galen1980guo
commented
1 month ago Has anyone encountered the same issue, or does anyone know how to solve this problem? Thanks.
Closed galen1980guo closed 1 month ago
galen1980guo
commented
1 month ago Has anyone encountered the same issue, or does anyone know how to solve this problem? Thanks.
natoverse
commented
1 month ago This is likely resolved with https://github.com/microsoft/graphrag/pull/801. We will push a release shortly, or you can update on main if you are running the codebase directly. Otherwise, there is lots of commentary linked to #657 about the JSON formats of non-OpenAI models
xxll88
commented
1 month ago This is likely resolved with #801. We will push a release shortly, or you can update on main if you are running the codebase directly. Otherwise, there is lots of commentary linked to #657 about the JSON formats of non-OpenAI models
although json_clean_up repair json faulty responses，it increase index time and global search time 70-80% v0.2.0 index time 35min , v0.2.1 62min v0.2.0 global search time 45s , v0.2.1 75s
Vaccy-Zhu
commented
1 month ago import re
def extract_json(input: str) -> str:
"""
Extract JSON content from a string, where JSON is embedded between ```json and ``` markers.
"""
text = input
# Define a regular expression pattern to match JSON blocks
pattern = r"```(.*?)```"
# Find all non-overlapping matches of the pattern in the string
matches = re.findall(pattern, text, re.DOTALL)
if not matches:
return input
# Return the matched JSON string, stripped of any leading or trailing whitespace
try:
return matches[0].strip()
except Exception:
raise ValueError(f"Failed to parse: {input}")======= 用以上函数处理一下你的llm输出。 Use above function to process your LLM output.
natoverse
commented
1 month ago We have resolved several issues related to text encoding and JSON parsing that are rolled up into version 0.2.2. Please try again with that version and re-open if this is still an issue.
zidanereal5
commented
1 week ago extract_json
请问这个是修改哪个文件
Is there an existing issue for this?
Describe the issue
When I deployed the LLM using Ollama and the embeddings using Xinference, the process of building the index failed at the step create_final_community_reports. According to the detailed logs, it seems that the issue occurred due to a failure in parsing JSON.
Steps to reproduce
No response
GraphRAG Config Used
Logs and screenshots