app.py

import streamlit as st
import pandas as pd
import numpy as np
import re
import pickle
import nltk
from nltk.corpus import stopwords
from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences

nltk.download('stopwords')
stop_words = set(stopwords.words('english'))

def cleanResume(resumeText):
    resumeText = re.sub(r'http\S+\s*', ' ', resumeText)
    resumeText = re.sub(r'RT|cc', ' ', resumeText)
    resumeText = re.sub(r'#\S+', '', resumeText)
    resumeText = re.sub(r'@\S+', ' ', resumeText)
    resumeText = re.sub(r'[%s]' % re.escape("!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~"), ' ', resumeText)
    resumeText = re.sub(r'[^\x00-\x7f]', r' ', resumeText)
    resumeText = re.sub(r'\s+', ' ', resumeText)
    resumeText = ' '.join([word for word in resumeText.split() if word.lower() not in stop_words])
    return resumeText

def load_resources():
    with open('tokenizer.pkl', 'rb') as f:
        tokenizer = pickle.load(f)
    with open('label_encoder.pkl', 'rb') as f:
        label_encoder = pickle.load(f)
    model = load_model('deeprank_model_v2.h5')
    return tokenizer, label_encoder, model

def infer(text, tokenizer, label_encoder, model):
    cleaned_text = cleanResume(text)
    sequence = tokenizer.texts_to_sequences([cleaned_text])
    padded_sequence = pad_sequences(sequence, maxlen=500)
    prediction = model.predict(padded_sequence)
    predicted_class = label_encoder.inverse_transform([np.argmax(prediction)])
    return predicted_class[0]

st.title("Resume Category Predictor")

st.write("Upload a resume text file or enter text below to predict the job category.")

uploaded_file = st.file_uploader("Upload Resume (TXT file)", type=["txt"])
user_input = st.text_area("Or paste resume text here:")

if uploaded_file is not None:
    resume_text = uploaded_file.read().decode("utf-8")
    st.session_state["resume_text"] = resume_text
elif user_input:
    resume_text = user_input
    st.session_state["resume_text"] = resume_text
else:
    resume_text = ""

if st.button("Predict Category") and resume_text:
    tokenizer, label_encoder, model = load_resources()
    prediction = infer(resume_text, tokenizer, label_encoder, model)
    st.write(f"Predicted Category: **{prediction}**")