From 398ceb887bd3ba99aff009b10babfbabdb0cbdc0 Mon Sep 17 00:00:00 2001 From: HuangHai <10402852@qq.com> Date: Mon, 30 Jun 2025 08:37:13 +0800 Subject: [PATCH] 'commit' --- dsRag/Start.py | 4 +- dsRag/Util/EsSearchUtil.py | 54 +++++------------- .../__pycache__/EsSearchUtil.cpython-310.pyc | Bin 6631 -> 5791 bytes 3 files changed, 17 insertions(+), 41 deletions(-) diff --git a/dsRag/Start.py b/dsRag/Start.py index 587a86f4..ca0b5094 100644 --- a/dsRag/Start.py +++ b/dsRag/Start.py @@ -114,11 +114,11 @@ async def rag(request: fastapi.Request): query = data.get('query', '') query_tags = data.get('tags', []) # 调用es进行混合搜索 - search_results = queryByEs(query, query_tags, logger) + search_results = EsSearchUtil.queryByEs(query, query_tags, logger) # 构建提示词 context = "\n".join([ f"结果{i + 1}: {res['tags']['full_content']}" - for i, res in enumerate(search_results['vector_results'] + search_results['text_results']) + for i, res in enumerate(search_results['text_results']) ]) # 添加图片识别提示 prompt = f""" diff --git a/dsRag/Util/EsSearchUtil.py b/dsRag/Util/EsSearchUtil.py index 21fe14e5..09d1c569 100644 --- a/dsRag/Util/EsSearchUtil.py +++ b/dsRag/Util/EsSearchUtil.py @@ -201,47 +201,23 @@ class EsSearchUtil: ) logger.info(f"2. 文本搜索结果数量: {len(text_results['hits']['hits'])}") - # 合并结果 - logger.info("\n=== 最终搜索结果 ===") - logger.info(f"向量搜索结果: {len(vector_results['hits']['hits'])}条") - for i, hit in enumerate(vector_results['hits']['hits'], 1): - logger.info(f" {i}. 文档ID: {hit['_id']}, 相似度分数: {hit['_score']:.2f}") - logger.info(f" 内容: {hit['_source']['user_input']}") - - logger.info("文本精确搜索结果:") - for i, hit in enumerate(text_results['hits']['hits']): - logger.info(f" {i + 1}. 文档ID: {hit['_id']}, 匹配分数: {hit['_score']:.2f}") - logger.info(f" 内容: {hit['_source']['user_input']}") - - # 去重处理：去除vector_results和text_results中重复的user_input - vector_sources = [hit['_source'] for hit in vector_results['hits']['hits']] - text_sources = [hit['_source'] for hit in text_results['hits']['hits']] - - # 构建去重后的结果 - unique_text_sources = [] - text_user_inputs = set() - - # 先处理text_results，保留所有 - for source in text_sources: - text_user_inputs.add(source['user_input']) - unique_text_sources.append(source) - - # 处理vector_results，只保留不在text_results中的 - unique_vector_sources = [] - for source in vector_sources: - if source['user_input'] not in text_user_inputs: - unique_vector_sources.append(source) - - # 计算优化掉的记录数量和节约的tokens - removed_count = len(vector_sources) - len(unique_vector_sources) - saved_tokens = sum(len(source['user_input']) for source in vector_sources - if source['user_input'] in text_user_inputs) - - logger.info(f"优化掉 {removed_count} 条重复记录，节约约 {saved_tokens} tokens") + # 合并vector和text结果 + all_sources = [hit['_source'] for hit in vector_results['hits']['hits']] + \ + [hit['_source'] for hit in text_results['hits']['hits']] + + # 去重处理 + unique_sources = [] + seen_user_inputs = set() + + for source in all_sources: + if source['user_input'] not in seen_user_inputs: + seen_user_inputs.add(source['user_input']) + unique_sources.append(source) + + logger.info(f"合并后去重结果数量: {len(unique_sources)}条") search_results = { - "vector_results": unique_vector_sources, - "text_results": unique_text_sources + "text_results": unique_sources } return search_results finally: diff --git a/dsRag/Util/__pycache__/EsSearchUtil.cpython-310.pyc b/dsRag/Util/__pycache__/EsSearchUtil.cpython-310.pyc index 5b37225ac2c0070aeda27e896b2c3ed68961e042..b8660cd06a341531f537f6061a2ff2d6bcf4a41c 100644 GIT binary patch delta 588 zcmY+B&ubGw6vyZ7EZgjEvKU>H4YV{-my&}AQ4xw2q^UQxsGtU9T{AHP?T8| zB3ArS3xgnNHMx3H0)q7B!JGX9g4d)3JbKsJR73~fcjo(f@4=gy-Q(|$8qG{bLt>fV z-PIq(j|S7B%%AED%A$aNrN5)U2Uy~FGi!E65LHBy34Q`XAqt@bF=lv8>f#=X<$L44 zW}JchWR?^kBBl_lfRd=74v^A>P5sOAC{C@QSPfGh6l*M1K{w1;XF8cFAg0|0f55b^ zy61ZUh^bwT>B*`6l}TNh910btcHlK4hqrLT$q}a3ZHUt>eHBdx)0vzkF8}B5pt9Yc z&YqHZRyoK|o3rX*$_bj0U!(k{X~Q@E$Xq^fK6!%+iIfsq7Q)-6Ilq&*OG2UV|2&n7VRvn)PQA5DVHA?(!-cp~4d`i-=#Bb$SrcG>u{7aSPw=@Ur6K{?-la=ne_#(^NWR7zX8mqpt%45 delta 1412 zcma)6U2NM_6u#H?+{8(oG)+SbjJe&KE@Lg-x^_}oCc(O)y=;R5lOjeI)n3V9PIGOS zHFA`0>DCTf2w0mYHDL{)N<*j`ln1sIUU@^}ae3ioPF5tuD^HAbozkB-toZnx?|%1u z_nc$rPEWmGO#6HSf-Cu3^1bts??lss?^+*wV*D_G1tfQrS-Psb zHgVZsV207T-S03tSI+J1>oi@!-EF$%J?#j*F7P6drF(%Z^BdXs0vZK$C*_nG!1E~c zi<4mAx=7QeI7oDDk$BN`$^K_(@=clj|5FBK{+4^>-f6U1#4wbX6Y}y)4!j}jz@9Ob z^Xl#bx`9j&9K)aU&7wuY^wE<$)p!>@rjL3>oOv18qA4_Z4bCi@z(q{Xnc^1SY3Qv` zz+lA!ZbLc0?wvV{E?~J&J}~W~2?}&f0jD1aO+QrHZw9~~D(`SaMbj$}l$czz*?bl` z!}OUPT^ZRHeTbriD}VS8GkN1V`YOa?F{xIXsz02q6-)IWzN|mAYG0HjfQ=BW+r`h; zWkQFZ}p;b=7{bWY1o& z-Mlv>#rcFbnkGt??HxII!4QE3IpDIdU$d9*!|tYpmNv+!Vzg5c$Xfcg{{7GO%JQ}| zWEoHnI2!ic{k3cJ4FPLcaBv4X4Y>7gunRx3rh{kt&Ok_Br>29NH#R&c?4)4?OqDus zg0Y(U)_gxW+FmeMkV!|3O)R5!1fxTQ(NQBR;I|P*8Aiq)lX5RjK{>kz^`=^_Kt`Q9e+TnvxhztEw`p zr_yRPXcb!y