app.py

import streamlit as st
import time
import pandas as pd
import io
from transformers import pipeline
from streamlit_extras.stylable_container import stylable_container
import plotly.express as px
import zipfile
from PyPDF2 import PdfReader
import docx
import os
from comet_ml import Experiment
import re
import numpy as np

st.set_page_config(layout="wide", page_title="Named Entity Recognition App")



# --- Configuration ---
COMET_API_KEY = os.environ.get("COMET_API_KEY")
COMET_WORKSPACE = os.environ.get("COMET_WORKSPACE")
COMET_PROJECT_NAME = os.environ.get("COMET_PROJECT_NAME")

comet_initialized = False
if COMET_API_KEY and COMET_WORKSPACE and COMET_PROJECT_NAME:
    comet_initialized = True

# --- Initialize session state ---
if 'file_upload_attempts' not in st.session_state:
    st.session_state['file_upload_attempts'] = 0

max_attempts = 10

# --- Helper function for model loading ---
@st.cache_resource
def load_ner_model():
    """Loads the pre-trained NER model and caches it."""
    return pipeline("token-classification", model="h2oai/deberta_finetuned_pii", aggregation_strategy="first")

# --- UI Elements ---
st.subheader("9-Personal Data Named Entity Recognition Web App", divider="rainbow")
st.link_button("by nlpblogs", "https://nlpblogs.com", type="tertiary")

expander = st.expander("**Important notes on the 9-Personal Data Named Entity Recognition Web App**")
expander.write('''
    
    **Named Entities:**
    This 9-Personal Data Named Entity Recognition Web App predicts nine (9) categories:
    
    1. **Account-related information**: Account name, account number, and transaction amounts 
    
    2. **Banking details**: BIC, IBAN, and Bitcoin or Ethereum addresses 
     
    3. **Personal information**: Full name, first name, middle name, last name, gender, and date of birth   
     
    4. **Contact information**: Email, phone number, and street address (including building number, city, county, state, and zip code) 
     
    5. **Job-related data**: Job title, job area, job descriptor, and job type
     
    6. **Financial data**: Credit card number, issuer, CVV, and currency information (code, name, and symbol)  
     
    7. **Digital identifiers**: IP addresses (IPv4 and IPv6), MAC addresses, and user agents 
     
    8. **Online presence**: URL, usernames, and passwords    
     
    9. **Other sensitive data**: SSN, vehicle VIN and VRM, phone IMEI, and nearby GPS coordinates 
    
    Results are presented in an easy-to-read table, visualized in an interactive tree map, pie chart, and bar chart, and are available for download along with a Glossary of tags.
    
    **How to Use:**
    Upload your .pdf or .docx file. Then, click the 'Results' button to extract and tag entities in your text data.
    
    **Usage Limits:**
     You can request results up to 10 times.
 
    **Customization:**
    To change the app's background color to white or black, click the three-dot menu on the right-hand side of your app, go to Settings and then Choose app theme, colors and fonts.
    
    **Technical issues:**
    If your connection times out, please refresh the page or reopen the app's URL.
    
    For any errors or inquiries, please contact us at info@nlpblogs.com
    
''')


     
    
with st.sidebar:
    container = st.container(border=True)
    container.write("**Named Entity Recognition (NER)** is the task of extracting and tagging entities in text data. Entities can be persons, organizations, locations, countries, products, events etc.")
    st.subheader("Related NLP Web Apps", divider="rainbow")
    st.link_button("8-Named Entity Recognition Web App", "https://nlpblogs.com/shop/named-entity-recognition-ner/8-named-entity-recognition-web-app/", type="primary")

# --- File Upload ---
upload_file = st.file_uploader("Upload your file. Accepted file formats include: .pdf, .docx", type=['pdf', 'docx'])
text = None
df = None

if upload_file is not None:
    file_extension = upload_file.name.split('.')[-1].lower()
    if file_extension == 'pdf':
        try:
            pdf_reader = PdfReader(upload_file)
            text = ""
            for page in pdf_reader.pages:
                text += page.extract_text()
            st.write("File uploaded successfully. Due to security protocols, the file content is hidden.")
        except Exception as e:
            st.error(f"An error occurred while reading PDF: {e}")
            text = None
    elif file_extension == 'docx':
        try:
            doc = docx.Document(upload_file)
            text = "\n".join([para.text for para in doc.paragraphs])
            st.write("File uploaded successfully. Due to security protocols, the file content is hidden.")
        except Exception as e:
            st.error(f"An error occurred while reading docx: {e}")
            text = None
    else:
        st.warning("Unsupported file type.")
        text = None

st.divider()

# --- Results Button and Processing Logic ---
if st.button("Results"):
    if not comet_initialized:
        st.warning("Comet ML not initialized. Check environment variables if you wish to log data.")

    if st.session_state['file_upload_attempts'] >= max_attempts:
        st.error(f"You have requested results {max_attempts} times. You have reached your daily request limit.")
        st.stop()

    if text is None:
        st.warning("Please upload a supported file (.pdf or .docx) before requesting results.")
        st.stop()

    st.session_state['file_upload_attempts'] += 1

    with st.spinner("Analyzing text...", show_time=True):
        # Load model (cached)
        model = load_ner_model()
        text_entities = model(text)
        df = pd.DataFrame(text_entities)

        # Clean and filter DataFrame
        pattern = r'[^\w\s]'
        df['word'] = df['word'].replace(pattern, '', regex=True)
        df = df.replace('', 'Unknown').dropna()

        if df.empty:
            st.warning("No entities were extracted from the uploaded text.")
            st.stop()

        if comet_initialized:
            experiment = Experiment(
                api_key=COMET_API_KEY,
                workspace=COMET_WORKSPACE,
                project_name=COMET_PROJECT_NAME,
            )
            experiment.log_parameter("input_text_length", len(text))
            experiment.log_table("predicted_entities", df)

        # --- Display Results ---
        properties = {"border": "2px solid gray", "color": "blue", "font-size": "16px"}
        df_styled = df.style.set_properties(**properties)
        st.dataframe(df_styled, use_container_width=True)

        with st.expander("See Glossary of tags"):
            st.write('''
            '**word**': ['entity extracted from your text data']
            
            '**score**': ['accuracy score; how accurately a tag has been assigned to a given entity']
            
            '**entity_group**': ['label (tag) assigned to a given extracted entity']
            
            '**start**': ['index of the start of the corresponding entity']
            
            '**end**': ['index of the end of the corresponding entity']
            ''')

        # --- Visualizations ---
        st.subheader("Tree map", divider="rainbow")
        fig_treemap = px.treemap(df, path=[px.Constant("all"), 'word', 'entity_group'],
                                 values='score', color='entity_group')
        fig_treemap.update_layout(margin=dict(t=50, l=25, r=25, b=25))
        st.plotly_chart(fig_treemap)
        if comet_initialized:
            experiment.log_figure(figure=fig_treemap, figure_name="entity_treemap")

        value_counts1 = df['entity_group'].value_counts()
        final_df_counts = value_counts1.reset_index().rename(columns={"index": "entity_group"})

        col1, col2 = st.columns(2)
        with col1:
            st.subheader("Pie Chart", divider="rainbow")
            fig_pie = px.pie(final_df_counts, values='count', names='entity_group', hover_data=['count'], labels={'count': 'count'}, title='Percentage of predicted labels')
            fig_pie.update_traces(textposition='inside', textinfo='percent+label')
            st.plotly_chart(fig_pie)
            if comet_initialized:
                experiment.log_figure(figure=fig_pie, figure_name="label_pie_chart")

        with col2:
            st.subheader("Bar Chart", divider="rainbow")
            fig_bar = px.bar(final_df_counts, x="count", y="entity_group", color="entity_group", text_auto=True, title='Occurrences of predicted labels')
            st.plotly_chart(fig_bar)
            if comet_initialized:
                experiment.log_figure(figure=fig_bar, figure_name="label_bar_chart")

        # --- Downloadable Content ---
        dfa = pd.DataFrame(
            data={
                'word': ['entity extracted from your text data'],
                'score': ['accuracy score; how accurately a tag has been assigned to a given entity'],
                'entity_group': ['label (tag) assigned to a given extracted entity'],
                'start': ['index of the start of the corresponding entity'],
                'end': ['index of the end of the corresponding entity'],
            })

        buf = io.BytesIO()
        with zipfile.ZipFile(buf, "w") as myzip:
            myzip.writestr("Summary of the results.csv", df.to_csv(index=False))
            myzip.writestr("Glossary of tags.csv", dfa.to_csv(index=False))

        with stylable_container(
            key="download_button",
            css_styles="""button { background-color: yellow; border: 1px solid black; padding: 5px; color: black; }""",
        ):
            st.download_button(
                label="Download zip file",
                data=buf.getvalue(),
                file_name="nlpblogs_ner_results.zip",
                mime="application/zip",
            )
            if comet_initialized:
                experiment.log_asset(buf.getvalue(), file_name="downloadable_results.zip")

        st.divider()
        if comet_initialized:
            experiment.end()

st.write(f"Number of times you requested results: **{st.session_state['file_upload_attempts']}/{max_attempts}**")