Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -46,7 +46,9 @@ def chatresponse(message, history):
|
|
| 46 |
# vectorstore = data_vectorstore + history_vectorstore
|
| 47 |
vectorstore = data_vectorstore
|
| 48 |
retriever = vectorstore.as_retriever()
|
| 49 |
-
|
|
|
|
|
|
|
| 50 |
# from langchain.prompts import PromptTemplate
|
| 51 |
|
| 52 |
from langchain_core.prompts import ChatPromptTemplate
|
|
@@ -57,13 +59,15 @@ def chatresponse(message, history):
|
|
| 57 |
If somebody asks "Who are you?" or a similar phrase, state "I am Rishi's assistant built using a Large Language Model!"
|
| 58 |
If the answer is not found in the context, kindly state "I don't know. Please ask Rishi on Discord. Discord Invite Link: https://discord.gg/6ezpZGeCcM. Or email at rishi@aiotsmartlabs.com" Don't try to make up an answer.
|
| 59 |
|
|
|
|
|
|
|
| 60 |
CONTEXT: {context}
|
| 61 |
|
| 62 |
QUESTION: {question}""")
|
| 63 |
|
| 64 |
from langchain_core.runnables import RunnablePassthrough
|
| 65 |
rag_chain = (
|
| 66 |
-
{"context":
|
| 67 |
| prompt
|
| 68 |
| llm
|
| 69 |
| StrOutputParser()
|
|
|
|
| 46 |
# vectorstore = data_vectorstore + history_vectorstore
|
| 47 |
vectorstore = data_vectorstore
|
| 48 |
retriever = vectorstore.as_retriever()
|
| 49 |
+
|
| 50 |
+
history_str = "\n".join([f"Human: {h[0]}\nAI: {h[1]}" for h in history])
|
| 51 |
+
|
| 52 |
# from langchain.prompts import PromptTemplate
|
| 53 |
|
| 54 |
from langchain_core.prompts import ChatPromptTemplate
|
|
|
|
| 59 |
If somebody asks "Who are you?" or a similar phrase, state "I am Rishi's assistant built using a Large Language Model!"
|
| 60 |
If the answer is not found in the context, kindly state "I don't know. Please ask Rishi on Discord. Discord Invite Link: https://discord.gg/6ezpZGeCcM. Or email at rishi@aiotsmartlabs.com" Don't try to make up an answer.
|
| 61 |
|
| 62 |
+
HISTORY: {history}
|
| 63 |
+
|
| 64 |
CONTEXT: {context}
|
| 65 |
|
| 66 |
QUESTION: {question}""")
|
| 67 |
|
| 68 |
from langchain_core.runnables import RunnablePassthrough
|
| 69 |
rag_chain = (
|
| 70 |
+
{"history": history_str, "context": retriever, "question": RunnablePassthrough()}
|
| 71 |
| prompt
|
| 72 |
| llm
|
| 73 |
| StrOutputParser()
|