Spaces:
Sleeping
Sleeping
File size: 3,135 Bytes
330c288 67cab39 330c288 67cab39 330c288 67cab39 330c288 67cab39 330c288 67cab39 330c288 67cab39 330c288 67cab39 330c288 67cab39 330c288 67cab39 330c288 67cab39 330c288 67cab39 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 |
import os
import pandas as pd
import google.generativeai as genai
import numpy as np
import gradio as gr
# Initialize an empty DataFrame with columns 'Title' and 'Text'
df = pd.DataFrame(columns=['Title', 'Text'])
# Mapping filenames to custom titles
title_mapping = {
'company.txt': 'company_data',
'products.txt': 'product_data',
'shipping.txt': 'shipping_data'
}
# Loop through each file in the current directory and process the relevant ones
for file_name in os.listdir('.'):
if file_name in title_mapping:
try:
with open(file_name, 'r', encoding='utf-8') as file:
text = file.read().replace('\n', ' ') # Replace newlines with spaces for cleaner text
custom_title = title_mapping[file_name]
new_row = pd.DataFrame({'Title': [custom_title], 'Text': [text]})
df = pd.concat([df, new_row], ignore_index=True)
except Exception as e:
print(f"Error processing file {file_name}: {e}")
GEMINI_API_KEY = os.getenv("GOOGLE_API_KEY")
if not GEMINI_API_KEY:
return "Error: Gemini API key not found. Please set the GEMINI_API_KEY environment variable."
# Configure the Gemini API
try:
genai.configure(api_key=GEMINI_API_KEY)
except Exception as e:
return f"Error: Failed to configure the Gemini API. Details: {e}"
# Function to embed text using the Google Generative AI API
def embed_text(text):
return genai.embed_content(model='models/embedding-001', content=text, task_type='retrieval_document')['embedding']
# Embed the text and ensure embeddings are stored in the DataFrame
if 'Embeddings' not in df.columns:
df['Embeddings'] = df['Text'].apply(embed_text)
# Function to calculate similarity score between the query and document embeddings
def query_similarity_score(query, vector):
query_embedding = embed_text(query)
return np.dot(query_embedding, vector)
# Function to get the most similar document based on the query
def most_similar_document(query):
# Create a local DataFrame copy to store similarity scores
local_df = df.copy()
# Calculate similarity for all rows
local_df['Similarity'] = local_df['Embeddings'].apply(lambda vector: query_similarity_score(query, vector))
# Sort by similarity score and retrieve the top match
most_similar = local_df.sort_values('Similarity', ascending=False).iloc[0]
return most_similar['Title'], most_similar['Text']
# Function to generate a response using the RAG approach
def RAG(query):
title, text = most_similar_document(query)
model = genai.GenerativeModel('gemini-pro')
prompt = f"Answer this query:\n{query}.\nOnly use this context to answer:\n{text}"
response = model.generate_content(prompt)
return f"{response.text}\n\nSource Document: {title}"
iface = gr.Interface(
fn=RAG, # Main function to handle the query
inputs=[
gr.Textbox(label="Enter Your Query"), # Input for the user's query
],
outputs=gr.Textbox(label="Response"), # Output for the generated response
title="Patrick's Multilingual Query Handler"
)
|