Spaces:
Runtime error
Runtime error
Upload 3 files
Browse files- main.py +74 -0
- requirements.txt +10 -0
- ui.py +49 -0
main.py
ADDED
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from langchain_groq import ChatGroq
|
2 |
+
from langchain.prompts import PromptTemplate
|
3 |
+
from langchain.chains import LLMChain
|
4 |
+
from langchain.prompts import PromptTemplate
|
5 |
+
from langchain.callbacks.manager import CallbackManager
|
6 |
+
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
|
7 |
+
from langchain.agents import Tool
|
8 |
+
from langchain_experimental.utilities import PythonREPL # type: ignore
|
9 |
+
from langchain_community.chat_models import ChatOllama
|
10 |
+
import autopep8 # type: ignore
|
11 |
+
import pandas as pd
|
12 |
+
import os
|
13 |
+
|
14 |
+
from dotenv import load_dotenv
|
15 |
+
load_dotenv()
|
16 |
+
|
17 |
+
groq_api_key = os.getenv("GROQ_API_KEY")
|
18 |
+
|
19 |
+
class datachat():
|
20 |
+
|
21 |
+
def __init__(self,file_path):
|
22 |
+
callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
|
23 |
+
|
24 |
+
self.llm = ChatGroq(temperature=0, model_name="mixtral-8x7b-32768",callback_manager=callback_manager)
|
25 |
+
self.instruction = """
|
26 |
+
As a python coder create a pythonic response for the query with reference to the columns in my pandas dataframe{columns}.
|
27 |
+
Instruction:
|
28 |
+
Do not write the whole script just give me a pythonic response for this query and do not extend more than asked. Assume a dataframe variable df_temp.
|
29 |
+
Enclose the generated code in Markdown code embedding format. Do not generate sample output. Answer the question and provide a one-line explanation and stop.
|
30 |
+
|
31 |
+
example:
|
32 |
+
```python
|
33 |
+
output = df['region'].unique()
|
34 |
+
```
|
35 |
+
|
36 |
+
question: {input}
|
37 |
+
|
38 |
+
answer:
|
39 |
+
|
40 |
+
"""
|
41 |
+
self.file_path=file_path
|
42 |
+
|
43 |
+
|
44 |
+
def extract_code(self,response):
|
45 |
+
start = 0
|
46 |
+
q = ""
|
47 |
+
temp_block=""
|
48 |
+
for line in response.splitlines():
|
49 |
+
if '```python' in line and start==0:
|
50 |
+
start=1
|
51 |
+
if '```' == line.strip() and start==1:
|
52 |
+
start =0
|
53 |
+
break
|
54 |
+
if start ==1 and '```' not in line:
|
55 |
+
q=q+'\n'+line
|
56 |
+
return q
|
57 |
+
|
58 |
+
|
59 |
+
def data_ops(self,query):
|
60 |
+
if os.path.isfile('./data/output.csv'):
|
61 |
+
df=pd.read_csv('./data/output.csv')
|
62 |
+
else:
|
63 |
+
df=pd.read_csv(self.file_path)
|
64 |
+
query = query
|
65 |
+
columns=df.columns.tolist()
|
66 |
+
prompt = PromptTemplate.from_template(self.instruction)
|
67 |
+
agent = LLMChain(llm=self.llm,prompt=prompt)
|
68 |
+
response = agent.invoke(input={"columns":columns,"input":query})
|
69 |
+
response = self.extract_code(response['text'])
|
70 |
+
gencode=autopep8.fix_code(response)
|
71 |
+
df_temp=df
|
72 |
+
exec(gencode)
|
73 |
+
df_temp.to_csv('./data/output.csv',index=False)
|
74 |
+
return df_temp
|
requirements.txt
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
#insert python library
|
2 |
+
langchain
|
3 |
+
autopep8
|
4 |
+
langchain_experimental
|
5 |
+
langchain-community
|
6 |
+
python-dotenv
|
7 |
+
langchain-groq
|
8 |
+
unstructured[md]
|
9 |
+
pandas
|
10 |
+
streamlit
|
ui.py
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
from main import datachat as dc
|
3 |
+
|
4 |
+
data_file = r"C:\Users\Naresh Kumar Lahajal\Desktop\DE-LLM\data\input\world_population_data.csv"
|
5 |
+
uploaded_file = st.file_uploader("Choose a file")
|
6 |
+
# Write the uploaded file to a specific location
|
7 |
+
if uploaded_file is not None:
|
8 |
+
with open(data_file, "wb") as f:
|
9 |
+
f.write(uploaded_file.read())
|
10 |
+
|
11 |
+
#chat_object= dc(file_path='./data/employees.csv')
|
12 |
+
chat_object= dc(file_path=data_file)
|
13 |
+
|
14 |
+
st.title("Data Engineering Chatbot")
|
15 |
+
|
16 |
+
# Initialize chat history
|
17 |
+
if "messages" not in st.session_state:
|
18 |
+
st.session_state.messages = []
|
19 |
+
|
20 |
+
# Display chat messages from history on app rerun
|
21 |
+
for message in st.session_state.messages:
|
22 |
+
if message["role"] == 'user':
|
23 |
+
with st.chat_message(message["role"]):
|
24 |
+
st.markdown(message["content"])
|
25 |
+
if message["role"] == 'assistant':
|
26 |
+
with st.chat_message(message["role"]):
|
27 |
+
st.dataframe(message["content"],hide_index=True)
|
28 |
+
|
29 |
+
|
30 |
+
# React to user input
|
31 |
+
if prompt := st.chat_input("What is up?"):
|
32 |
+
# Display user message in chat message container
|
33 |
+
with st.chat_message("user"):
|
34 |
+
st.markdown(prompt)
|
35 |
+
# Add user message to chat history
|
36 |
+
st.session_state.messages.append({"role": "user", "content": prompt})
|
37 |
+
|
38 |
+
response = chat_object.data_ops(prompt)
|
39 |
+
# Display assistant response in chat message container
|
40 |
+
with st.chat_message("assistant"):
|
41 |
+
#st.markdown(response)
|
42 |
+
st.dataframe(response,hide_index=True)
|
43 |
+
# Add assistant response to chat history
|
44 |
+
st.session_state.messages.append({"role": "assistant", "content": response})
|
45 |
+
|
46 |
+
# split the salary and define 10% as HRA, 70% as Basic and 20% as Allowance.
|
47 |
+
# mask the SSN columns as *********1234
|
48 |
+
# convert the hire date column from string to date time and format it as DD-MON-YYYY
|
49 |
+
# combine the first name and last name columns
|