Spaces:
Sleeping
Sleeping
File size: 8,257 Bytes
d950883 |
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 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 |
import gradio as gr
from groq import Groq, RateLimitError
import pandas as pd
from PIL import Image
import pytesseract
import pdfplumber
from pdf2image import convert_from_path
import os
import time
from dotenv import load_dotenv
# Load environment variables from .env file
load_dotenv()
# Set the path to Tesseract executable
pytesseract.pytesseract.tesseract_cmd = os.getenv("TESSERACT_CMD")
# Set the path to Poppler for PDF image extraction
poppler_path = os.getenv("POPPLER_PATH")
# Your Groq API key
YOUR_GROQ_API_KEY = os.getenv("GROQ_API_KEY")
# Initialize Groq client
client = Groq(api_key=YOUR_GROQ_API_KEY)
# Global variable to store extracted text
extracted_text = ""
def extract_text_from_image(image):
return pytesseract.image_to_string(image)
def remove_header_footer(image, header_height=3.9, footer_height=2.27):
width, height = image.size
header_height_pixels = int(header_height * 96) # Convert inches to pixels (assuming 96 DPI)
footer_height_pixels = int(footer_height * 96)
cropping_box = (0, header_height_pixels, width, height - footer_height_pixels)
return image.crop(cropping_box)
def handle_file(file, page_range=None):
global extracted_text
extracted_text = ""
if file is None:
return None, "No file uploaded"
file_name = file.name.lower()
if file_name.endswith(('png', 'jpg', 'jpeg')):
image = Image.open(file)
extracted_text = extract_text_from_image(image)
return image, extracted_text
elif file_name.endswith('pdf'):
text = ""
pdf_images = []
start_page = 1
end_page = None
if page_range:
try:
start_page, end_page = map(int, page_range.split('-'))
except ValueError:
start_page = int(page_range)
end_page = start_page
with pdfplumber.open(file) as pdf_file:
total_pages = len(pdf_file.pages)
end_page = end_page or total_pages
for page_number in range(start_page - 1, end_page):
page = pdf_file.pages[page_number]
page_text = page.extract_text() or ""
text += f"Page {page_number + 1}:\n{page_text}\n"
try:
page_images = convert_from_path(file.name, first_page=page_number + 1, last_page=page_number + 1, poppler_path=poppler_path)
page_images = [remove_header_footer(img) for img in page_images]
pdf_images.extend(page_images)
for img in page_images:
image_text = extract_text_from_image(img)
text += f"Page {page_number + 1} (Image):\n{image_text}\n"
except Exception as e:
text += f"Error processing images on page {page_number + 1}: {e}\n"
extracted_text = text
if pdf_images:
return pdf_images[0], extracted_text
else:
return None, extracted_text
elif file_name.endswith(('xls', 'xlsx')):
df = pd.read_excel(file)
extracted_text = df.to_string()
return None, extracted_text
elif file_name.endswith('csv'):
df = pd.read_csv(file)
extracted_text = df.to_string()
return None, extracted_text
else:
return None, "Unsupported file type"
def split_text(text, max_length=2000):
words = text.split()
chunks = []
current_chunk = []
current_length = 0
for word in words:
word_length = len(word) + 1 # +1 for the space or punctuation
if current_length + word_length > max_length:
chunks.append(" ".join(current_chunk))
current_chunk = [word]
current_length = word_length
else:
current_chunk.append(word)
current_length += word_length
if current_chunk:
chunks.append(" ".join(current_chunk))
return chunks
def is_rate_limited():
# Implement a method to check rate limit status if needed
return False
def chat_groq_sync(user_input, history, extracted_text):
retries = 5
while retries > 0:
rate_limit_status = is_rate_limited()
if rate_limit_status:
return f"{rate_limit_status} Please try again later."
messages = [{"role": "system", "content": "The following text is extracted from the uploaded file:\n" + extracted_text}]
for msg in history:
messages.append({"role": "user", "content": msg[0]})
messages.append({"role": "assistant", "content": msg[1]})
messages.append({"role": "user", "content": user_input})
try:
response = client.chat.completions.create(
model="llama3-70b-8192",
messages=messages,
max_tokens=1000,
temperature=0.4
)
response_content = response.choices[0].message.content
return response_content
except RateLimitError as e:
error_info = e.args[0] if e.args else {}
error_message = error_info.get('error', {}).get('message', '') if isinstance(error_info, dict) else str(error_info)
wait_time = 60
if 'try again in' in error_message:
try:
wait_time = float(error_message.split('try again in ')[-1].split('s')[0])
except ValueError:
pass
print(f"Rate limit error: {error_message}")
print(f"Retrying in {wait_time:.2f} seconds...")
retries -= 1
if retries > 0:
time.sleep(wait_time)
else:
return "Rate limit exceeded. Please try again later."
except Exception as e:
print(f"An unexpected error occurred: {e}")
return "An unexpected error occurred. Please try again later."
def update_chat(user_input, history):
global extracted_text
response = chat_groq_sync(user_input, history, extracted_text)
history.append((user_input, response))
return history, history, ""
with gr.Blocks() as demo:
with gr.Row():
with gr.Column(scale=1):
gr.Markdown("# RAG Chatbot")
gr.Markdown("Check out the [GitHub](https://github.com/anshh-arora?tab=repositories) for more information.")
file = gr.File(label="Upload your file")
page_range = gr.Textbox(label="If the uploaded document is a PDF and has more than 10 pages, enter the page range (e.g., 1-3) or specific page number (e.g., 2):", lines=1, visible=False, interactive=True)
file_upload_button = gr.Button("Upload File")
image_display = gr.Image(label="Uploaded Image", visible=False)
extracted_text_display = gr.Textbox(label="Extracted Text", interactive=False)
with gr.Column(scale=3):
gr.Markdown("# Chat with your file")
history = gr.State([])
with gr.Column():
chatbot = gr.Chatbot(height=500, bubble_full_width=False)
user_input = gr.Textbox(placeholder="Enter Your Query", visible=True, scale=7, interactive=True)
clear_btn = gr.Button("Clear")
undo_btn = gr.Button("Undo")
user_input.submit(update_chat, [user_input, history], [chatbot, history, user_input])
clear_btn.click(lambda: ([], []), None, [chatbot, history])
undo_btn.click(lambda h: h[:-2], history, history)
def show_page_range_input(file):
if file and file.name.lower().endswith('pdf'):
with pdfplumber.open(file) as pdf_file:
if len(pdf_file.pages) > 10:
return gr.update(visible=True)
return gr.update(visible=False)
file.change(show_page_range_input, inputs=file, outputs=page_range)
file_upload_button.click(handle_file, [file, page_range], [image_display, extracted_text_display])
if __name__ == "__main__":
demo.launch(share=True)
|