mirror of
https://github.com/maglore9900/max_headroom.git
synced 2025-06-06 19:45:31 +00:00
68 lines
2.3 KiB
Python
68 lines
2.3 KiB
Python
from langchain_core.prompts import ChatPromptTemplate, HumanMessagePromptTemplate
|
|
from langchain_core.messages import SystemMessage
|
|
from langchain_core.output_parsers import StrOutputParser
|
|
import adapter
|
|
|
|
ad = adapter.Adapter()
|
|
|
|
class Journal:
|
|
def __init__(self):
|
|
self.sys_prompt = """
|
|
You are tasked with creating a detailed and neutral journal entry. The entry should capture key points, observations, any relevant metrics, and recommended actions or next steps. The language should remain neutral, avoiding any self-referential language.
|
|
|
|
Example 1:
|
|
Date: [DD-MM-YYYY]
|
|
|
|
**Key Points:**
|
|
- [Summary of the main points or events, capturing essential details]
|
|
|
|
**Observations:**
|
|
- [Notes on observations, trends, or insights drawn from the data]
|
|
|
|
**Metrics:**
|
|
- [List of any relevant metrics, figures, or statistics related to the data]
|
|
|
|
**Recommended Actions:**
|
|
- [Suggested actions, strategies, or next steps based on the data analysis]
|
|
|
|
|
|
Example 2:
|
|
Date: [DD-MM-YYYY]
|
|
|
|
**Summary:**
|
|
- [Overview of the data or event, focusing on significant highlights]
|
|
|
|
**Analysis:**
|
|
- [Detailed analysis, including patterns, anomalies, or key insights]
|
|
|
|
**Figures:**
|
|
- [Relevant numerical data, metrics, or charts]
|
|
|
|
**Next Steps:**
|
|
- [Proposed actions, decisions, or follow-up activities derived from the analysis]
|
|
|
|
Ensure that the tone remains neutral, and avoid using 'I' or 'my' in the notes.
|
|
|
|
The format MUST be in markdown
|
|
"""
|
|
|
|
self.chat_template = ChatPromptTemplate.from_messages(
|
|
[
|
|
SystemMessage(
|
|
content=(
|
|
self.sys_prompt
|
|
)
|
|
),
|
|
HumanMessagePromptTemplate.from_template("{text}"),
|
|
]
|
|
)
|
|
|
|
self.journal_llm = self.chat_template | ad.llm_chat | StrOutputParser()
|
|
|
|
def journal(self, text):
|
|
message = self.chat_template.format_messages(text=text)
|
|
response = self.journal_llm.invoke(message)
|
|
return response
|
|
|
|
|