vamseeachanta
vamseeachanta/workspace-hub
相关职业SOC
基于 SOC 职业分类
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/vamseeachanta/workspace-hub --skill rag-system-builder-advanced-reranking命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | rag-system-builder-advanced-reranking |
| description | Sub-skill of rag-system-builder: Advanced: Reranking. |
| version | 1.2.0 |
| category | data |
| type | reference |
| scripts_exempt | true |
Add a reranking step for improved precision:
from sentence_transformers import CrossEncoder
class Reranker:
def __init__(self, model_name='cross-encoder/ms-marco-MiniLM-L-6-v2'):
self.model = CrossEncoder(model_name)
def rerank(self, query, candidates, top_k=5):
"""Rerank candidates using cross-encoder."""
pairs = [(query, c['text']) for c in candidates]
scores = self.model.predict(pairs)
for i, score in enumerate(scores):
candidates[i]['rerank_score'] = float(score)
reranked = sorted(candidates, key=lambda x: x['rerank_score'], reverse=True)
return reranked[:top_k]
# Usage in RAG pipeline
def query_with_rerank(self, question, initial_k=20, final_k=5):
# First pass: retrieve more candidates
candidates = semantic_search(self.db_path, question, self.model, top_k=initial_k)
# Second pass: rerank for precision
reranked = self.reranker.rerank(question, candidates, top_k=final_k)
return reranked