AI Image Caption Generator

 import numpy as np

import tensorflow as tf

from tensorflow.keras.applications.inception_v3 import InceptionV3, preprocess_input

from tensorflow.keras.preprocessing import image

from tensorflow.keras.models import Model, load_model

import pickle

import cv2

import streamlit as st

from PIL import Image

# Load Pretrained InceptionV3 Model for Image Feature Extraction

base_model = InceptionV3(weights='imagenet')

model = Model(inputs=base_model.input, outputs=base_model.layers[-2].output)

# Load Pretrained Captioning Model

captioning_model = load_model("image_captioning_model.h5")

# Load Tokenizer & Word Mappings

with open("tokenizer.pickle", "rb") as handle:

    tokenizer = pickle.load(handle)

max_length = 35  # Max caption length

# Extract Features from Image

def extract_features(img_path):

    img = image.load_img(img_path, target_size=(299, 299))

    img = image.img_to_array(img)

    img = np.expand_dims(img, axis=0)

    img = preprocess_input(img)

    feature_vector = model.predict(img)

    return feature_vector

# Generate Caption

def generate_caption(img_path):

    image_features = extract_features(img_path)

    caption = "startseq"

    

    for i in range(max_length):

        sequence = [tokenizer.word_index[word] for word in caption.split() if word in tokenizer.word_index]

        sequence = tf.keras.preprocessing.sequence.pad_sequences([sequence], maxlen=max_length)

        predicted_index = np.argmax(captioning_model.predict([image_features, sequence]), axis=-1)

        word = tokenizer.index_word.get(predicted_index[0], "")

        if word == "endseq":

            break

        caption += " " + word

    

    return caption.replace("startseq", "").replace("endseq", "").strip()

# Streamlit Web Interface

st.title("🖼️ AI Image Caption Generator")

uploaded_file = st.file_uploader("Upload an image...", type=["jpg", "png", "jpeg"])

if uploaded_file is not None:

    img = Image.open(uploaded_file)

    st.image(img, caption="Uploaded Image", use_column_width=True)

    

    # Save the uploaded image temporarily

    img_path = "temp.jpg"

    img.save(img_path)

    

    # Generate Caption

    with st.spinner("Generating Caption..."):

        caption_text = generate_caption(img_path)

    

    st.subheader("📝 Generated Caption:")

    st.write(caption_text)

AI Resume Scorer

 import fitz  # PyMuPDF for PDF parsing

import docx2txt

import spacy

import re

from collections import Counter

import tkinter as tk

from tkinter import filedialog, messagebox

# Load NLP Model (English)

nlp = spacy.load("en_core_web_sm")

# Job Description (Example)

job_description = """

We are looking for a Data Scientist with expertise in Python, Machine Learning, and Data Analysis.

Candidates must have experience with Pandas, NumPy, and Scikit-learn.

Strong communication and teamwork skills are required.

"""

# Function to extract text from PDF

def extract_text_from_pdf(pdf_path):

    text = ""

    doc = fitz.open(pdf_path)

    for page in doc:

        text += page.get_text()

    return text

# Function to extract text from DOCX

def extract_text_from_docx(docx_path):

    return docx2txt.process(docx_path)

# Function to clean and preprocess text

def clean_text(text):

    text = re.sub(r"\s+", " ", text)  # Remove extra spaces

    text = text.lower()  # Convert to lowercase

    return text

# Function to extract keywords using NLP

def extract_keywords(text):

    doc = nlp(text)

    keywords = [token.text for token in doc if token.is_alpha and not token.is_stop]

    return Counter(keywords)

# Function to score the resume

def score_resume(resume_text, job_description):

    resume_keywords = extract_keywords(resume_text)

    job_keywords = extract_keywords(job_description)

    # Calculate Keyword Match Score

    matched_keywords = sum((resume_keywords & job_keywords).values())

    total_keywords = sum(job_keywords.values())

    keyword_score = (matched_keywords / total_keywords) * 100 if total_keywords else 0

    # Readability Score (Basic: Word Count / Sentence Count)

    sentence_count = len(re.findall(r"[.!?]", resume_text))

    word_count = len(resume_text.split())

    readability_score = (word_count / (sentence_count + 1)) * 2  # Simplified readability measure

    # Final Score Calculation (Weighted Average)

    final_score = (keyword_score * 0.7) + (readability_score * 0.3)

    return round(final_score, 2), keyword_score, readability_score

# GUI for File Upload

def upload_file():

    file_path = filedialog.askopenfilename(filetypes=[("PDF Files", "*.pdf"), ("Word Files", "*.docx")])

    

    if file_path:

        if file_path.endswith(".pdf"):

            resume_text = extract_text_from_pdf(file_path)

        elif file_path.endswith(".docx"):

            resume_text = extract_text_from_docx(file_path)

        else:

            messagebox.showerror("Error", "Unsupported file format!")

            return

        

        # Clean and score resume

        cleaned_resume = clean_text(resume_text)

        final_score, keyword_score, readability_score = score_resume(cleaned_resume, job_description)

        

        # Show results

        messagebox.showinfo("Resume Score", f"📄 Resume Score: {final_score}%\n\n"

                                             f"🔑 Keyword Match: {keyword_score:.2f}%\n"

                                             f"📖 Readability Score: {readability_score:.2f}%")

# GUI Setup

root = tk.Tk()

root.title("AI Resume Scorer")

root.geometry("300x200")

upload_btn = tk.Button(root, text="Upload Resume", command=upload_file, padx=10, pady=5)

upload_btn.pack(pady=20)

root.mainloop()

Job Application Tracker

 import tkinter as tk

from tkinter import ttk, messagebox

import sqlite3

import pandas as pd

import smtplib

# Database Setup

conn = sqlite3.connect("job_tracker.db")

cursor = conn.cursor()

cursor.execute("""

    CREATE TABLE IF NOT EXISTS jobs (

        id INTEGER PRIMARY KEY AUTOINCREMENT,

        company TEXT,

        role TEXT,

        date_applied TEXT,

        status TEXT

    )

""")

conn.commit()

# GUI Application

class JobTrackerApp:

    def __init__(self, root):

        self.root = root

        self.root.title("Job Application Tracker")

        self.root.geometry("600x400")

        # Labels

        ttk.Label(root, text="Company:").grid(row=0, column=0)

        ttk.Label(root, text="Role:").grid(row=1, column=0)

        ttk.Label(root, text="Date Applied:").grid(row=2, column=0)

        ttk.Label(root, text="Status:").grid(row=3, column=0)

        # Entry Fields

        self.company_entry = ttk.Entry(root)

        self.role_entry = ttk.Entry(root)

        self.date_entry = ttk.Entry(root)

        self.status_combo = ttk.Combobox(root, values=["Pending", "Interview", "Rejected", "Hired"])

        

        self.company_entry.grid(row=0, column=1)

        self.role_entry.grid(row=1, column=1)

        self.date_entry.grid(row=2, column=1)

        self.status_combo.grid(row=3, column=1)

        # Buttons

        ttk.Button(root, text="Add Job", command=self.add_job).grid(row=4, column=0)

        ttk.Button(root, text="Show Jobs", command=self.show_jobs).grid(row=4, column=1)

        ttk.Button(root, text="Export to CSV", command=self.export_csv).grid(row=5, column=0)

        ttk.Button(root, text="Send Follow-up", command=self.send_followup).grid(row=5, column=1)

    def add_job(self):

        company = self.company_entry.get()

        role = self.role_entry.get()

        date = self.date_entry.get()

        status = self.status_combo.get()

        if not company or not role or not date or not status:

            messagebox.showerror("Error", "All fields are required!")

            return

        

        cursor.execute("INSERT INTO jobs (company, role, date_applied, status) VALUES (?, ?, ?, ?)", 

                       (company, role, date, status))

        conn.commit()

        messagebox.showinfo("Success", "Job Application Added!")

    def show_jobs(self):

        jobs_window = tk.Toplevel(self.root)

        jobs_window.title("Job Applications")

        tree = ttk.Treeview(jobs_window, columns=("ID", "Company", "Role", "Date", "Status"), show="headings")

        tree.heading("ID", text="ID")

        tree.heading("Company", text="Company")

        tree.heading("Role", text="Role")

        tree.heading("Date", text="Date Applied")

        tree.heading("Status", text="Status")

        tree.pack(fill="both", expand=True)

        cursor.execute("SELECT * FROM jobs")

        for row in cursor.fetchall():

            tree.insert("", "end", values=row)

    def export_csv(self):

        cursor.execute("SELECT * FROM jobs")

        data = cursor.fetchall()

        df = pd.DataFrame(data, columns=["ID", "Company", "Role", "Date Applied", "Status"])

        df.to_csv("job_applications.csv", index=False)

        messagebox.showinfo("Exported", "Job Applications saved as CSV!")

    def send_followup(self):

        email = "your-email@gmail.com"  # Change to your email

        password = "your-password"  # Use App Password for security

        cursor.execute("SELECT company, role FROM jobs WHERE status='Pending'")

        pending_jobs = cursor.fetchall()

        if not pending_jobs:

            messagebox.showinfo("No Follow-ups", "No pending applications to follow up on.")

            return

        

        msg = "Subject: Follow-up on Job Applications\n\n"

        msg += "Here are your pending job applications:\n"

        for company, role in pending_jobs:

            msg += f"- {role} at {company}\n"

        try:

            server = smtplib.SMTP("smtp.gmail.com", 587)

            server.starttls()

            server.login(email, password)

            server.sendmail(email, email, msg)

            server.quit()

            messagebox.showinfo("Email Sent", "Follow-up email sent successfully!")

        except Exception as e:

            messagebox.showerror("Error", f"Failed to send email: {e}")

# Run App

root = tk.Tk()

app = JobTrackerApp(root)

root.mainloop()

Automated Email Responder

 import imaplib

import smtplib

import email

from email.mime.text import MIMEText

import openai

# Gmail Credentials

EMAIL_USER = "your_email@gmail.com"

EMAIL_PASS = "your_app_password"

# OpenAI API Key (Optional)

OPENAI_API_KEY = "your_openai_api_key"

openai.api_key = OPENAI_API_KEY

# Connect to Gmail Inbox

def check_inbox():

    mail = imaplib.IMAP4_SSL("imap.gmail.com")

    mail.login(EMAIL_USER, EMAIL_PASS)

    mail.select("inbox")

    _, messages = mail.search(None, "UNSEEN")

    email_ids = messages[0].split()

    for email_id in email_ids:

        _, msg_data = mail.fetch(email_id, "(RFC822)")

        for response_part in msg_data:

            if isinstance(response_part, tuple):

                msg = email.message_from_bytes(response_part[1])

                sender = msg["From"]

                subject = msg["Subject"]

                body = extract_body(msg)

                print(f"New Email from: {sender}")

                print(f"Subject: {subject}")

                print(f"Body: {body}")

                reply_message = generate_reply(body)

                send_email(sender, reply_message)

    mail.logout()

# Extract Email Body

def extract_body(msg):

    if msg.is_multipart():

        for part in msg.walk():

            if part.get_content_type() == "text/plain":

                return part.get_payload(decode=True).decode()

    return msg.get_payload(decode=True).decode()

# Generate AI-Based Response (Optional)

def generate_reply(user_query):

    prompt = f"Reply professionally to this email: {user_query}"

    response = openai.ChatCompletion.create(

        model="gpt-3.5-turbo",

        messages=[{"role": "system", "content": "You are a professional email assistant."},

                  {"role": "user", "content": prompt}]

    )

    return response["choices"][0]["message"]["content"]

# Send Email Response

def send_email(to_email, message):

    smtp_server = smtplib.SMTP_SSL("smtp.gmail.com", 465)

    smtp_server.login(EMAIL_USER, EMAIL_PASS)

    msg = MIMEText(message)

    msg["Subject"] = "Re: Your Inquiry"

    msg["From"] = EMAIL_USER

    msg["To"] = to_email

    smtp_server.sendmail(EMAIL_USER, to_email, msg.as_string())

    smtp_server.quit()

    print(f"Auto-reply sent to {to_email}")

# Run the Email Bot

if __name__ == "__main__":

    check_inbox()

YouTube Video Summarizer

 pip install youtube-transcript-api nltk sumy

import tkinter as tk

from tkinter import messagebox

from youtube_transcript_api import YouTubeTranscriptApi

from sumy.parsers.plaintext import PlaintextParser

from sumy.nlp.tokenizers import Tokenizer

from sumy.summarizers.lsa import LsaSummarizer

# Function to extract video ID from URL

def extract_video_id(url):

    if "youtube.com/watch?v=" in url:

        return url.split("v=")[1].split("&")[0]

    elif "youtu.be/" in url:

        return url.split("youtu.be/")[1].split("?")[0]

    else:

        return None

# Function to fetch transcript

def get_transcript(video_url):

    video_id = extract_video_id(video_url)

    if not video_id:

        messagebox.showerror("Error", "Invalid YouTube URL")

        return None

    try:

        transcript = YouTubeTranscriptApi.get_transcript(video_id)

        full_text = " ".join([entry["text"] for entry in transcript])

        return full_text

    except Exception as e:

        messagebox.showerror("Error", f"Could not fetch transcript: {str(e)}")

        return None

# Function to summarize text

def summarize_text(text, num_sentences=3):

    parser = PlaintextParser.from_string(text, Tokenizer("english"))

    summarizer = LsaSummarizer()

    summary = summarizer(parser.document, num_sentences)

    return " ".join(str(sentence) for sentence in summary)

# Function to fetch and summarize

def summarize_video():

    video_url = url_entry.get()

    transcript = get_transcript(video_url)

    

    if transcript:

        summary = summarize_text(transcript)

        output_text.delete("1.0", tk.END)

        output_text.insert(tk.END, summary)

# GUI

root = tk.Tk()

root.title("YouTube Video Summarizer")

tk.Label(root, text="Enter YouTube Video URL:").pack()

url_entry = tk.Entry(root, width=50)

url_entry.pack()

tk.Button(root, text="Summarize", command=summarize_video).pack()

output_text = tk.Text(root, height=10, width=60)

output_text.pack()

root.mainloop()