File size: 4,293 Bytes
57b7c34 b22506c a9402fb b22506c 57b7c34 b22506c 57b7c34 b22506c 0427a5f 0b04e53 4fa7dc2 b22506c b2ff95a b22506c b2ff95a d02fb12 8b1095d b2ff95a d02fb12 991ebc9 b22506c b2ff95a b22506c b2464dd 437e9bc b2ff95a b22506c b2ff95a 5b6d1c4 b22506c 0b04e53 b22506c |
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 |
import gradio as gr
import openpyxl
import PyPDF2
import pandas as pd
from PIL import Image
import pytesseract # Replaced EasyOCR
import io
import os
from huggingface_hub import InferenceClient
# Access the Hugging Face token from the environment variable
hf_token = os.environ.get("HF_TOKEN")
def reconcile_statements(erp_file, bank_file):
yield "⏳ Processing your request... Please wait.", ""
# your existing code block...
try:
# File parsing...
# Extract ERP statement
erp_statement = ""
erp_filename = erp_file.name
if erp_filename.endswith((".xlsx", ".xls")):
workbook = openpyxl.load_workbook(erp_filename)
sheet = workbook.active
for row in sheet.iter_rows():
for cell in row:
erp_statement += str(cell.value) + "\t"
erp_statement += "\n"
elif erp_filename.endswith(".pdf"):
pdf_reader = PyPDF2.PdfReader(erp_filename)
for page in pdf_reader.pages:
erp_statement += page.extract_text() or ""
elif erp_filename.endswith((".jpg", ".jpeg", ".png")):
image = Image.open(io.BytesIO(erp_file.read()))
erp_statement = pytesseract.image_to_string(image) # Tesseract OCR
elif erp_filename.endswith(".csv"):
df = pd.read_csv(erp_filename)
erp_statement = df.to_string()
else:
raise ValueError("Unsupported ERP file format.")
# Extract bank statement (similar logic as above)
bank_statement = ""
bank_filename = bank_file.name
if bank_filename.endswith((".xlsx", ".xls")):
workbook = openpyxl.load_workbook(bank_filename)
sheet = workbook.active
for row in sheet.iter_rows():
for cell in row:
bank_statement += str(cell.value) + "\t"
bank_statement += "\n"
elif bank_filename.endswith(".pdf"):
pdf_reader = PyPDF2.PdfReader(bank_filename)
for page in pdf_reader.pages:
bank_statement += page.extract_text() or ""
elif bank_filename.endswith((".jpg", ".jpeg", ".png")):
image = Image.open(io.BytesIO(bank_file.read()))
bank_statement = pytesseract.image_to_string(image) # Tesseract OCR
elif bank_filename.endswith(".csv"):
df = pd.read_csv(bank_filename)
bank_statement = df.to_string()
else:
raise ValueError("Unsupported bank file format.")
# Hugging Face request...
prompt = f"Reconcile these statements:\nERP:\n{erp_statement}\nBank:\n{bank_statement}"
client = InferenceClient(provider="together", api_key=hf_token)
completion = client.chat.completions.create(
model="deepseek-ai/DeepSeek-R1",
messages=[{"role": "user", "content": prompt}],
)
if completion.choices:
reconciliation_results = completion.choices[0].message.get('content', '')
else:
reconciliation_results = "⚠️ No response received from the model."
output = f"""
<div style="font-family: 'Segoe UI', ...">
<h2>🔍 Reconciliation Results</h2>
<div style="...">
<pre>{reconciliation_results}</pre>
</div>
</div>
"""
yield "✅ Processing complete!", output
except Exception as e:
yield f"❌ Error: {e}", f"<h1>Error</h1><p>{e}</p>"
with gr.Blocks(css="""
#company-logo {
width: 25%;
margin: auto;
display: block;
}
""") as iface:
gr.Image("logo_Icon.png", elem_id="company-logo", label="Beiing Human")
status_text = gr.Markdown("👋 Upload your files to begin reconciliation.")
with gr.Row():
erp_input = gr.File(label="📂 Upload ERP Statement", type="filepath")
bank_input = gr.File(label="📂 Upload Bank Statement", type="filepath")
submit_btn = gr.Button("🔄 Start Reconciliation")
result_output = gr.HTML()
submit_btn.click(
fn=reconcile_statements,
inputs=[erp_input, bank_input],
outputs=[status_text, result_output]
)
if __name__ == "__main__":
iface.launch(debug=True) |