\u30c7\u30fc\u30bf\u3092\u4ed8\u52a0\u3059\u308b\u3053\u3068\u304c\u53ef\u80fd\u3067\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>#JSONNodeParser\nfrom llama_index.core.node_parser import JSONNodeParser\nimport os\nimport requests\n\nos.environ&#091;\"TAVILY_API_KEY\"] = \"tvly-*********************\"\napi_key = os.getenv(\"TAVILY_API_KEY\")\n\nurl = \"https:\/\/api.tavily.com\/search\"\n\nheaders = {\n    \"Content-Type\": \"application\/json\"\n}\n\ndata = {\n    \"api_key\": api_key,\n    \"query\": \"who is fukuoka takamaro\uff1f\",\n    \"search_depth\": \"basic\",\n    \"include_answer\": True,\n    \"max_results\": 5,\n    \"include_domains\": &#091;],\n    \"exclude_domains\": &#091;]\n}\n\nresponse = requests.post(url, headers=headers, json=data)\n\nif response.status_code == 200:\n    document = Document(id_=url, text=response.text)\n    parser = JSONNodeParser()\n\n    nodes = parser.get_nodes_from_documents(&#091;document])\n    print(nodes&#091;0])\nelse:\n    print(\"Failed to fetch JSON content:\", response.status_code)<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>Node ID: ed5baeef-e60e-4035-96a7-1e2425ea7c39\nText: query who is fukuoka takamaro\uff1f follow_up_questions None answer\nTakamaro Fukuoka is a Japanese politician who is a member of the\nLiberal Democratic Party and has been serving as the Minister of\nHealth, Labour, and Welfare since 2024. He has been a member of the\nHouse of Representatives since 2005. results title Takamaro Fukuoka -\nSimple English W...<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">os.environ[&#8220;TAVILY_API_KEY&#8221;]<\/span><\/p>\n\n\n\n<p>\u74b0\u5883\u5909\u6570<code>TAVILY_API_KEY<\/code>\u306bAPI\u30ad\u30fc\u3092\u8a2d\u5b9a\u3057\u307e\u3059\u3002\u3053\u306eAPI\u30ad\u30fc\u306f<a href=\"https:\/\/tavily.com\/\">Tavily\u306e\u516c\u5f0f\u30da\u30fc\u30b8<\/a>\u304b\u3089\u767a\u884c\u3067\u304d\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">url = &#8220;https:\/\/api.tavily.com\/search&#8221;<\/span><\/p>\n\n\n\n<p>API\u30a8\u30f3\u30c9\u30dd\u30a4\u30f3\u30c8\u306eURL\u3067\u3059\u3002\u3053\u3053\u3067\u306f <code>Tavily<\/code> \u3068\u3044\u3046\u691c\u7d22API\u306eURL\u3092\u6307\u5b9a\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">data = {&#8230;<\/span><\/p>\n\n\n\n<p> API\u30ea\u30af\u30a8\u30b9\u30c8\u306e\u30da\u30a4\u30ed\u30fc\u30c9\uff08\u30c7\u30fc\u30bf\uff09\u3067\u3059\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>query<\/code>: \u691c\u7d22\u30af\u30a8\u30ea\u3067\u3059\u3002\u3053\u306e\u4f8b\u3067\u306f\u300cFukuoka Takamaro\u3068\u306f\u8ab0\u304b\uff1f\u300d\u3068\u3044\u3046\u691c\u7d22\u3092\u884c\u3063\u3066\u3044\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>search_depth<\/code>: \u691c\u7d22\u306e\u6df1\u3055\u3092\u6307\u5b9a\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>include_answer<\/code>: \u691c\u7d22\u7d50\u679c\u306b\u56de\u7b54\u3092\u542b\u3081\u308b\u304b\u3069\u3046\u304b\u3092\u6307\u5b9a\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>max_results<\/code>: \u6700\u5927\u691c\u7d22\u7d50\u679c\u6570\u3092\u6307\u5b9a\u3057\u307e\u3059\u3002\u3053\u3053\u3067\u306f5\u4ef6\u306e\u7d50\u679c\u3092\u30ea\u30af\u30a8\u30b9\u30c8\u3057\u3066\u3044\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>include_domains<\/code>, <code>exclude_domains<\/code>: \u7279\u5b9a\u306e\u30c9\u30e1\u30a4\u30f3\u3092\u542b\u3081\u305f\u308a\u9664\u5916\u3057\u305f\u308a\u3059\u308b\u8a2d\u5b9a\u3067\u3059\u304c\u3001\u3053\u3053\u3067\u306f\u7a7a\u306e\u30ea\u30b9\u30c8\u304c\u6307\u5b9a\u3055\u308c\u3066\u304a\u308a\u3001\u30c9\u30e1\u30a4\u30f3\u306e\u5236\u9650\u306f\u3042\u308a\u307e\u305b\u3093\u3002<\/li>\n<\/ul>\n\n\n\n<p><span class=\"swl-format-2\">if response.status_code == 200&#8230;<\/span><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>response.status_code<\/code>: API\u304b\u3089\u306e\u30ec\u30b9\u30dd\u30f3\u30b9\u306e\u30b9\u30c6\u30fc\u30bf\u30b9\u30b3\u30fc\u30c9\u3092\u78ba\u8a8d\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>Document(id_=url, text=response.text)<\/code>: <code>response.text<\/code>\u306b\u306fAPI\u304b\u3089\u306eJSON\u30ec\u30b9\u30dd\u30f3\u30b9\u304c\u542b\u307e\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u306eJSON\u30c7\u30fc\u30bf\u3092<code>Document<\/code>\u30aa\u30d6\u30b8\u30a7\u30af\u30c8\u306b\u5909\u63db\u3057\u307e\u3059\u3002<code>id_<\/code>\u306b\u306fAPI\u306eURL\u3001<code>text<\/code>\u306b\u306f\u53d6\u5f97\u3057\u305fJSON\u30c7\u30fc\u30bf\u304c\u542b\u307e\u308c\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>JSONNodeParser()<\/code>: <code>JSONNodeParser<\/code>\u306e\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u3092\u4f5c\u6210\u3057\u307e\u3059\u3002\u3053\u306e\u30d1\u30fc\u30b5\u3092\u4f7f\u7528\u3057\u3066\u3001\u53d6\u5f97\u3057\u305fJSON\u30c7\u30fc\u30bf\u3092\u30ce\u30fc\u30c9\u306b\u5206\u5272\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>get_nodes_from_documents([document])<\/code>: JSON\u30c7\u30fc\u30bf\u304c\u683c\u7d0d\u3055\u308c\u305f<code>Document<\/code>\u30aa\u30d6\u30b8\u30a7\u30af\u30c8\u3092\u6e21\u3057\u3001\u305d\u306e\u5185\u5bb9\u3092\u30ce\u30fc\u30c9\u306b\u5206\u5272\u3057\u307e\u3059\u3002\u5206\u5272\u3055\u308c\u305f\u7d50\u679c\u306f<code>nodes<\/code>\u306b\u4fdd\u5b58\u3055\u308c\u307e\u3059\u3002<\/li>\n<\/ul>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">Markdown NodeParser<\/h3>\n\n\n\n<p>MarkdownNodeParser \u306f\u3001Markdown\u5f62\u5f0f\u306e\u30c7\u30fc\u30bf\u3092\u89e3\u6790\u3057\u3001\u30ce\u30fc\u30c9\uff08\u30c1\u30e3\u30f3\u30af\uff09\u306b\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u30af\u30e9\u30b9\u3067\u3059\u3002<\/p>\n\n\n\n<p>\u3053\u306e\u30d1\u30fc\u30b5\u30fc\u306f\u3001Markdown\u306e\u69cb\u9020\u3092\u7406\u89e3\u3057\u3001\u898b\u51fa\u3057\u3084\u6bb5\u843d\u306a\u3069\u306e\u8981\u7d20\u306b\u57fa\u3065\u3044\u3066\u30c7\u30fc\u30bf\u3092\u5206\u5272\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code># MarkdownNodeParser\nfrom llama_index.core.node_parser import MarkdownNodeParser\n\nmd_docs = FlatReader().load_data(Path(\"README.md\"))\nparser = MarkdownNodeParser()\n\nnodes = parser.get_nodes_from_documents(md_docs)\nnodes&#091;0].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'&#x1f5c2;&#xfe0f; LlamaIndex &#x1f999;\\n\\n&#091;!&#091;PyPI - Downloads](https:\/\/img.shields.io\/pypi\/dm\/llama-index)](https:\/\/pypi.org\/project\/llama-index\/)\\n&#091;!&#091;GitHub contributors](https:\/\/img.shields.io\/github\/contributors\/jerryjliu\/llama_index)](https:\/\/github.com\/jerryjliu\/llama_index\/graphs\/contributors)\\n&#091;!&#091;Discord](https:\/\/img.shields.io\/discord\/1059199217496772688)](https:\/\/discord.gg\/dGcwcsnxhU)\\n&#091;!&#091;Ask AI](https:\/\/img.shields.io\/badge\/Phorm-Ask_AI-%23F2777A.svg?&amp;logo=data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iNSIgaGVpZ2h0PSI0IiBmaWxsPSJub25lIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPgogIDxwYXRoIGQ9Ik00LjQzIDEuODgyYTEuNDQgMS40NCAwIDAgMS0uMDk4LjQyNmMtLjA1LjEyMy0uMTE1LjIzLS4xOTIuMzIyLS4wNzUuMDktLjE2LjE2NS0uMjU1LjIyNmExLjM1MyAxLjM1MyAwIDAgMS0uNTk1LjIxMmMtLjA5OS4wMTItLjE5Mi4wMTQtLjI3OS4wMDZsLTEuNTkzLS4xNHYtLjQwNmgxLjY1OGMuMDkuMDAxLjE3LS4xNjkuMjQ2LS4xOTFhLjYwMy42MDMgMCAwIDAgLjItLjEwNi41MjkuNTI5IDAgMCAwIC4xMzgtLjE3LjY1NC42NTQgMCAwIDAgLjA2NS0uMjRsLjAyOC0uMzJhLjkzLjkzIDAgMCAwLS4wMzYtLjI0OS41NjcuNTY3IDAgMCAwLS4xMDMtLjIuNTAyLjUwMiAwIDAgMC0uMTY4LS4xMzguNjA4LjYwOCAwIDAgMC0uMjQtLjA2N0wyLjQzNy43MjkgMS42MjUuNjcxYS4zMjIuMzIyIDAgMCAwLS4yMzIuMDU4LjM3NS4zNzUgMCAwIDAtLjExNi4yMzJsLS4xMTYgMS40NS0uMDU4LjY5Ny0uMDU4Ljc1NEwuNzA1IDRsLS4zNTctLjA3OUwuNjAyLjkwNkMuNjE3LjcyNi42NjMuNTc0LjczOS40NTRhLjk1OC45NTggMCAwIDEgLjI3NC0uMjg1Ljk3MS45NzEgMCAwIDEgLjMzNy0uMTRjLjExOS0uMDI2LjIyNy0uMDM0LjMyNS0uMDI2TDMuMjMyLjE2Yy4xNTkuMDE0LjMzNi4wMy40NTkuMDgyYTEuMTczIDEuMTczIDAgMCAxIC41NDUuNDQ3Yy4wNi4wOTQuMTA5LjE5Mi4xNDQuMjkzYTEuMzkyIDEuMzkyIDAgMCAxIC4wNzguNThsLS4wMjkuMzJaIiBmaWxsPSIjRjI3NzdBIi8+CiAgPHBhdGggZD0iTTQuMDgyIDIuMDA3YTEuNDU1IDEuNDU1IDAgMCAxLS4wOTguNDI3Yy0uMDUuMTI0LS4xMTQuMjMyLS4xOTIuMzI0YTEuMTMgMS4xMyAwIDAgMS0uMjU0LjIyNyAxLjM1MyAxLjM1MyAwIDAgMS0uNTk1LjIxNGMtLjEuMDEyLS4xOTMuMDE0LS4yOC4wMDZsLTEuNTYtLjEwOC4wMzQtLjQwNi4wMy0uMzQ4IDEuNTU5LjE1NGMuMDkgMCAuMTczLS4wMS4yNDgtLjAzM2EuNjAzLjYwMyAwIDAgMCAuMi0uMTA2LjUzMi41MzIgMCAwIDAgLjEzOS0uMTcyLjY2LjY2IDAgMCAwIC4wNjQtLjI0MWwuMDI5LS4zMjFhLjk0Ljk0IDAgMCAwLS4wMzYtLjI1LjU3LjU3IDAgMCAwLS4xMDMtLjIwMi41MDIuNTAyIDAgMCAwLS4xNjgtLjEzOC42MDUuNjA1IDAgMCAwLS4yNC0uMDY3TDEuMjczLjgyN2MtLjA5NC0uMDA4LS4xNjguMDEtLjIyMS4wNTUtLjA1My4wNDUtLjA4NC4xMTQtLjA5Mi4yMDZMLjcwNSA0IDAgMy45MzhsLjI1NS0yLjkxMUExLjAxIDEuMDEgMCAwIDEgLjM5My41NzIuOTYyLjk2MiAwIDAgMSAuNjY2LjI4NmEuOTcuOTcgMCAwIDEgLjMzOC0uMTRDMS4xMjIuMTIgMS4yMy4xMSAxLjMyOC4xMTlsMS41OTMuMTRjLjE2LjAxNC4zLjA0Ny40MjMuMWExLjE3IDEuMTcgMCAwIDEgLjU0NS40NDhjLjA2MS4wOTUuMTA5LjE5My4xNDQuMjk1YTEuNDA2IDEuNDA2IDAgMCAxIC4wNzcuNTgzbC0uMDI4LjMyMloiIGZpbGw9IndoaXRlIi8+CiAgPHBhdGggZD0iTTQuMDgyIDIuMDA3YTEuNDU1IDEuNDU1IDAgMCAxLS4wOTguNDI3Yy0uMDUuMTI0LS4xMTQuMjMyLS4xOTIuMzI0YTEuMTMgMS4xMyAwIDAgMS0uMjU0LjIyNyAxLjM1MyAxLjM1MyAwIDAgMS0uNTk1LjIxNGMtLjEuMDEyLS4xOTMuMDE0LS4yOC4wMDZsLTEuNTYtLjEwOC4wMzQtLjQwNi4wMy0uMzQ4IDEuNTU5LjE1NGMuMDkgMCAuMTczLS4wMS4yNDgtLjAzM2EuNjAzLjYwMyAwIDAgMCAuMi0uMTA2LjUzMi41MzIgMCAwIDAgLjEzOS0uMTcyLjY2LjY2IDAgMCAwIC4wNjQtLjI0MWwuMDI5LS4zMjFhLjk0Ljk0IDAgMCAwLS4wMzYtLjI1LjU3LjU3IDAgMCAwLS4xMDMtLjIwMi41MDIuNTAyIDAgMCAwLS4xNjgtLjEzOC42MDUuNjA1IDAgMCAwLS4yNC0uMDY3TDEuMjczLjgyN2MtLjA5NC0uMDA4LS4xNjguMDEtLjIyMS4wNTUtLjA1My4wNDUtLjA4NC4xMTQtLjA5Mi4yMDZMLjcwNSA0IDAgMy45MzhsLjI1NS0yLjkxMUExLjAxIDEuMDEgMCAwIDEgLjM5My41NzIuOTYyLjk2MiAwIDAgMSAuNjY2LjI4NmEuOTcuOTcgMCAwIDEgLjMzOC0uMTRDMS4xMjIuMTIgMS4yMy4xMSAxLjMyOC4xMTlsMS41OTMuMTRjLjE2LjAxNC4zLjA0Ny40MjMuMWExLjE3IDEuMTcgMCAwIDEgLjU0NS40NDhjLjA2MS4wOTUuMTA5LjE5My4xNDQuMjk1YTEuNDA2IDEuNDA2IDAgMCAxIC4wNzcuNTgzbC0uMDI4LjMyMloiIGZpbGw9IndoaXRlIi8+Cjwvc3ZnPgo=)](https:\/\/www.phorm.ai\/query?projectId=c5863b56-6703-4a5d-87b6-7e6031bf16b6)\\n\\nLlamaIndex (GPT Index) is a data framework for your LLM application. Building with LlamaIndex typically involves working with LlamaIndex core and a chosen set of integrations (or plugins). There are two ways to start building with LlamaIndex in\\nPython:\\n\\n1. **Starter**: `llama-index` (https:\/\/pypi.org\/project\/llama-index\/). A starter Python package that includes core LlamaIndex as well as a selection of integrations.\\n\\n2. **Customized**: `llama-index-core` (https:\/\/pypi.org\/project\/llama-index-core\/). Install core LlamaIndex and add your chosen LlamaIndex integration packages on \n\uff1c\u4ee5\u4e0b\u7701\u7565\uff1e<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">md_docs = FlatReader().load_data(Path(&#8220;README.md&#8221;))<\/span><\/p>\n\n\n\n<p>\u30d5\u30a1\u30a4\u30eb\u3092\u8aad\u307f\u8fbc\u3080\u305f\u3081\u306e\u30e1\u30bd\u30c3\u30c9\u3067\u3059\u3002\u3053\u3053\u3067\u306f <code>README.md<\/code> \u30d5\u30a1\u30a4\u30eb\u306e\u5185\u5bb9\u3092\u8aad\u307f\u8fbc\u307f\u3001\u305d\u306e\u5185\u5bb9\u3092 <code>md_docs<\/code> \u306b\u4fdd\u5b58\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">parser = MarkdownNodeParser()<\/span><\/p>\n\n\n\n<p><code>MarkdownNodeParser<\/code>\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u3092\u4f7f\u3063\u3066\u3001Markdown\u5f62\u5f0f\u306e\u30c7\u30fc\u30bf\u3092\u30ce\u30fc\u30c9\uff08\u30c1\u30e3\u30f3\u30af\uff09\u306b\u5206\u5272\u3059\u308b\u6e96\u5099\u3092\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">nodes = parser.get_nodes_from_documents(md_docs)<\/span><\/p>\n\n\n\n<p>\u8aad\u307f\u8fbc\u3093\u3060 Markdown \u30d5\u30a1\u30a4\u30eb\u306e\u5185\u5bb9 (<code>md_docs<\/code>) \u3092\u89e3\u6790\u3057\u3001\u30ce\u30fc\u30c9\u306b\u5206\u5272\u3057\u307e\u3059\u3002\u30d5\u30a1\u30a4\u30eb\u5185\u306e\u69cb\u9020\uff08\u898b\u51fa\u3057\u3001\u6bb5\u843d\u306a\u3069\uff09\u306b\u57fa\u3065\u3044\u3066\u30c6\u30ad\u30b9\u30c8\u3092\u30c1\u30e3\u30f3\u30af\u306b\u5206\u5272\u3057\u307e\u3059\u3002<code>nodes<\/code>\u306b\u5206\u5272\u3055\u308c\u305f\u30ce\u30fc\u30c9\u304c\u30ea\u30b9\u30c8\u5f62\u5f0f\u3067\u683c\u7d0d\u3055\u308c\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">\u9577\u3055\u30d9\u30fc\u30b9\u306eNode Parser<\/h2>\n\n\n\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" width=\"600\" height=\"300\" src=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/08\/cat15_midashi.jpg\" alt=\"\u898b\u51fa\u3057\u753b\u50cf\" class=\"wp-image-9908\" srcset=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/08\/cat15_midashi.jpg 600w, https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/08\/cat15_midashi-300x150.jpg 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/><\/figure>\n\n\n\n<p>\u9577\u3055\u306b\u57fa\u3065\u304f\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u306f\u3001\u6307\u5b9a\u3055\u308c\u305f\u6587\u5b57\u6570\u3084\u30c8\u30fc\u30af\u30f3\u6570\u306b\u5f93\u3063\u3066\u30c6\u30ad\u30b9\u30c8\u3092\u5206\u5272\u3059\u308b\u624b\u6cd5\u3067\u3059\u3002<\/p>\n\n\n\n<p>\u3053\u308c\u3089\u306f\u30b3\u30fc\u30c9\u3084\u30c6\u30ad\u30b9\u30c8\u3092\u6307\u5b9a\u3055\u308c\u305f\u884c\u6570\u3084\u6587\u5358\u4f4d\u3001\u6587\u5b57\u6570\u306b\u57fa\u3065\u3044\u3066\u5206\u5272\u3057\u3001\u5404\u30c1\u30e3\u30f3\u30af\u306e\u9593\u3067\u5fc5\u8981\u306b\u5fdc\u3058\u3066\u91cd\u8907\u3092\u8ffd\u52a0\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u9577\u3055\u306b\u57fa\u3065\u304f\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u306f\u4ee5\u4e0b\u306e\u3088\u3046\u306a\u3082\u306e\u304c\u3042\u308a\u307e\u3059\u3002<\/p>\n\n\n\n<div class=\"wp-block-group is-style-big_icon_good\"><div class=\"wp-block-group__inner-container\">\n<ul class=\"wp-block-list\">\n<li>Code Splitter<\/li>\n\n\n\n<li>Sentence Splitter<\/li>\n\n\n\n<li>TokenText Splitter<\/li>\n\n\n\n<li>SentenceWindowNodeParser<\/li>\n<\/ul>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">\u4e8b\u524d\u6e96\u5099<\/h3>\n\n\n\n<p>\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>pip install llama_index llama-index-core<\/code><\/pre>\n\n\n\n<h3 class=\"wp-block-heading\">Code Splitter<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">CodeSplitter \u306f\u3001\u30b3\u30fc\u30c9\u3092\u89e3\u6790\u3057\u3001\u7279\u5b9a\u306e\u8a00\u8a9e\u306b\u57fa\u3065\u3044\u3066\u30c1\u30e3\u30f3\u30af\uff08\u30ce\u30fc\u30c9\uff09\u306b\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u30af\u30e9\u30b9\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>Python\u3084JavaScript\u306a\u3069\u3001\u3055\u307e\u3056\u307e\u306a\u30d7\u30ed\u30b0\u30e9\u30df\u30f3\u30b0\u8a00\u8a9e\u306e\u69cb\u9020\u306b\u57fa\u3065\u3044\u3066\u30b3\u30fc\u30c9\u3092\u5206\u5272\u3067\u304d\u3001\u30b3\u30fc\u30c9\u89e3\u6790\u3084\u691c\u7d22\u3001\u51e6\u7406\u3092\u884c\u3046\u969b\u306b\u5f79\u7acb\u3061\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>pip install -U \"tree-sitter&lt;0.22.0\" \"tree_sitter_languages&lt;1.10.0\"<\/code><\/pre>\n\n\n\n\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>!curl -O https:\/\/raw.githubusercontent.com\/lancedb\/vectordb-recipes\/main\/applications\/talk-with-podcast\/app.py<\/code><\/pre>\n\n\n\n\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code># CodeSplitter\nfrom llama_index.core.node_parser import CodeSplitter\n\ndocuments = FlatReader().load_data(Path(\"sentence_transformer.py\"))\nsplitter = CodeSplitter(\n    language=\"python\",\n    chunk_lines=40,\n    chunk_lines_overlap=15,\n    max_chars=1500,\n)\nnodes = splitter.get_nodes_from_documents(documents)\nnodes&#091;0].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'\"\"\"Sentence Transformer Finetuning Engine.\"\"\"\\n\\nfrom typing import Any, Optional\\n\\nfrom llama_index.core.base.embeddings.base import BaseEmbedding\\nfrom llama_index.core.embeddings.utils import resolve_embed_model\\nfrom llama_index.finetuning.embeddings.common import (\\n    EmbeddingQAFinetuneDataset,\\n)\\nfrom llama_index.finetuning.types import BaseEmbeddingFinetuneEngine'<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">documents = FlatReader().load_data(Path(&#8220;sentence_transformer.py&#8221;))<\/span><\/p>\n\n\n\n<p>\u6307\u5b9a\u3057\u305f\u30b3\u30fc\u30c9\u30d5\u30a1\u30a4\u30eb\u3092\u8aad\u307f\u8fbc\u3080\u305f\u3081\u306e\u30e1\u30bd\u30c3\u30c9\u3067\u3059\u3002\u3053\u3053\u3067\u306f\u3001<code>sentence_transformer.py<\/code> \u3068\u3044\u3046Python\u30d5\u30a1\u30a4\u30eb\u3092\u8aad\u307f\u8fbc\u307f\u3001\u305d\u306e\u5185\u5bb9\u3092 <code>documents<\/code> \u306b\u683c\u7d0d\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">splitter = CodeSplitter(&#8230;<\/span><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>language=\"python\"<\/code>: \u5206\u5272\u5bfe\u8c61\u306e\u8a00\u8a9e\u3092Python\u306b\u6307\u5b9a\u3057\u3066\u3044\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>chunk_lines=40<\/code>: 1\u30c1\u30e3\u30f3\u30af\u306b\u542b\u307e\u308c\u308b\u30b3\u30fc\u30c9\u306e\u884c\u6570\u3092\u6307\u5b9a\u3057\u307e\u3059\u3002\u3053\u3053\u3067\u306f\u300140\u884c\u3054\u3068\u306b\u30b3\u30fc\u30c9\u3092\u30c1\u30e3\u30f3\u30af\u306b\u5206\u5272\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>chunk_lines_overlap=15<\/code>: \u5404\u30c1\u30e3\u30f3\u30af\u9593\u3067\u91cd\u8907\u3059\u308b\u884c\u6570\u3092\u6307\u5b9a\u3057\u3066\u3044\u307e\u3059\u3002\u3053\u3053\u3067\u306f\u300115\u884c\u306e\u30b3\u30fc\u30c9\u304c\u6b21\u306e\u30c1\u30e3\u30f3\u30af\u306b\u3082\u91cd\u8907\u3057\u3066\u542b\u307e\u308c\u308b\u8a2d\u5b9a\u3067\u3059\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u30b3\u30fc\u30c9\u306e\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u304c\u9014\u5207\u308c\u308b\u3053\u3068\u306a\u304f\u6b21\u306e\u30c1\u30e3\u30f3\u30af\u306b\u7d9a\u304f\u5f62\u306b\u306a\u308a\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>max_chars=1500<\/code>: 1\u30c1\u30e3\u30f3\u30af\u306e\u6700\u5927\u6587\u5b57\u6570\u3092\u6307\u5b9a\u3057\u307e\u3059\u3002\u6587\u5b57\u6570\u304c1,500\u6587\u5b57\u3092\u8d85\u3048\u305f\u5834\u5408\u3001\u3055\u3089\u306b\u5206\u5272\u304c\u884c\u308f\u308c\u307e\u3059\u3002\u3053\u306e\u8a2d\u5b9a\u306b\u3088\u308a\u3001\u30c1\u30e3\u30f3\u30af\u306e\u30b5\u30a4\u30ba\u304c\u5236\u5fa1\u3055\u308c\u307e\u3059\u3002<\/li>\n<\/ul>\n\n\n\n<p><span class=\"swl-format-2\">nodes = splitter.get_nodes_from_documents(documents)<\/span><\/p>\n\n\n\n<p>\u3053\u306e\u30e1\u30bd\u30c3\u30c9\u306f\u3001\u8aad\u307f\u8fbc\u3093\u3060 <code>documents<\/code>\u3092\u30c1\u30e3\u30f3\u30af\u306b\u5206\u5272\u3057\u307e\u3059\u3002\u6307\u5b9a\u3055\u308c\u305f\u30d1\u30e9\u30e1\u30fc\u30bf\u306b\u57fa\u3065\u3044\u3066\u3001\u30b3\u30fc\u30c9\u3092\u5206\u5272\u3057\u3001\u30ce\u30fc\u30c9\u30ea\u30b9\u30c8\u3092\u4f5c\u6210\u3057\u307e\u3059\u3002 \u5206\u5272\u3055\u308c\u305f\u30c1\u30e3\u30f3\u30af\uff08\u30ce\u30fc\u30c9\uff09\u304c <code>nodes<\/code> \u3068\u3044\u3046\u30ea\u30b9\u30c8\u306b\u683c\u7d0d\u3055\u308c\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">Sentence Splitter<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">SentenceSplitter \u306f\u3001LlamaIndex\u3067\u30c6\u30ad\u30b9\u30c8\u3092<\/span><span class=\"swl-format-2\">\u6587\u5358\u4f4d<\/span><span class=\"swl-format-2\">\u306b\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u30af\u30e9\u30b9\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u30c6\u30ad\u30b9\u30c8\u3092\u6587\u3054\u3068\u306b\u5206\u5272\u3057\u3001\u30ce\u30fc\u30c9\uff08\u30c1\u30e3\u30f3\u30af\uff09\u3068\u3057\u3066\u6271\u3046\u969b\u306b\u975e\u5e38\u306b\u5f79\u7acb\u3061\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u7279\u306b\u5927\u898f\u6a21\u306a\u30c6\u30ad\u30b9\u30c8\u30c7\u30fc\u30bf\u3092\u51e6\u7406\u3059\u308b\u969b\u3001\u81ea\u7136\u306a\u6587\u306e\u533a\u5207\u308a\u3092\u4fdd\u6301\u3057\u306a\u304c\u3089\u30c6\u30ad\u30b9\u30c8\u3092\u52b9\u7387\u7684\u306b\u51e6\u7406\u3059\u308b\u305f\u3081\u306b\u4f7f\u308f\u308c\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code># SentenceSplitter\nfrom llama_index.core.node_parser import SentenceSplitter\n\ndocuments = FlatReader().load_data(Path(\"\u7b2c1\u7ae0_\u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069.txt\"))\nsplitter = SentenceSplitter(\n    chunk_size=254,\n    chunk_overlap=20,\n)\nnodes = splitter.get_nodes_from_documents(documents)\nnodes&#091;0].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'\u7b2c1\u7ae0 \u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069\\n\\t \u7b2c1\u7bc0\\t \u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u5f85\u9047\u6539\u5584\u3001\u9577\u6642\u9593\u52b4\u50cd\u306e\u662f\u6b63\u7b49\\n\\t1\\t\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u5bfe\u7b56\\n\uff081\uff09\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u8ab2\u984c\\n\u8fd1\u5e74\u3001\u30d1\u30fc\u30c8\u30bf\u30a4\u30e0\u52b4\u50cd\u8005\u3001\u6709\u671f\u96c7\u7528\u52b4\u50cd\u8005\u3001\u6d3e\u9063\u52b4\u50cd\u8005\u3068\u3044\u3063\u305f\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306f\\n\u5168\u4f53\u3068\u3057\u3066\u5897\u52a0\u50be\u5411\u306b\u3042\u308a\u3001\u96c7\u7528\u8005\u306e\u7d044\u5272\u3092\u5360\u3081\u308b\u72b6\u6cc1\u306b\u3042\u308b\u3002'<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">documents = FlatReader().load_data()<\/span><\/p>\n\n\n\n<p>\u6307\u5b9a\u3055\u308c\u305f\u30c6\u30ad\u30b9\u30c8\u30d5\u30a1\u30a4\u30eb\u3092\u8aad\u307f\u8fbc\u307f\u3001\u305d\u306e\u5185\u5bb9\u3092\u30c6\u30ad\u30b9\u30c8\u30c7\u30fc\u30bf\u3068\u3057\u3066\u53d6\u5f97\u3057\u307e\u3059\u3002\u3053\u306e\u30c6\u30ad\u30b9\u30c8\u30c7\u30fc\u30bf\u306f <code>documents<\/code> \u306b\u683c\u7d0d\u3055\u308c\u3001\u5f8c\u306b<code>SentenceSplitter<\/code> \u3067\u51e6\u7406\u3055\u308c\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">splitter = SentenceSplitter(&#8230;<\/span><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>chunk_size=254<\/code>: 1\u3064\u306e\u30c1\u30e3\u30f3\u30af\u306b\u542b\u307e\u308c\u308b\u6700\u5927\u6587\u5b57\u6570\u3092\u6307\u5b9a\u3057\u307e\u3059\u3002\u3053\u3053\u3067\u306f254\u6587\u5b57\u3054\u3068\u306b\u30c1\u30e3\u30f3\u30af\u306b\u5206\u5272\u3055\u308c\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>chunk_overlap=20<\/code>: \u5404\u30c1\u30e3\u30f3\u30af\u9593\u3067\u91cd\u8907\u3059\u308b\u6587\u5b57\u6570\u3092\u6307\u5b9a\u3057\u307e\u3059\u3002\u3053\u3053\u3067\u306f\u300120\u6587\u5b57\u5206\u304c\u6b21\u306e\u30c1\u30e3\u30f3\u30af\u306b\u91cd\u8907\u3057\u3066\u542b\u307e\u308c\u308b\u3088\u3046\u306b\u8a2d\u5b9a\u3055\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u30c1\u30e3\u30f3\u30af\u9593\u3067\u6587\u8108\u304c\u9014\u5207\u308c\u308b\u3053\u3068\u306a\u304f\u51e6\u7406\u3067\u304d\u307e\u3059\u3002<\/li>\n<\/ul>\n\n\n\n<p><span class=\"swl-format-2\">nodes = splitter.get_nodes_from_documents(documents)<\/span><\/p>\n\n\n\n<p>\u8aad\u307f\u8fbc\u3093\u3060\u30c6\u30ad\u30b9\u30c8\u30c7\u30fc\u30bf<code>documents<\/code>\u3092\u6587\u5358\u4f4d\u3067\u5206\u5272\u3057\u3001\u305d\u308c\u305e\u308c\u3092\u30c1\u30e3\u30f3\u30af\u3068\u3057\u3066\u30ce\u30fc\u30c9\u306b\u5909\u63db\u3057\u307e\u3059\u3002\u5404\u30ce\u30fc\u30c9\u306b\u306f\u3001\u6307\u5b9a\u3055\u308c\u305f <code>chunk_size<\/code> \u3068 <code>chunk_overlap<\/code> \u306b\u57fa\u3065\u3044\u3066\u5206\u5272\u3055\u308c\u305f\u30c6\u30ad\u30b9\u30c8\u304c\u542b\u307e\u308c\u307e\u3059\u3002\u7d50\u679c\u306f <code>nodes<\/code> \u3068\u3044\u3046\u30ea\u30b9\u30c8\u306b\u683c\u7d0d\u3055\u308c\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">TokenText Splitter<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">TokenTextSplitter\u306f\u3001\u30c6\u30ad\u30b9\u30c8\u3092\u30c8\u30fc\u30af\u30f3\u6570\u306b\u57fa\u3065\u3044\u3066\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u6a5f\u80fd\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u591a\u304f\u306eLLM\u306f\u3001\u7279\u5b9a\u306e\u30c8\u30fc\u30af\u30f3\u6570\uff08\u5358\u8a9e\u3084\u6587\u306e\u65ad\u7247\u306a\u3069\uff09\u3092\u8d85\u3048\u308b\u3068\u51e6\u7406\u3067\u304d\u306a\u304f\u306a\u308a\u307e\u3059\u3002\u3053\u306e\u5236\u9650\u3092\u8003\u616e\u3057\u3066\u3001\u30c6\u30ad\u30b9\u30c8\u3092\u6307\u5b9a\u3055\u308c\u305f\u30c8\u30fc\u30af\u30f3\u6570\u306b\u57fa\u3065\u3044\u3066\u30c1\u30e3\u30f3\u30af\u306b\u5206\u5272\u3059\u308b\u5f79\u5272\u3092\u679c\u305f\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code># TokenTextSplitting\nfrom llama_index.core.node_parser import TokenTextSplitter\n\nsplitter = TokenTextSplitter(\n    chunk_size=254,\n    chunk_overlap=20,\n    separator=\" \",\n)\nnodes = splitter.get_nodes_from_documents(documents)\nnodes&#091;0].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'\u7b2c1\u7ae0 \u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069\\n\\t \u7b2c1\u7bc0\\t \u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u5f85\u9047\u6539\u5584\u3001\u9577\u6642\u9593\u52b4\u50cd\u306e\u662f\u6b63\u7b49\\n\\t1\\t\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u5bfe\u7b56\\n\uff081\uff09\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u8ab2\u984c\\n\u8fd1\u5e74\u3001\u30d1\u30fc\u30c8\u30bf\u30a4\u30e0\u52b4\u50cd\u8005\u3001\u6709\u671f\u96c7\u7528\u52b4\u50cd\u8005\u3001\u6d3e\u9063\u52b4\u50cd\u8005\u3068\u3044\u3063\u305f\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306f'<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">splitter = TokenTextSplitter(&#8230;<\/span><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>chunk_size=254<\/code>: 1\u3064\u306e\u30c1\u30e3\u30f3\u30af\u306b\u542b\u307e\u308c\u308b\u30c8\u30fc\u30af\u30f3\u306e\u6700\u5927\u6570\u3092254\u30c8\u30fc\u30af\u30f3\u306b\u8a2d\u5b9a\u3057\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u30c6\u30ad\u30b9\u30c8\u304c254\u30c8\u30fc\u30af\u30f3\u305a\u3064\u5206\u5272\u3055\u308c\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>chunk_overlap=20<\/code>: \u96a3\u63a5\u3059\u308b\u30c1\u30e3\u30f3\u30af\u9593\u306720\u30c8\u30fc\u30af\u30f3\u306e\u91cd\u8907\u304c\u8a2d\u5b9a\u3055\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u6587\u8108\u306e\u9023\u7d9a\u6027\u3092\u4fdd\u3064\u3053\u3068\u304c\u3067\u304d\u3001\u30e2\u30c7\u30eb\u304c\u3088\u308a\u610f\u5473\u3092\u7406\u89e3\u3057\u3084\u3059\u304f\u306a\u308a\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>separator=\" \"<\/code>: \u30c8\u30fc\u30af\u30f3\u3092\u5206\u5272\u3059\u308b\u969b\u306e\u533a\u5207\u308a\u6587\u5b57\u3068\u3057\u3066\u3001\u30b9\u30da\u30fc\u30b9\uff08<code>\" \"<\/code>\uff09\u3092\u4f7f\u7528\u3057\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306f\u3001\u4e00\u822c\u7684\u306b\u5358\u8a9e\u306e\u533a\u5207\u308a\u3092\u610f\u5473\u3057\u307e\u3059\u3002<\/li>\n<\/ul>\n\n\n\n<p><span class=\"swl-format-2\">nodes = splitter.get_nodes_from_documents(documents)<\/span><\/p>\n\n\n\n<p>\u6307\u5b9a\u3057\u305f<code>documents<\/code>\u3092\u30c8\u30fc\u30af\u30f3\u5358\u4f4d\u3067\u5206\u5272\u3057\u3001\u305d\u308c\u305e\u308c\u3092\u30ce\u30fc\u30c9\u306b\u5909\u63db\u3057\u307e\u3059\u3002\u30ce\u30fc\u30c9\u306f\u30c6\u30ad\u30b9\u30c8\u306e\u4e00\u90e8\u5206\u3092\u4fdd\u6301\u3057\u3001\u5f8c\u3067\u30e2\u30c7\u30eb\u306b\u5165\u529b\u3059\u308b\u305f\u3081\u306e\u30c1\u30e3\u30f3\u30af\u3068\u306a\u308a\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">SentenceWindow NodeParser<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">SentenceWindowNodeParser \u306f\u3001\u30c6\u30ad\u30b9\u30c8\u3092\u6587\u5358\u4f4d\u3067\u5206\u5272\u3057\u3001\u3055\u3089\u306b\u96a3\u63a5\u3059\u308b\u6587\u3092\u300c\u30a6\u30a3\u30f3\u30c9\u30a6\u300d\u3068\u3057\u3066\u30b0\u30eb\u30fc\u30d7\u5316\u3059\u308b\u305f\u3081\u306e\u30af\u30e9\u30b9\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u901a\u5e38\u306e\u6587\u5358\u4f4d\u306e\u5206\u5272\u306b\u52a0\u3048\u3001\u524d\u5f8c\u306e\u6587\u8108\u3092\u4fdd\u6301\u3059\u308b\u305f\u3081\u306b\u300c\u30a6\u30a3\u30f3\u30c9\u30a6\u300d\u3092\u4f5c\u308a\u51fa\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u5404\u30ce\u30fc\u30c9\u304c\u6301\u3064\u60c5\u5831\u306b\u306f\u3001\u305d\u306e\u30ce\u30fc\u30c9\u5185\u306e\u6587\u306b\u52a0\u3048\u3001\u524d\u5f8c\u306e\u6587\u3082\u542b\u307e\u308c\u308b\u305f\u3081\u3001\u6587\u8108\u3092\u9014\u5207\u308c\u3055\u305b\u305a\u306b\u51e6\u7406\u304c\u53ef\u80fd\u3067\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code># SentenceWindowNodeParser\nimport nltk\nfrom llama_index.core.node_parser import SentenceWindowNodeParser\n\nnode_parser = SentenceWindowNodeParser.from_defaults(\n    window_size=3,\n    window_metadata_key=\"window\",\n    original_text_metadata_key=\"original_sentence\",\n)\nnodes = node_parser.get_nodes_from_documents(documents)\nnodes&#091;1].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'\u8133\u30fb\u5fc3\u81d3\u75be\u60a3\u3068\u306f\u3001\u696d\u52d9\u306b\u304a\u3051\u308b\u904e\u91cd\u8ca0\u8377\u306b\u3088\u308a\u8133\u30fb\u5fc3\u81d3\u75be\u60a3\uff08\u8ca0\u50b7\u306b\u8d77\u56e0\u3059\u308b\u3082\u306e\u3092\u9664\u304f\u3002\uff09\u3092\u767a\u75c7\u3057\u305f\u4e8b\u6848\uff08\u6b7b\\n\u4ea1\u3092\u542b\u3080\u3002\uff09\u3092\u3044\u3046\u3002\\n\uff12. '<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">node_parser = SentenceWindowNodeParser.from_defaults(&#8230;<\/span><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>window_size=3<\/code>: \u5404\u30ce\u30fc\u30c9\u306b\u542b\u307e\u308c\u308b\u6587\u306e\u6570\u30923\u306b\u8a2d\u5b9a\u3057\u3066\u3044\u307e\u3059\u3002\u3053\u306e\u8a2d\u5b9a\u306b\u3088\u308a\u3001\u9023\u7d9a\u3059\u308b3\u6587\u304c1\u3064\u306e\u30ce\u30fc\u30c9\uff08\u30a6\u30a3\u30f3\u30c9\u30a6\uff09\u3068\u3057\u3066\u51e6\u7406\u3055\u308c\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>window_metadata_key=\"window\"<\/code>: \u5404\u30ce\u30fc\u30c9\u306b\u30a6\u30a3\u30f3\u30c9\u30a6\u306b\u95a2\u3059\u308b\u30e1\u30bf\u30c7\u30fc\u30bf\u304c\u4fdd\u5b58\u3055\u308c\u307e\u3059\u3002\u30e1\u30bf\u30c7\u30fc\u30bf\u30ad\u30fc\u306f\u300cwindow\u300d\u3067\u3059\u3002<\/li>\n\n\n\n<li><code>original_text_metadata_key=\"original_sentence\"<\/code>: \u5143\u306e\u6587\u306b\u95a2\u3059\u308b\u30e1\u30bf\u30c7\u30fc\u30bf\u3092\u8ffd\u52a0\u3057\u307e\u3059\u3002\u30e1\u30bf\u30c7\u30fc\u30bf\u30ad\u30fc\u306f\u300coriginal_sentence\u300d\u3067\u3059\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u5404\u30ce\u30fc\u30c9\u304c\u5143\u306e\u3069\u306e\u6587\u304b\u3089\u4f5c\u6210\u3055\u308c\u305f\u304b\u304c\u8ffd\u8de1\u53ef\u80fd\u3067\u3059\u3002<\/li>\n<\/ul>\n\n\n\n<p><span class=\"swl-format-2\">nodes = node_parser.get_nodes_from_documents(documents)<\/span><\/p>\n\n\n\n<p><code>documents<\/code> \u304b\u3089\u30c6\u30ad\u30b9\u30c8\u3092\u53d6\u5f97\u3057\u3001\u6587\u5358\u4f4d\u3067\u5206\u5272\u3057\u3066\u30ce\u30fc\u30c9\u306b\u5909\u63db\u3057\u307e\u3059\u3002\u3053\u306e\u5834\u5408\u3001<code>window_size=3<\/code>\u306b\u57fa\u3065\u3044\u3066\u3001\u5404\u30ce\u30fc\u30c9\u306b\u306f3\u3064\u306e\u9023\u7d9a\u3057\u305f\u6587\u304c\u542b\u307e\u308c\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">\u610f\u5473\u30d9\u30fc\u30b9\u306eNode Parser<\/h2>\n\n\n\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" width=\"600\" height=\"300\" src=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/09\/17_midashi_anime0921.jpg\" alt=\"\u898b\u51fa\u3057\u753b\u50cf\" class=\"wp-image-11898\" srcset=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/09\/17_midashi_anime0921.jpg 600w, https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/09\/17_midashi_anime0921-300x150.jpg 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/><\/figure>\n\n\n\n<p><span class=\"swl-format-2\">\u610f\u5473\u30d9\u30fc\u30b9\u306e\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u306f\u3001\u30c6\u30ad\u30b9\u30c8\u306e\u610f\u5473\u7684\u306a\u69cb\u9020\u306b\u57fa\u3065\u3044\u3066\u5206\u5272\u3092\u884c\u3046\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u5358\u306b\u6587\u3084\u6587\u5b57\u6570\u3067\u533a\u5207\u308b\u306e\u3067\u306f\u306a\u304f\u3001\u30c6\u30ad\u30b9\u30c8\u5185\u306e\u610f\u5473\u7684\u306a\u307e\u3068\u307e\u308a\u3084\u5909\u5316\u306b\u5fdc\u3058\u3066\u30c1\u30e3\u30f3\u30af\uff08\u30ce\u30fc\u30c9\uff09\u3092\u751f\u6210\u3059\u308b\u305f\u3081\u3001\u81ea\u7136\u306a\u6587\u8108\u3092\u7dad\u6301\u3057\u306a\u304c\u3089\u60c5\u5831\u3092\u51e6\u7406\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u610f\u5473\u30d9\u30fc\u30b9\u306e\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u306b\u306f\u3001SemanticSplitter NodeParser\u304c\u3042\u308a\u307e\u3059\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">\u4e8b\u524d\u6e96\u5099<\/h3>\n\n\n\n<p>\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>pip install llama_index llama-index-core llama-index-embeddings-huggingface<\/code><\/pre>\n\n\n\n<h3 class=\"wp-block-heading\">SemanticSplitter NodeParser<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">SemanticSplitterNodeParser \u306f\u3001\u30c6\u30ad\u30b9\u30c8\u3092\u610f\u5473\u7684\u306b\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u30af\u30e9\u30b9\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u57cb\u3081\u8fbc\u307f\u30e2\u30c7\u30eb\u3092\u4f7f\u3063\u3066\u30c6\u30ad\u30b9\u30c8\u5185\u306e\u610f\u5473\u306e\u5909\u308f\u308a\u76ee\u3084\u985e\u4f3c\u6027\u3092\u57fa\u306b\u3057\u3066\u3001\u5206\u5272\u4f4d\u7f6e\u3092\u52d5\u7684\u306b\u6c7a\u5b9a\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u306e\u610f\u5473\u7684\u306a\u307e\u3068\u307e\u308a\u3092\u7dad\u6301\u3057\u306a\u304c\u3089\u5206\u5272\u3067\u304d\u308b\u3053\u3068\u304c\u7279\u5fb4\u3067\u3059\u3002<\/p>\n\n\n\n<p class=\"is-style-balloon_box\">HuggingFace\u306e\u57cb\u3081\u8fbc\u307f\u30e2\u30c7\u30eb\u306b\u306fGPU\u74b0\u5883\u304c\u5fc5\u8981\u306b\u306a\u308a\u307e\u3059\u3002<\/p>\n\n\n<div class=\"swell-block-postLink\">\t\t\t<div class=\"p-blogCard -internal\" data-type=\"type1\" data-onclick=\"clickLink\">\n\t\t\t\t<div class=\"p-blogCard__inner\">\n\t\t\t\t\t<span class=\"p-blogCard__caption\">\u3042\u308f\u305b\u3066\u8aad\u307f\u305f\u3044<\/span>\n\t\t\t\t\t<div class=\"p-blogCard__thumb c-postThumb\"><figure class=\"c-postThumb__figure\"><img decoding=\"async\" src=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/06\/ubuntudocker-eyecatch-300x158.jpg\" alt=\"\" class=\"c-postThumb__img u-obf-cover\" width=\"320\" height=\"180\"><\/figure><\/div>\t\t\t\t\t<div class=\"p-blogCard__body\">\n\t\t\t\t\t\t<a class=\"p-blogCard__title\" href=\"https:\/\/highreso.jp\/edgehub\/machinelearning\/ubuntudocker.html\">Ubuntu\u306bDocker\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\uff01GPU\u3092\u30b3\u30f3\u30c6\u30ca\u304b\u3089\u4f7f\u3046\u65b9\u6cd5<\/a>\n\t\t\t\t\t\t<span class=\"p-blogCard__excerpt\">GPU\u306e\u74b0\u5883\u69cb\u7bc9\u3067\u306f\u3001\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u4e0d\u4e00\u81f4\u306b\u3088\u308b\u30a8\u30e9\u30fc\u304c\u983b\u767a\u3057\u307e\u3059\u3002 Docker\u3092\u4f7f\u3046\u3053\u3068\u3067\u3001\u3053\u306e\u3088\u3046\u306a\u4f9d\u5b58\u95a2\u4fc2\u306e\u30a8\u30e9\u30fc\u3092\u56de\u907f\u3057\u3084\u3059\u304f\u306a\u308a\u307e\u3059\u3002 \u3053\u306e\u8a18\u4e8b\u3067&#8230;<\/span>\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code># SemanticSplitterNodeParser\nfrom llama_index.core.node_parser import SemanticSplitterNodeParser\nfrom llama_index.embeddings.huggingface import HuggingFaceEmbedding\n\nembed_model = HuggingFaceEmbedding(model_name=\"intfloat\/multilingual-e5-large\")\nsplitter = SemanticSplitterNodeParser(\n    buffer_size=1, breakpoint_percentile_threshold=95, embed_model=embed_model\n)\n\nnodes = splitter.get_nodes_from_documents(documents)\nnodes&#091;0].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'\u7b2c1\u7ae0 \u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069\\n\\t \u7b2c1\u7bc0\\t \u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u5f85\u9047\u6539\u5584\u3001\u9577\u6642\u9593\u52b4\u50cd\u306e\u662f\u6b63\u7b49\\n\\t1\\t\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u5bfe\u7b56\\n\uff081\uff09\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u8ab2\u984c\\n\u8fd1\u5e74\u3001\u30d1\u30fc\u30c8\u30bf\u30a4\u30e0\u52b4\u50cd\u8005\u3001\u6709\u671f\u96c7\u7528\u52b4\u50cd\u8005\u3001\u6d3e\u9063\u52b4\u50cd\u8005\u3068\u3044\u3063\u305f\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306f\\n\u5168\u4f53\u3068\u3057\u3066\u5897\u52a0\u50be\u5411\u306b\u3042\u308a\u3001\u96c7\u7528\u8005\u306e\u7d044\u5272\u3092\u5360\u3081\u308b\u72b6\u6cc1\u306b\u3042\u308b\u3002\u3053\u308c\u306f\u3001\u9ad8\u9f62\u8005\u304c\u5897\u3048\\n\u308b\u4e2d\u3001\u9ad8\u9f62\u5c64\u3067\u306e\u7d99\u7d9a\u96c7\u7528\u306b\u3088\u308a\u975e\u6b63\u898f\u96c7\u7528\u304c\u5897\u52a0\u3057\u3066\u3044\u308b\u3053\u3068\u3084\u3001\u5973\u6027\u3092\u4e2d\u5fc3\u306b\u30d1\u30fc\u30c8\\n\u306a\u3069\u3067\u50cd\u304d\u59cb\u3081\u308b\u52b4\u50cd\u8005\u304c\u5897\u52a0\u3057\u3066\u3044\u308b\u3053\u3068\u306a\u3069\u306e\u8981\u56e0\u304c\u5927\u304d\u3044\u3002\u306a\u304a\u3001\u65b0\u578b\u30b3\u30ed\u30ca\u30a6\u30a4\\n\u30eb\u30b9\u611f\u67d3\u75c7\u306e\u611f\u67d3\u62e1\u5927\u306e\u5f71\u97ff\u3082\u3042\u308a\u30012020\uff08\u4ee4\u548c2\uff09\u5e74\u30012021\uff08\u4ee4\u548c3\uff09\u5e74\u306e\u975e\u6b63\u898f\u96c7\\n\u7528\u52b4\u50cd\u8005\u306f\u5bfe\u524d\u5e74\u6bd4\u3067\u6e1b\u5c11\u3057\u305f\u304c\u30012022\uff08\u4ee4\u548c4\uff09\u5e74\u4ee5\u964d\u306f\u5897\u52a0\u3057\u30012023\uff08\u4ee4\u548c5\uff09\u5e74\\n\u306f\u30012,124\u4e07\u4eba\u3068\u306a\u3063\u3066\u3044\u308b\u3002\\n\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306f\u3001\u96c7\u7528\u304c\u4e0d\u5b89\u5b9a\u3001\u8cc3\u91d1\u304c\u4f4e\u3044\u3001\u80fd\u529b\u958b\u767a\u6a5f\u4f1a\u304c\u4e4f\u3057\u3044\u306a\u3069\u306e\u8ab2\u984c\u304c\\n\u3042\u308a\u3001\u6b63\u898f\u96c7\u7528\u3092\u5e0c\u671b\u3057\u306a\u304c\u3089\u305d\u308c\u304c\u304b\u306a\u308f\u305a\u3001\u975e\u6b63\u898f\u96c7\u7528\u3067\u50cd\u304f\u8005\uff08\u4e0d\u672c\u610f\u975e\u6b63\u898f\u96c7\u7528\\n\u52b4\u50cd\u8005\uff09\u304c9.6\uff05\uff082023\u5e74\uff09\u5b58\u5728\u3057\u3001\u5e74\u9f62\u968e\u7d1a\u5225\u3067\u306f25\uff5e34\u6b73\u306e\u82e5\u5e74\u5c64\u306713.1\uff05\uff082023\\n\u5e74\uff09\u3068\u9ad8\u304f\u306a\u3063\u3066\u3044\u308b\u3002\u4e00\u65b9\u3001\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u4e2d\u306b\u306f\u300c\u81ea\u5206\u306e\u90fd\u5408\u306e\u3088\u3044\u6642\u9593\u306b\u50cd\u304d\\n\u305f\u3044\u304b\u3089\u300d\u7b49\u306e\u7406\u7531\u306b\u3088\u308a\u81ea\u3089\u975e\u6b63\u898f\u96c7\u7528\u3092\u9078\u3076\u65b9\u3082\u304a\u308a\u3001\u591a\u69d8\u306a\u50cd\u304d\u65b9\u304c\u9032\u3080\u4e2d\u3067\u3001\u3069\\n\u306e\u3088\u3046\u306a\u96c7\u7528\u5f62\u614b\u3092\u9078\u629e\u3057\u3066\u3082\u7d0d\u5f97\u304c\u5f97\u3089\u308c\u308b\u51e6\u9047\u3092\u53d7\u3051\u3089\u308c\u308b\u3053\u3068\u304c\u91cd\u8981\u3067\u3042\u308b\u3002\\n\uff082\uff09\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u3078\u306e\u7dcf\u5408\u7684\u306a\u5bfe\u7b56\u306e\u63a8\u9032\\n1 \u6b63\u793e\u54e1\u8ee2\u63db\u30fb\u5f85\u9047\u6539\u5584\u306e\u63a8\u9032\\n\u6b63\u793e\u54e1\u3092\u5e0c\u671b\u3059\u308b\u65b9\u306e\u6b63\u793e\u54e1\u8ee2\u63db\u3084\u975e\u6b63\u898f\u96c7\u7528\u3092\u9078\u629e\u3059\u308b\u65b9\u306e\u5f85\u9047\u6539\u5584\u3092\u63a8\u9032\u3059\u308b\u305f\\n\u3081\u3001\u30ad\u30e3\u30ea\u30a2\u30a2\u30c3\u30d7\u52a9\u6210\u91d1\u306b\u304a\u3044\u3066\u3001\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u6b63\u793e\u54e1\u8ee2\u63db\u3001\u51e6\u9047\u6539\u5584\u306e\u53d6\u7d44\u307f\\n\u3092\u56f3\u308b\u4e8b\u696d\u4e3b\u306b\u5bfe\u3057\u3066\u52a9\u6210\u3092\u884c\u3063\u3066\u3044\u308b\u3002\\n\u307e\u305f\u3001\u3069\u306e\u3088\u3046\u306a\u50cd\u304d\u65b9\u3092\u9078\u629e\u3057\u3066\u3082\u516c\u6b63\u306a\u5f85\u9047\u3092\u53d7\u3051\u3089\u308c\u308b\u3088\u3046\u306b\u3057\u3001\u4eba\u3005\u304c\u81ea\u5206\u306e\\n\u5e0c\u671b\u306b\u5408\u308f\u305b\u3066\u591a\u69d8\u306a\u50cd\u304d\u65b9\u3092\u81ea\u7531\u306b\u9078\u629e\u3067\u304d\u308b\u3088\u3046\u306b\u3059\u308b\u3053\u3068\u304c\u91cd\u8981\u3067\u3042\u308b\u3002\\n2020\uff08\u4ee4\u548c2\uff09\u5e744\u67081\u65e5\u306b\u65bd\u884c\u3055\u308c\u305f\u300c\u77ed\u6642\u9593\u52b4\u50cd\u8005\u53ca\u3073\u6709\u671f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u96c7\u7528\u7ba1\u7406\\n\u306e\u6539\u5584\u7b49\u306b\u95a2\u3059\u308b\u6cd5\u5f8b\u300d\uff08\u5e73\u62105\u5e74\u6cd5\u5f8b\u7b2c76\u53f7\u3002\u4ee5\u4e0b\u300c\u30d1\u30fc\u30c8\u30bf\u30a4\u30e0\u30fb\u6709\u671f\u96c7\u7528\u52b4\u50cd\u6cd5\u300d\\n\u3068\u3044\u3046\u3002\u540c\u6cd5\u306e\u4e2d\u5c0f\u4f01\u696d\u3078\u306e\u9069\u7528\u306f2021\uff08\u4ee4\u548c3\uff09\u5e744\u67081\u65e5\u3002\uff09\u53ca\u3073\u300c\u52b4\u50cd\u8005\u6d3e\u9063\u4e8b\u696d\\n\u306e\u9069\u6b63\u306a\u904b\u55b6\u306e\u78ba\u4fdd\u53ca\u3073\u6d3e\u9063\u52b4\u50cd\u8005\u306e\u4fdd\u8b77\u7b49\u306b\u95a2\u3059\u308b\u6cd5\u5f8b\u300d\uff08\u662d\u548c60\u5e74\u6cd5\u5f8b\u7b2c88\u53f7\uff09\u3067\\n\u306f\u3001\u96c7\u7528\u5f62\u614b\u306b\u304b\u304b\u308f\u3089\u306a\u3044\u516c\u6b63\u306a\u5f85\u9047\u306e\u78ba\u4fdd\u306b\u5411\u3051\u3001\u2460\u4e0d\u5408\u7406\u306a\u5f85\u9047\u5dee\u3092\u89e3\u6d88\u3059\u308b\u305f\u3081\\n\u306e\u898f\u5b9a\u306e\u6574\u5099\u3001\u2461\u52b4\u50cd\u8005\u306b\u5bfe\u3059\u308b\u5f85\u9047\u306b\u95a2\u3059\u308b\u8aac\u660e\u7fa9\u52d9\u306e\u5f37\u5316\u3001\u2462\u884c\u653f\u306b\u3088\u308b\u6cd5\u306e\u5c65\u884c\u78ba\\n\u4fdd\u63aa\u7f6e\u53ca\u3073\u88c1\u5224\u5916\u7d1b\u4e89\u89e3\u6c7a\u624b\u7d9a\uff08\u884c\u653fADR\uff09\u304c\u6574\u5099\u3055\u308c\u305f\u3002\\n\u30d5\u30ea\u30fc\u30bf\u30fc\u7b49\uff0a1\u306e\u6b63\u793e\u54e1\u5c31\u8077\u652f\u63f4\u306e\u305f\u3081\u3001\u300c\u308f\u304b\u3082\u306e\u30cf\u30ed\u30fc\u30ef\u30fc\u30af\u300d\uff082024\uff08\u4ee4\u548c6\uff09\u5e74\\n4\u67081\u65e5\u73fe\u572821\u304b\u6240\uff09\u7b49\u3092\u62e0\u70b9\u306b\u3001\u62c5\u5f53\u8005\u5236\u306b\u3088\u308b\u500b\u5225\u652f\u63f4\u3001\u6b63\u793e\u54e1\u5c31\u8077\u306b\u5411\u3051\u305f\u30bb\u30df\\n\uff0a1 \u304a\u304a\u3080\u306d35\u6b73\u672a\u6e80\u3067\u6b63\u793e\u54e1\u3067\u306e\u5c31\u8077\u3092\u5e0c\u671b\u3059\u308b\u6c42\u8077\u8005\uff08\u65b0\u898f\u5b66\u5352\u8005\u3001\u6b63\u898f\u96c7\u7528\u306e\u5728\u8077\u6c42\u8077\u8005\u306f\u9664\u304f\u3002\uff09\u306e\u3046\u3061\u3001\u5b89\u5b9a\u3057\u305f\u5c31\u52b4\u306e\u7d4c\u9a13\u304c\\n\u5c11\u306a\u3044\u8005\u3002\\n\u4ee4\u548c6\u5e74\u7248\\u3000\u539a\u751f\u52b4\u50cd\u767d\u66f8 187\\n\u7b2c2\u90e8\\u3000\u73fe\u4e0b\u306e\u653f\u7b56\u8ab2\u984c\u3078\u306e\u5bfe\u5fdc\\n\u7b2c\\n1\\n\u7ae0\u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069\\n\\n\u30ca\u30fc\u3084\u30b0\u30eb\u30fc\u30d7\u30ef\u30fc\u30af\u7b49\u5404\u7a2e\u652f\u63f4\u3001\u5c31\u8077\u5f8c\u306e\u5b9a\u7740\u652f\u63f4\u3092\u5b9f\u65bd\u3057\u3066\u304a\u308a\u30012023\uff08\u4ee4\u548c5\uff09\\n\u5e74\u5ea6\u306f\u7d049.8\u4e07\u4eba\u304c\u5c31\u8077\u3057\u305f\u3002\\n\u307e\u305f\u3001\u8077\u696d\u7d4c\u9a13\u3001\u6280\u80fd\u3001\u77e5\u8b58\u306e\u4e0d\u8db3\u7b49\u304b\u3089\u5b89\u5b9a\u7684\u306a\u5c31\u8077\u304c\u56f0\u96e3\u306a\u6c42\u8077\u8005\u306b\u3064\u3044\u3066\u3001\u6b63\u898f\\n\u96c7\u7528\u5316\u7b49\u306e\u65e9\u671f\u5b9f\u73fe\u3092\u56f3\u308b\u305f\u3081\u3001\u3053\u308c\u3089\u306e\u8005\u3092\u30cf\u30ed\u30fc\u30ef\u30fc\u30af\u7b49\u306e\u7d39\u4ecb\u3092\u901a\u3058\u3066\u4e00\u5b9a\u671f\u9593\u8a66\\n\u884c\u96c7\u7528\u3059\u308b\u4e8b\u696d\u4e3b\u306b\u5bfe\u3057\u3066\u52a9\u6210\u63aa\u7f6e\uff08\u30c8\u30e9\u30a4\u30a2\u30eb\u96c7\u7528\u52a9\u6210\u91d1\uff09\u3092\u8b1b\u3058\u3066\u3044\u308b\u3002\\n2\n\uff1c\u4ee5\u4e0b\u4f59\u767d\uff1e<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">embed_model = HuggingFaceEmbedding(model_name=&#8221;intfloat\/multilingual-e5-large&#8221;)<\/span><\/p>\n\n\n\n<p>Hugging Face\u306e\u57cb\u3081\u8fbc\u307f\u30e2\u30c7\u30eb\u3092\u4f7f\u7528\u3057\u3066\u3001\u30c6\u30ad\u30b9\u30c8\u306e\u610f\u5473\u7684\u306a\u30d9\u30af\u30c8\u30eb\u3092\u751f\u6210\u3057\u307e\u3059\u3002<code>intfloat\/multilingual-e5-large<\/code>\u306fHugging Face\u3067\u63d0\u4f9b\u3055\u308c\u3066\u3044\u308b\u591a\u8a00\u8a9e\u5bfe\u5fdc\u306e\u57cb\u3081\u8fbc\u307f\u30e2\u30c7\u30eb\u3067\u3001\u65e5\u672c\u8a9e\u306b\u3082\u5bfe\u5fdc\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">splitter = SemanticSplitterNodeParser(&#8230;<\/span><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>SemanticSplitterNodeParser<\/code>: \u610f\u5473\u7684\u306a\u985e\u4f3c\u6027\u306b\u57fa\u3065\u3044\u3066\u30c6\u30ad\u30b9\u30c8\u3092\u5206\u5272\u3059\u308b\u30d1\u30fc\u30b5\u3092\u521d\u671f\u5316\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>buffer_size=1<\/code>: \u30c1\u30e3\u30f3\u30af\u306e\u30d0\u30c3\u30d5\u30a1\u30b5\u30a4\u30ba\u3092\u8a2d\u5b9a\u3057\u307e\u3059\u3002\u3053\u3053\u3067\u306f\u6700\u5c0f\u30b5\u30a4\u30ba\u3068\u3057\u3066\u300c1\u300d\u304c\u8a2d\u5b9a\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>breakpoint_percentile_threshold=95<\/code>: \u30c1\u30e3\u30f3\u30af\u9593\u306e\u610f\u5473\u7684\u306a\u9055\u3044\u304c95%\u3092\u8d85\u3048\u305f\u5834\u5408\u306b\u30c6\u30ad\u30b9\u30c8\u3092\u5206\u5272\u3057\u307e\u3059\u3002\u30c6\u30ad\u30b9\u30c8\u306e\u610f\u5473\u7684\u306a\u5909\u5316\u304c\u5927\u304d\u3044\u7b87\u6240\u3067\u5206\u5272\u3055\u308c\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>embed_model=embed_model<\/code>: Hugging Face\u306e\u57cb\u3081\u8fbc\u307f\u30e2\u30c7\u30eb\u3092\u4f7f\u7528\u3057\u3066\u3001\u30c6\u30ad\u30b9\u30c8\u306e\u610f\u5473\u7684\u306a\u30d9\u30af\u30c8\u30eb\u3092\u8a08\u7b97\u3057\u307e\u3059\u3002<\/li>\n<\/ul>\n\n\n\n<p><span class=\"swl-format-2\">nodes = splitter.get_nodes_from_documents(documents)<\/span><\/p>\n\n\n\n<p><code>documents<\/code>\u3092\u610f\u5473\u7684\u306b\u5206\u5272\u3057\u3001\u305d\u308c\u305e\u308c\u306e\u30c1\u30e3\u30f3\u30af\u3092\u30ce\u30fc\u30c9\u3068\u3057\u3066\u53d6\u5f97\u3057\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">TokenText Splitter<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">TokenTextSplitter\u306f\u3001\u30c6\u30ad\u30b9\u30c8\u3092\u30c8\u30fc\u30af\u30f3\u6570\u306b\u57fa\u3065\u3044\u3066\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u6a5f\u80fd\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u591a\u304f\u306eLLM\u306f\u3001\u7279\u5b9a\u306e\u30c8\u30fc\u30af\u30f3\u6570\uff08\u5358\u8a9e\u3084\u6587\u306e\u65ad\u7247\u306a\u3069\uff09\u3092\u8d85\u3048\u308b\u3068\u51e6\u7406\u3067\u304d\u306a\u304f\u306a\u308a\u307e\u3059\u3002\u3053\u306e\u5236\u9650\u3092\u8003\u616e\u3057\u3066\u3001\u30c6\u30ad\u30b9\u30c8\u3092\u6307\u5b9a\u3055\u308c\u305f\u30c8\u30fc\u30af\u30f3\u6570\u306b\u57fa\u3065\u3044\u3066\u30c1\u30e3\u30f3\u30af\u306b\u5206\u5272\u3059\u308b\u5f79\u5272\u3092\u679c\u305f\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code># TokenTextSplitting\nfrom llama_index.core.node_parser import TokenTextSplitter\n\nsplitter = TokenTextSplitter(\n    chunk_size=254,\n    chunk_overlap=20,\n    separator=\" \",\n)\nnodes = splitter.get_nodes_from_documents(documents)\nnodes&#091;0].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'\u7b2c1\u7ae0 \u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069\\n\\t \u7b2c1\u7bc0\\t \u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u5f85\u9047\u6539\u5584\u3001\u9577\u6642\u9593\u52b4\u50cd\u306e\u662f\u6b63\u7b49\\n\\t1\\t\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u5bfe\u7b56\\n\uff081\uff09\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u8ab2\u984c\\n\u8fd1\u5e74\u3001\u30d1\u30fc\u30c8\u30bf\u30a4\u30e0\u52b4\u50cd\u8005\u3001\u6709\u671f\u96c7\u7528\u52b4\u50cd\u8005\u3001\u6d3e\u9063\u52b4\u50cd\u8005\u3068\u3044\u3063\u305f\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306f'<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">splitter = TokenTextSplitter(&#8230;<\/span><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>chunk_size=254<\/code>: 1\u3064\u306e\u30c1\u30e3\u30f3\u30af\u306b\u542b\u307e\u308c\u308b\u30c8\u30fc\u30af\u30f3\u306e\u6700\u5927\u6570\u3092254\u30c8\u30fc\u30af\u30f3\u306b\u8a2d\u5b9a\u3057\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u30c6\u30ad\u30b9\u30c8\u304c254\u30c8\u30fc\u30af\u30f3\u305a\u3064\u5206\u5272\u3055\u308c\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>chunk_overlap=20<\/code>: \u96a3\u63a5\u3059\u308b\u30c1\u30e3\u30f3\u30af\u9593\u306720\u30c8\u30fc\u30af\u30f3\u306e\u91cd\u8907\u304c\u8a2d\u5b9a\u3055\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u6587\u8108\u306e\u9023\u7d9a\u6027\u3092\u4fdd\u3064\u3053\u3068\u304c\u3067\u304d\u3001\u30e2\u30c7\u30eb\u304c\u3088\u308a\u610f\u5473\u3092\u7406\u89e3\u3057\u3084\u3059\u304f\u306a\u308a\u307e\u3059\u3002<\/li>\n\n\n\n<li><code>separator=\" \"<\/code>: \u30c8\u30fc\u30af\u30f3\u3092\u5206\u5272\u3059\u308b\u969b\u306e\u533a\u5207\u308a\u6587\u5b57\u3068\u3057\u3066\u3001\u30b9\u30da\u30fc\u30b9\uff08<code>\" \"<\/code>\uff09\u3092\u4f7f\u7528\u3057\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306f\u3001\u4e00\u822c\u7684\u306b\u5358\u8a9e\u306e\u533a\u5207\u308a\u3092\u610f\u5473\u3057\u307e\u3059\u3002<\/li>\n<\/ul>\n\n\n\n<p><span class=\"swl-format-2\">nodes = splitter.get_nodes_from_documents(documents)<\/span><\/p>\n\n\n\n<p>\u6307\u5b9a\u3057\u305f<code>documents<\/code>\u3092\u30c8\u30fc\u30af\u30f3\u5358\u4f4d\u3067\u5206\u5272\u3057\u3001\u305d\u308c\u305e\u308c\u3092\u30ce\u30fc\u30c9\u306b\u5909\u63db\u3057\u307e\u3059\u3002\u30ce\u30fc\u30c9\u306f\u30c6\u30ad\u30b9\u30c8\u306e\u4e00\u90e8\u5206\u3092\u4fdd\u6301\u3057\u3001\u5f8c\u3067\u30e2\u30c7\u30eb\u306b\u5165\u529b\u3059\u308b\u305f\u3081\u306e\u30c1\u30e3\u30f3\u30af\u3068\u306a\u308a\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">\u30ea\u30ec\u30fc\u30b7\u30e7\u30f3\u30d9\u30fc\u30b9\u306eNode Parser<\/h2>\n\n\n\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" width=\"600\" height=\"300\" src=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/09\/18_midashi_sf0923.jpg\" alt=\"\u898b\u51fa\u3057\u753b\u50cf\" class=\"wp-image-11929\" srcset=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/09\/18_midashi_sf0923.jpg 600w, https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/09\/18_midashi_sf0923-300x150.jpg 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/><\/figure>\n\n\n\n<p><span class=\"swl-format-2\">\u30ea\u30ec\u30fc\u30b7\u30e7\u30f3\u30d9\u30fc\u30b9\u306e\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u306f\u3001\u30c6\u30ad\u30b9\u30c8\u3084\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3092\u968e\u5c64\u7684\u306b\u5206\u5272\u3057\u3001\u30ce\u30fc\u30c9\u9593\u306b\u89aa\u5b50\u95a2\u4fc2\u306a\u3069\u306e\u30ea\u30ec\u30fc\u30b7\u30e7\u30f3\uff08\u95a2\u4fc2\u6027\uff09\u3092\u69cb\u7bc9\u3059\u308b\u305f\u3081\u306e\u624b\u6cd5\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u30c6\u30ad\u30b9\u30c8\u306e\u5185\u5bb9\u3092\u3088\u308a\u5927\u304d\u306a\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u304b\u3089\u7d30\u90e8\u307e\u3067\u7406\u89e3\u3059\u308b\u305f\u3081\u306b\u3001\u30ce\u30fc\u30c9\u3092\u968e\u5c64\u7684\u306b\u7ba1\u7406\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u4ee3\u8868\u7684\u306a\u30ea\u30ec\u30fc\u30b7\u30e7\u30f3\u30d9\u30fc\u30b9\u306e\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u306b\u306fHierarchical NodeParse\u304c\u3042\u308a\u307e\u3059\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">\u4e8b\u524d\u6e96\u5099<\/h3>\n\n\n\n<p>\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code>pip install llama_index llama-index-core llama-index-embeddings-huggingface<\/code><\/pre>\n\n\n\n<h3 class=\"wp-block-heading\">Hierarchical NodeParser<\/h3>\n\n\n\n<p><span class=\"swl-format-2\">HierarchicalNodeParser\u306f\u3001\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3092\u968e\u5c64\u7684\u306b\u5206\u5272\u3059\u308b\u305f\u3081\u306e\u30ce\u30fc\u30c9\u30d1\u30fc\u30b5\u30fc\u3067\u3059\u3002<\/span><\/p>\n\n\n\n<p>\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3092\u5927\u304d\u306a\u30c1\u30e3\u30f3\u30af\u304b\u3089\u5c0f\u3055\u306a\u30c1\u30e3\u30f3\u30af\u3078\u3068\u5206\u5272\u3057\u3001\u305d\u308c\u305e\u308c\u306e\u30c1\u30e3\u30f3\u30af\u9593\u306b\u89aa\u5b50\u95a2\u4fc2\u3092\u69cb\u7bc9\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u5927\u304d\u306a\u30c1\u30e3\u30f3\u30af\u304c\u89aa\u30ce\u30fc\u30c9\u3001\u5c0f\u3055\u306a\u30c1\u30e3\u30f3\u30af\u304c\u5b50\u30ce\u30fc\u30c9\u3068\u3057\u3066\u968e\u5c64\u69cb\u9020\u3092\u5f62\u6210\u3059\u308b\u305f\u3081\u3001\u5168\u4f53\u306e\u6587\u8108\u3092\u4fdd\u6301\u3057\u3064\u3064\u3001\u8a73\u7d30\u306a\u90e8\u5206\u306e\u89e3\u6790\u3082\u53ef\u80fd\u3067\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code has-black-background-color has-background\"><code># HierarchicalNodeParser\nfrom llama_index.core.node_parser import HierarchicalNodeParser\n\nnode_parser = HierarchicalNodeParser.from_defaults(chunk_sizes=&#091;512, 254, 128])\n\nnodes = node_parser.get_nodes_from_documents(documents)\nnodes&#091;0].text<\/code><\/pre>\n\n\n\n<pre class=\"wp-block-code has-swl-gray-background-color has-background\"><code>'\u7b2c1\u7ae0 \u50cd\u304d\u65b9\u6539\u9769\u306e\u63a8\u9032\u306a\u3069\u3092\u901a\u3058\u305f\u52b4\u50cd\u74b0\u5883\u306e\u6574\u5099\u306a\u3069\\n\\t \u7b2c1\u7bc0\\t \u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306e\u5f85\u9047\u6539\u5584\u3001\u9577\u6642\u9593\u52b4\u50cd\u306e\u662f\u6b63\u7b49\\n\\t1\\t\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u5bfe\u7b56\\n\uff081\uff09\u975e\u6b63\u898f\u96c7\u7528\u306e\u73fe\u72b6\u3068\u8ab2\u984c\\n\u8fd1\u5e74\u3001\u30d1\u30fc\u30c8\u30bf\u30a4\u30e0\u52b4\u50cd\u8005\u3001\u6709\u671f\u96c7\u7528\u52b4\u50cd\u8005\u3001\u6d3e\u9063\u52b4\u50cd\u8005\u3068\u3044\u3063\u305f\u975e\u6b63\u898f\u96c7\u7528\u52b4\u50cd\u8005\u306f\\n\u5168\u4f53\u3068\u3057\u3066\u5897\u52a0\u50be\u5411\u306b\u3042\u308a\u3001\u96c7\u7528\u8005\u306e\u7d044\u5272\u3092\u5360\u3081\u308b\u72b6\u6cc1\u306b\u3042\u308b\u3002\u3053\u308c\u306f\u3001\u9ad8\u9f62\u8005\u304c\u5897\u3048\\n\u308b\u4e2d\u3001\u9ad8\u9f62\u5c64\u3067\u306e\u7d99\u7d9a\u96c7\u7528\u306b\u3088\u308a\u975e\u6b63\u898f\u96c7\u7528\u304c\u5897\u52a0\u3057\u3066\u3044\u308b\u3053\u3068\u3084\u3001\u5973\u6027\u3092\u4e2d\u5fc3\u306b\u30d1\u30fc\u30c8\\n\u306a\u3069\u3067\u50cd\u304d\u59cb\u3081\u308b\u52b4\u50cd\u8005\u304c\u5897\u52a0\u3057\u3066\u3044\u308b\u3053\u3068\u306a\u3069\u306e\u8981\u56e0\u304c\u5927\u304d\u3044\u3002\u306a\u304a\u3001\u65b0\u578b\u30b3\u30ed\u30ca\u30a6\u30a4\\n\u30eb\u30b9\u611f\u67d3\u75c7\u306e\u611f\u67d3\u62e1\u5927\u306e\u5f71\u97ff\u3082\u3042\u308a\u30012020\uff08\u4ee4\u548c2\uff09\u5e74\u30012021\uff08\u4ee4\u548c3\uff09\u5e74\u306e\u975e\u6b63\u898f\u96c7\\n\u7528\u52b4\u50cd\u8005\u306f\u5bfe\u524d\u5e74\u6bd4\u3067\u6e1b\u5c11\u3057\u305f\u304c\u30012022\uff08\u4ee4\u548c4\uff09\u5e74\u4ee5\u964d\u306f\u5897\u52a0\u3057\u30012023\uff08\u4ee4\u548c5\uff09\u5e74\\n\u306f\u30012,124\u4e07\u4eba\u3068\u306a\u3063\u3066\u3044\u308b\u3002'<\/code><\/pre>\n\n\n\n<div class=\"swell-block-capbox cap_box is-style-onborder_ttl2\" data-colset=\"col3\"><div class=\"cap_box_ttl\"><span>\u30b3\u30fc\u30c9\u306e\u8aac\u660e<\/span><\/div><div class=\"cap_box_content\">\n<p><span class=\"swl-format-2\">node_parser = HierarchicalNodeParser.from_defaults(chunk_sizes=[512, 254, 128])<\/span><\/p>\n\n\n\n<p><code>chunk_sizes<\/code> \u30d1\u30e9\u30e1\u30fc\u30bf\u306b <code>[512, 254, 128]<\/code> \u304c\u6307\u5b9a\u3055\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u306f\u3001\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u304c3\u3064\u306e\u7570\u306a\u308b\u30b5\u30a4\u30ba\u306e\u30c1\u30e3\u30f3\u30af\u306b\u5206\u5272\u3055\u308c\u308b\u3053\u3068\u3092\u610f\u5473\u3057\u307e\u3059\u3002512\u304c\u89aa\u30ce\u30fc\u30c9, 254\u304c\u5b50\u30ce\u30fc\u30c9, 128\u5b6b\u30ce\u30fc\u30c9\u306e\u3088\u3046\u306b\u968e\u5c64\u304c\u4f5c\u3089\u308c\u3001\u89aa\u5b50\u95a2\u4fc2\u304c\u69cb\u7bc9\u3055\u308c\u307e\u3059\u3002<\/p>\n\n\n\n<p><span class=\"swl-format-2\">nodes = node_parser.get_nodes_from_documents(documents)<\/span><\/p>\n\n\n\n<p><code>documents<\/code> \u306e\u5185\u5bb9\u3092\u6307\u5b9a\u3055\u308c\u305f\u30c1\u30e3\u30f3\u30af\u30b5\u30a4\u30ba\u306b\u57fa\u3065\u3044\u3066\u5206\u5272\u3057\u307e\u3059\u3002\u3053\u306e\u30e1\u30bd\u30c3\u30c9\u306f\u3001\u968e\u5c64\u7684\u306b\u5206\u5272\u3055\u308c\u305f\u30ce\u30fc\u30c9\u306e\u30ea\u30b9\u30c8\u3092\u8fd4\u3057\u307e\u3059\u3002<\/p>\n<\/div><\/div>\n\n\n\n\n\n\n\n<h2 class=\"wp-block-heading\" id=\"index_id8\">\u751f\u6210AI\u30fbLLM\u306e\u30b3\u30b9\u30c8\u3067\u304a\u56f0\u308a\u306a\u3089<\/h2>\n\n\n\n<p>GPU\u306e\u30b9\u30da\u30c3\u30af\u4e0d\u8db3\u3067\u751f\u6210AI\u306e\u958b\u767a\u304c\u601d\u3046\u3088\u3046\u306b\u9032\u307e\u306a\u3044\u3053\u3068\u306f\u3042\u308a\u307e\u305b\u3093\u304b\uff1f<\/p>\n\n\n\n<p>\u305d\u3093\u306a\u3068\u304d\u306b\u306f\u3001<strong><span class=\"swl-format-2\">\u9ad8\u6027\u80fd\u306aGPU\u3092\u30ea\u30fc\u30ba\u30ca\u30d6\u30eb\u306a\u4fa1\u683c\u3067\u4f7f\u3048\u308bGPU\u30af\u30e9\u30a6\u30c9\u30b5\u30fc\u30d3\u30b9\u304c\u304a\u3059\u3059\u3081\u3067\u3059\uff01<\/span><\/strong><\/p>\n\n\n\n<div class=\"p-blogParts post_content\" data-partsID=\"4543\">\n<div class=\"swell-block-capbox cap_box\" data-colset=\"col2\"><div class=\"cap_box_ttl\"><span>GPUSOROBAN<\/span><\/div><div class=\"cap_box_content\">\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" width=\"1462\" height=\"508\" src=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/06\/h100_banner.jpg\" alt=\"GPUSOROBAN\" class=\"wp-image-10134\" srcset=\"https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/06\/h100_banner.jpg 1462w, https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/06\/h100_banner-300x104.jpg 300w, https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/06\/h100_banner-1024x356.jpg 1024w, https:\/\/highreso.jp\/edgehub\/wp-content\/uploads\/2024\/06\/h100_banner-768x267.jpg 768w\" sizes=\"(max-width: 1462px) 100vw, 1462px\" \/><\/figure>\n\n\n\n<p>GPUSOROBAN\u306f\u3001<span class=\"swl-marker mark_yellow\"><span class=\"swl-format-1\">\u9ad8\u6027\u80fdGPU\u300cNVIDIA H200\u300d\u3092\u696d\u754c\u6700\u5b89\u7d1a\u306e\u6599\u91d1\u3067\u4f7f\u7528\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/span><\/span><\/p>\n\n\n\n<p>NVIDIA H200\u306f\u3001\u751f\u6210AI\u30fbLLM\u306e\u8a08\u7b97\u306b\u304b\u304b\u308b\u6642\u9593\u3092\u5927\u5e45\u306b\u77ed\u7e2e\u3059\u308b\u3053\u3068\u304c\u53ef\u80fd\u3067\u3059\u3002<\/p>\n\n\n\n<p>\u30af\u30e9\u30a6\u30c9\u3067\u4f7f\u3048\u308b\u305f\u3081\u3001\u5927\u898f\u6a21\u306a\u8a2d\u5099\u6295\u8cc7\u306e\u5fc5\u8981\u304c\u306a\u304f\u3001\u7169\u96d1\u306a\u30b5\u30fc\u30d0\u30fc\u7ba1\u7406\u304b\u3089\u3082\u89e3\u653e\u3055\u308c\u307e\u3059\u3002<\/p>\n\n\n\n<div class=\"swell-block-columns\"><div class=\"swell-block-columns__inner\">\n<div class=\"swell-block-column swl-has-mb--s\">\n<div class=\"swell-block-button blue_ -size-l is-style-btn_normal\"><a href=\"https:\/\/soroban.highreso.jp\/?utm_medium=edgehub&amp;utm_source=edgehub&amp;utm_campaign=highend_boxb&amp;adid=edgehub\" class=\"swell-block-button__link\"><span>\u516c\u5f0f\u30b5\u30a4\u30c8\u3092\u307f\u308b<\/span><\/a><\/div>\n<\/div>\n\n\n\n<div class=\"swell-block-column swl-has-mb--s\">\n<div class=\"swell-block-button red_ -size-l is-style-btn_normal\"><a href=\"https:\/\/soroban.highreso.jp\/download\/compute-catalog\/?utm_medium=edgehub&amp;utm_source=edgehub&amp;utm_campaign=highend_boxb&amp;adid=edgehub\" class=\"swell-block-button__link\"><span>\u8cc7\u6599\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9<\/span><\/a><\/div>\n<\/div>\n<\/div><\/div>\n<\/div><\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>\u3053\u306e\u8a18\u4e8b\u3067\u306f\u3001LlamaIndex\u3067\u30c1\u30e3\u30f3\u30af\u5206\u5272\u3092\u3059\u308bNode Parser\u306e\u4f7f\u3044\u65b9\u306b\u3064\u3044\u3066\u89e3\u8aac\u3057\u3066\u3044\u307e\u3059\u3002 \u69cb\u6587\u30d9\u30fc\u30b9\u3001\u9577\u3055\u30d9\u30fc\u30b9\u3001\u610f\u5473\u30d9\u30fc\u30b9\u3001\u30ea\u30ec\u30fc\u30b7\u30e7\u30f3\u30d9\u30fc\u30b9\u306e\u305d\u308c\u305e\u308c\u306b\u5bfe\u5fdc\u3059\u308b\u30d1\u30fc\u30b5\u30fc\u3092\u7d39\u4ecb\u3057\u3001\u30c6\u30ad\u30b9\u30c8\u3084\u30c9\u30ad\u30e5\u30e1 [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":12376,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"swell_btn_cv_data":"","footnotes":""},"categories":[20],"tags":[],"class_list":["post-12273","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-machinelearning"],"_links":{"self":[{"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/posts\/12273","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/comments?post=12273"}],"version-history":[{"count":3,"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/posts\/12273\/revisions"}],"predecessor-version":[{"id":13323,"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/posts\/12273\/revisions\/13323"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/media\/12376"}],"wp:attachment":[{"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/media?parent=12273"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/categories?post=12273"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/highreso.jp\/edgehub\/wp-json\/wp\/v2\/tags?post=12273"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}