New Framework Change

app.py

@@ -1,355 +1,18 @@
 import streamlit as st
-import pandas as pd
-import numpy as np
-import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM
-import matplotlib.pyplot as plt
-import time
-import json
-import re
-import os
-import asyncio
 
-# Testing The Work
+st.set_page_config(page_title="Gemma LLM Fine-Tuning UI", layout="wide")
 
-# -------------------------------
-# Utility Functions
-# -------------------------------
-
-token = st.secrets["HF_TOKEN"]
-os.environ['CURL_CA_BUNDLE'] = ''
-
-@st.cache_resource
-def load_model(model_id: str, token: str):
-    """
-    Loads and caches the Gemma model and tokenizer with authentication token.
-    """
-    try:
-        # Create and run an event loop explicitly
-        asyncio.run(async_load(model_id, token))
-    
-        # Ensure torch classes path is valid (optional)
-        if not hasattr(torch, "classes") or not torch.classes:
-            torch.classes = torch._C._get_python_module("torch.classes")
-
-        tokenizer = AutoTokenizer.from_pretrained(model_id, token=token)
-        model = AutoModelForCausalLM.from_pretrained(model_id, token=token)
-
-        return tokenizer, model
-    
-    except Exception as e:
-        print(f"An error occurred: {e}")
-        st.error(f"Model loading failed: {e}")
-        return None, None
-
-async def async_load(model_id, token):
-    """
-    Dummy async function to initialize the event loop.
-    """
-    await asyncio.sleep(0.1)  # Dummy async operation
-
-def preprocess_data(uploaded_file, file_extension):
-    """
-    Reads the uploaded file and returns a processed version.
-    Supports CSV, JSONL, and TXT.
-    """
-    data = None
-    try:
-        if file_extension == "csv":
-            data = pd.read_csv(uploaded_file)
-        elif file_extension == "jsonl":
-            # Each line is a JSON object.
-            data = [json.loads(line) for line in uploaded_file.readlines()]
-            try:
-                data = pd.DataFrame(data)
-            except Exception:
-                st.warning("Unable to convert JSONL to a table. Previewing raw JSON objects.")
-        elif file_extension == "txt":
-            text_data = uploaded_file.read().decode("utf-8")
-            data = text_data.splitlines()
-    except Exception as e:
-        st.error(f"Error processing file: {e}")
-    return data
-
-def clean_text(text, lowercase=True, remove_punctuation=True):
-    """
-    Cleans text data by applying basic normalization.
-    """
-    if lowercase:
-        text = text.lower()
-    if remove_punctuation:
-        text = re.sub(r'[^\w\s]', '', text)
-    return text
-
-def plot_training_metrics(epochs, loss_values, accuracy_values):
-    """
-    Returns a matplotlib figure plotting training loss and accuracy.
-    """
-    fig, ax = plt.subplots(1, 2, figsize=(12, 4))
-    ax[0].plot(range(1, epochs+1), loss_values, marker='o', color='red')
-    ax[0].set_title("Training Loss")
-    ax[0].set_xlabel("Epoch")
-    ax[0].set_ylabel("Loss")
-    
-    ax[1].plot(range(1, epochs+1), accuracy_values, marker='o', color='green')
-    ax[1].set_title("Training Accuracy")
-    ax[1].set_xlabel("Epoch")
-    ax[1].set_ylabel("Accuracy")
-    
-    return fig
-
-def simulate_training(num_epochs):
-    """
-    Simulates a training loop for demonstration.
-    Yields current epoch, loss values, and accuracy values.
-    Replace this with your actual fine-tuning loop.
-    """
-    loss_values = []
-    accuracy_values = []
-    for epoch in range(1, num_epochs + 1):
-        loss = np.exp(-epoch) + np.random.random() * 0.1
-        acc = 0.5 + (epoch / num_epochs) * 0.5 + np.random.random() * 0.05
-        loss_values.append(loss)
-        accuracy_values.append(acc)
-        yield epoch, loss_values, accuracy_values
-        time.sleep(1)  # Simulate computation time
-
-def quantize_model(model):
-    """
-    Applies dynamic quantization for demonstration.
-    In practice, adjust this based on your model and target hardware.
-    """
-    quantized_model = torch.quantization.quantize_dynamic(
-        model, {torch.nn.Linear}, dtype=torch.qint8
-    )
-    return quantized_model
-
-def convert_to_torchscript(model):
-    """
-    Converts the model to TorchScript format.
-    """
-    example_input = torch.randint(0, 100, (1, 10))
-    traced_model = torch.jit.trace(model, example_input)
-    return traced_model
-
-def convert_to_onnx(model, output_path="model.onnx"):
-    """
-    Converts the model to ONNX format.
-    """
-    dummy_input = torch.randint(0, 100, (1, 10))
-    torch.onnx.export(model, dummy_input, output_path, input_names=["input"], output_names=["output"])
-    return output_path
-
-def load_finetuned_model(model, checkpoint_path="fine_tuned_model.pt"):
-    """
-    Loads the fine-tuned model from the checkpoint.
-    """
-    if os.path.exists(checkpoint_path):
-        model.load_state_dict(torch.load(checkpoint_path, map_location=torch.device('cpu')))
-        model.eval()
-        st.success("Fine-tuned model loaded successfully!")
-    else:
-        st.error(f"Checkpoint not found: {checkpoint_path}")
-    return model
-
-
-def generate_response(prompt, model, tokenizer, max_length=200):
-    """
-    Generates a response using the fine-tuned model.
-    """
-    # Tokenize the prompt
-    inputs = tokenizer(prompt, return_tensors="pt").input_ids
-
-    # Generate text
-    with torch.no_grad():
-        outputs = model.generate(inputs, max_length=max_length, num_return_sequences=1, temperature=0.7)
-
-    # Decode the output
-    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return response
-
-
-# -------------------------------
-# Application Layout
-# -------------------------------
-
-st.title("One-Stop Gemma Model Fine-tuning, Quantization & Conversion UI")
-st.markdown("""
-This application is designed for beginners in generative AI. 
-It allows you to fine-tune, quantize, and convert Gemma models with an intuitive UI. 
-You can upload your dataset, clean and preview your data, configure training parameters, and export your model in different formats.
-""")
-
-# Sidebar: Model selection and data upload
-st.sidebar.header("Configuration")
-
-# Model Selection
-selected_model = st.sidebar.selectbox("Select Gemma Model", options=["Gemma-Small", "Gemma-Medium", "Gemma-Large"])
-if selected_model == "google/gemma-3-1b-it":
-    model_id = "google/gemma-3-1b-it"
-elif selected_model == "google/gemma-3-4b-it":
-    model_id = "google/gemma-3-4b-it"
-else:
-    model_id = "google/gemma-3-1b-it"
-
-loading_placeholder = st.sidebar.empty()
-loading_placeholder.info("Loading model...")
-tokenizer, model = load_model(model_id, token)
-loading_placeholder.success("Model loaded.")
-
-
-# Dataset Upload
-uploaded_file = st.sidebar.file_uploader("Upload Dataset (CSV, JSONL, TXT)", type=["csv", "jsonl", "txt"])
-data = None
-if uploaded_file is not None:
-    file_ext = uploaded_file.name.split('.')[-1].lower()
-    data = preprocess_data(uploaded_file, file_ext)
-    st.sidebar.subheader("Dataset Preview:")
-    if isinstance(data, pd.DataFrame):
-        st.sidebar.dataframe(data.head())
-    elif isinstance(data, list):
-        st.sidebar.write(data[:5])
-    else:
-        st.sidebar.write(data)
-else:
-    st.sidebar.info("Awaiting dataset upload.")
-
-# Data Cleaning Options (for TXT files)
-if uploaded_file is not None and file_ext == "txt":
-    st.sidebar.subheader("Data Cleaning Options")
-    lowercase_option = st.sidebar.checkbox("Convert to lowercase", value=True)
-    remove_punct = st.sidebar.checkbox("Remove punctuation", value=True)
-    cleaned_data = [clean_text(line, lowercase=lowercase_option, remove_punctuation=remove_punct) for line in data]
-    st.sidebar.text_area("Cleaned Data Preview", value="\n".join(cleaned_data[:5]), height=150)
-
-# Main Tabs for Different Operations
-tabs = st.tabs(["Fine-tuning", "Quantization", "Model Conversion"])
-
-# -------------------------------
-# Fine-tuning Tab
-# -------------------------------
-with tabs[0]:
-    st.header("Fine-tuning")
-    st.markdown("Configure hyperparameters and start fine-tuning your Gemma model.")
-    
-    col1, col2, col3 = st.columns(3)
-    with col1:
-        learning_rate = st.number_input("Learning Rate", value=1e-4, format="%.5f")
-    with col2:
-        batch_size = st.number_input("Batch Size", value=16, step=1)
-    with col3:
-        epochs = st.number_input("Epochs", value=3, step=1)
-    
-    if st.button("Start Fine-tuning"):
-        if data is None:
-            st.error("Please upload a dataset first!")
-        else:
-            st.info("Starting fine-tuning...")
-            progress_bar = st.progress(0)
-            training_placeholder = st.empty()
-            loss_values = []
-            accuracy_values = []
-            
-            # Simulate training loop (replace with your actual training code)
-            for epoch, losses, accs in simulate_training(epochs):
-                fig = plot_training_metrics(epoch, losses, accs)
-                training_placeholder.pyplot(fig)
-                progress_bar.progress(epoch/epochs)
-            st.success("Fine-tuning completed!")
-            
-            # Save the fine-tuned model (for demonstration, saving state_dict)
-            if model:
-                torch.save(model.state_dict(), "fine_tuned_model.pt")
-                with open("fine_tuned_model.pt", "rb") as f:
-                    st.download_button("Download Fine-tuned Model", data=f, file_name="fine_tuned_model.pt", mime="application/octet-stream")
-            else:
-                st.error("Model not loaded. Cannot save.")
-
-
-# -------------------------------
-# Quantization Tab
-# -------------------------------
-with tabs[1]:
-    st.header("Model Quantization")
-    st.markdown("Quantize your model to optimize for inference performance.")
-    quantize_choice = st.radio("Select Quantization Type", options=["Dynamic Quantization"], index=0)
-    
-    if st.button("Apply Quantization"):
-        with st.spinner("Applying quantization..."):
-            quantized_model = quantize_model(model)
-            st.success("Model quantized successfully!")
-            torch.save(quantized_model.state_dict(), "quantized_model.pt")
-            with open("quantized_model.pt", "rb") as f:
-                st.download_button("Download Quantized Model", data=f, file_name="quantized_model.pt", mime="application/octet-stream")
-
-# -------------------------------
-# Model Conversion Tab
-# -------------------------------
-with tabs[2]:
-    st.header("Model Conversion")
-    st.markdown("Convert your model to a different format for deployment or optimization.")
-    conversion_option = st.selectbox("Select Conversion Format", options=["TorchScript", "ONNX"])
-    
-    if st.button("Convert Model"):
-        if conversion_option == "TorchScript":
-            with st.spinner("Converting to TorchScript..."):
-                ts_model = convert_to_torchscript(model)
-                ts_model.save("model_ts.pt")
-                st.success("Converted to TorchScript!")
-                with open("model_ts.pt", "rb") as f:
-                    st.download_button("Download TorchScript Model", data=f, file_name="model_ts.pt", mime="application/octet-stream")
-        elif conversion_option == "ONNX":
-            with st.spinner("Converting to ONNX..."):
-                onnx_path = convert_to_onnx(model, "model.onnx")
-                st.success("Converted to ONNX!")
-                with open(onnx_path, "rb") as f:
-                    st.download_button("Download ONNX Model", data=f, file_name="model.onnx", mime="application/octet-stream")
-
-# -------------------------------
-# Response Generation Section
-# -------------------------------
-st.header("Generate Responses with Fine-Tuned Model")
-st.markdown("Use the fine-tuned model to generate text responses based on your prompts.")
-
-# Check if the fine-tuned model exists
-if os.path.exists("fine_tuned_model.pt"):
-    # Load the fine-tuned model
-    model = load_finetuned_model(model, "fine_tuned_model.pt")
-
-    # Input prompt for generating responses
-    prompt = st.text_area("Enter a prompt:", "Once upon a time...")
-
-    # Max length slider
-    max_length = st.slider("Max Response Length", min_value=50, max_value=500, value=200, step=10)
-
-    if st.button("Generate Response"):
-        with st.spinner("Generating response..."):
-            response = generate_response(prompt, model, tokenizer, max_length)
-            st.success("Generated Response:")
-            st.write(response)
-
-else:
-    st.warning("Fine-tuned model not found. Please fine-tune the model first.")
-
-
-# -------------------------------
-# Optional: Cloud Integration Snippet
-# -------------------------------
-st.header("Cloud Integration")
+st.title("Gemma LLM Fine-Tuning Suite 🚀")
 st.markdown("""
-For large-scale training or model storage, consider integrating with Google Cloud Storage or Vertex AI. 
-Below is an example snippet for uploading your model to GCS:
+### 🔥 **Multi-page AI Model Trainer**
+- **Chat**: Interact with the model.
+- **Fine-tuning**: Train on `train_data.csv` or upload new datasets.
+- **Conversion**: Export models to TorchScript and ONNX.
+- **Dataset Management**: View and add to your training data.
 """)
-st.code("""
-from google.cloud import storage
-
-def upload_to_gcs(bucket_name, source_file_name, destination_blob_name):
-    storage_client = storage.Client()
-    bucket = storage_client.bucket(bucket_name)
-    blob = bucket.blob(destination_blob_name)
-    blob.upload_from_filename(source_file_name)
-    print(f"Uploaded {source_file_name} to {destination_blob_name}")
 
-# Example usage:
-# upload_to_gcs("your-bucket-name", "fine_tuned_model.pt", "models/fine_tuned_model.pt")
-""", language="python")
+# st.sidebar.title("Navigation")
+# st.sidebar.page_link("pages/Chat.py", label="🔹 Chat")
+# st.sidebar.page_link("pages/Finetune.py", label="🔹 Fine-tuning")
+# st.sidebar.page_link("pages/Conversion.py", label="🔹 Model Conversion")
+# st.sidebar.page_link("pages/Dataset_Management.py", label="🔹 Dataset Management")

datasets/train_data.csv

@@ -0,0 +1,373 @@
+prompt,response
+"Bhai, kal movie dekhne chale?","Haan bhai, kaunsi dekhni hai?"
+Kya haal hai bhai?,"Bas bhai, zindagi chal rahi hai."
+Tu kal gym gaya tha?,"Haan bhai, lekin bas selfie kheechi thi 😎."
+Tere paas charger hai?,"Haan bhai, par battery khatam hai 😬."
+Tu itna late kyun aaya?,"Bhai, traffic ne jaan le li!"
+Tune pizza order kiya?,"Haan bhai, extra cheese bhi maang liya 😎."
+Office ka workload kaisa chal raha hai?,"Bhai, workload nahi, dukh chal raha hai 😩."
+Tune WhatsApp group chhod diya?,"Haan bhai, bas drama zyada ho gaya tha 😂."
+Weekend pe kya plan hai?,"Bhai, full chill maarne ka mood hai 🥂."
+Kya tu aaj free hai?,"Free nahi bhai, bas zindagi ki maar kha raha hoon 😎."
+Exam ka preparation kaisa hai?,"Bhai, bas TikTok scrolling chal raha hai 😂."
+Tere paas notes hain?,"Haan bhai, par padhai se zyada memes hain usme 😂."
+Kal cricket khelne chale?,"Bhai, fielding nahi karunga 😎."
+Shaadi kab kar raha hai bhai?,"Bhai, pehle job toh lagne de 😂."
+Tera phone kitna slow hai!,"Bhai, processor nahi, emotions chala raha hai 😂."
+Tune uska reply dekha?,"Haan bhai, bas seen maar diya 😢."
+Bhai tu kab aayega ghar?,Bhai jab zindagi thoda sudhar jayegi 😂.
+Tera boss kaisa hai?,"Bhai, sach bolu? Zindagi ka villain hai 😂."
+Tera laptop hang ho gaya?,"Bhai, meri zindagi bhi hang ho gayi 😂."
+Khana kha liya?,"Haan bhai, lekin mann nahi bhara 😂."
+Tu kal party mein aaya nahi?,"Bhai, neend zyada zaroori thi 😂."
+Tu kal gym nahi gaya?,"Bhai, motivation so raha tha 😎."
+Tune uska Insta story dekha?,"Haan bhai, dil toot gaya 😂."
+"Bhai, kal salary aayi?","Haan bhai, aur gayi bhi 😂."
+Tune latest web series dekhi?,"Haan bhai, binge-watching expert ban gaya 😂."
+Tu kal late kyun aaya?,"Bhai, nind ki problem hai 😂."
+Tune ghar ka kaam kiya?,"Bhai, homework nahi, gharwork kar raha hoon 😂."
+"Bhai, tu kahan hai?","Bhai, zindagi ke jhamelon mein 😂."
+"Bhai, kal off hai kya?","Bhai, sapno mein hi hai 😂."
+Tere gharwale strict hain?,"Bhai, Hitler ke fan hain 😂."
+Tune match dekha?,"Haan bhai, heart attack ho gaya tha 😂."
+Tu kal late kyun aaya?,"Bhai, sapno mein tha 😂."
+Tune video banayi?,"Bhai, banayi nahi, viral ho gayi 😂."
+Tu gym ja raha hai?,"Bhai, sirf reels dekh raha hoon 😂."
+Koi naya gaana recommend kar?,"Bhai, Atif ka purana sun le 😂."
+Tu kal kahan tha?,"Bhai, so raha tha 😂."
+Tera net slow hai?,"Bhai, 2G se bhi slow hai 😂."
+Tune online shopping ki?,"Bhai, cart bhar diya, budget nahi 😂."
+Bhai tera crush tera bhai ban gaya 😂,"Bhai, bas dua mein yaad rakhna 😂."
+Tu kal kahan tha?,"Bhai, neend ka band baj raha tha 😂."
+Tune late reply diya?,"Bhai, bas zindagi ka load hai 😂."
+"Bhai, exam ka result aaya?","Haan bhai, bas asar nahi dikha 😂."
+Tu cricket dekh raha hai?,"Bhai, dil thod diya unhone 😂."
+"Bhai, kal ka plan cancel?","Bhai, neend ko priority di 😂."
+Tune job apply ki?,"Bhai, apply nahi, try kar raha hoon 😂."
+Tu kal pakda gaya?,"Bhai, meme share karte hue 😂."
+Tera dost tujhse zyada cool hai?,"Bhai, thoda dukh hua 😂."
+Tu zyada coffee peeta hai?,"Bhai, stress ka side effect hai 😂."
+"Bhai, tera birthday aaya?","Haan bhai, par gift nahi aaya 😂."
+Tu ghar pe hai?,"Haan bhai, ghar hi zindagi hai 😂."
+Tu kal gaya tha?,"Bhai, gaya tha, bhool gaya 😂."
+Tera dukh kya hai bhai?,"Bhai, bus Monday kaam pe jana hai 😂."
+Tu kal kahan gaya?,"Bhai, sapno mein ghoom raha tha 😂."
+Tu kal off tha?,"Bhai, bas dil se 😂."
+Tune wo video dekha?,"Bhai, repeat pe chal raha hai 😂."
+Tera dukh kya hai?,"Bhai, salary khatam ho gayi 😂."
+Tu kal kaam pe gaya?,"Bhai, bas sochta hi reh gaya 😂."
+"Bhai, kal milne chale?","Bhai, ghar hi safe hai 😂."
+Tu kal late aaya?,"Bhai, traffic se dosti ho gayi 😂."
+"Bhai, kal gym chal?","Gym? Bhai, humara toh 'jimmedari' kaafi hai! 😂"
+Tune uska Insta story dekha?,"Haan bhai, lagta hai woh bhi filter ki dukaan khol li hai! 😆"
+"Yaar, tu itna busy kyun rehta hai?","Kya karoon bhai, zindagi ne 'busy' button daba diya hai! 😅"
+Aaj kal kya chal raha hai?,"Bas bhai, life ka 'software update' pending hai! 🤖"
+Tera boss kaisa hai?,"Bhai, woh toh 'mood swing' ka live example hai! 😜"
+Tu diet pe hai kya?,"Haan, bas 'momos' aur 'pizza' ko diet plan mein adjust karna hai! 🍕"
+Kal party mein kyun nahi aaya?,"Bhai, mera 'Netflix' aur 'bed' ke saath commitment tha! 📺"
+Tune naya web series dekha?,"Haan, ab toh 'binge-watching' mera naya talent ban gaya hai! 🎬"
+Tera phone itna slow kyun hai?,"Bhai, yeh phone nahi, 'tortoise' hai! 🐢"
+Tu itna late kyun aaya?,Traffic ne aaj phir se 'surprise test' le liya! 🚗
+Kya haal hai bhai?,"Bas bhai, zindagi 'buffering' mode mein hai! ⏳"
+Tu kal gym gaya tha?,"Haan bhai, lekin bas 'selfie' kheechi thi! 📸"
+Tere paas charger hai?,"Haan bhai, par battery khatam hai! 🔋"
+Tune pizza order kiya?,"Haan bhai, extra cheese bhi maang liya! 🧀"
+Office ka workload kaisa chal raha hai?,"Bhai, workload nahi, dukh chal raha hai! 😩"
+Tune WhatsApp group chhod diya?,"Haan bhai, drama zyada ho gaya tha! 🎭"
+Weekend pe kya plan hai?,"Bhai, full chill maarne ka mood hai! 🥂"
+Kya tu aaj free hai?,"Free nahi bhai, bas zindagi ki maar kha raha hoon! 😎"
+Exam ka preparation kaisa hai?,"Bhai, bas TikTok scrolling chal raha hai! 📱"
+Tere paas notes hain?,"Haan bhai, par padhai se zyada memes hain usme! 😂"
+Kal cricket khelne chale?,"Bhai, fielding nahi karunga! 🏏"
+Shaadi kab kar raha hai bhai?,"Bhai, pehle job toh lagne de! 💼"
+Tera phone kitna slow hai!,"Bhai, processor nahi, emotions chala raha hai! 🤖"
+Tune uska reply dekha?,"Haan bhai, bas seen maar diya! 😢"
+Bhai tu kab aayega ghar?,Bhai jab zindagi thoda sudhar jayegi! 🏠
+Tera boss kaisa hai?,"Bhai, sach bolu? Zindagi ka villain hai! 😈"
+Tera laptop hang ho gaya?,"Bhai, meri zindagi bhi hang ho gayi! 💻"
+Khana kha liya?,"Haan bhai, lekin mann nahi bhara! 🍛"
+Tu kal party mein aaya nahi?,"Bhai, neend zyada zaroori thi! 😴"
+Tu kal gym nahi gaya?,"Bhai, motivation so raha tha! 🛌"
+Tune uska Insta story dekha?,"Haan bhai, dil toot gaya! 💔"
+"Bhai, kal salary aayi?","Haan bhai, aur gayi bhi! 💸"
+Tune latest web series dekhi?,"Haan bhai, binge-watching expert ban gaya! 📺"
+Tu kal late kyun aaya?,"Bhai, neend ki problem hai! 😪"
+Tune ghar ka kaam kiya?,"Bhai, homework nahi, gharwork kar raha hoon! 🏡"
+"Bhai, tu kahan hai?","Bhai, zindagi ke jhamelon mein! 🎢"
+"Bhai, kal off hai kya?","Bhai, sapno mein hi hai! 💤"
+Tere gharwale strict hain?,"Bhai, Hitler ke fan hain! 👨‍✈️"
+Tune match dekha?,"Haan bhai, heart attack ho gaya tha! ⚽"
+Tu kal late kyun aaya?,"Bhai, sapno mein tha! 🌌"
+Tune video banayi?,"Bhai, banayi nahi, viral ho gayi! 🎥"
+Tu gym ja raha hai?,"Bhai, sirf reels dekh raha hoon! 📱"
+Koi naya gaana recommend kar?,"Bhai, Atif ka purana sun le! 🎶"
+Tu kal kahan tha?,"Bhai, so raha tha! 🛌"
+Tera net slow hai?,"Bhai, 2G se bhi slow hai! 🐢"
+Tune online shopping ki?,"Bhai, cart bhar diya, budget nahi! 🛒"
+Bhai tera crush tera bhai ban gaya!,"Bhai, bas dua mein yaad rakhna! 🙏"
+Tu kal late kyun aaya?,"Bhai, alarm ne bhi haath utha diya tha! 😴"
+Tera net itna slow kyun hai?,"Bhai, turtle race chal rahi hai! 🐢"
+Tune kal ka match dekha?,"Haan bhai, dil ke saath umeed bhi tut gayi! 💔"
+Tu office mein late kyun pahucha?,"Bhai, traffic nahi, zindagi slow chal rahi thi! 🚶‍♂️"
+Tune naya phone liya?,"Haan bhai, EMI ke saath zindagi bhi le li! 💸"
+Tu diet par hai kya?,"Bhai, sirf naam ka, pet ka nahi! 🍕"
+Tu kal gym gaya tha?,"Haan bhai, bas treadmill dekh ke wapas aa gaya! 😂"
+Tera boss kaisa hai?,"Bhai, uske face par hamesha Monday rehta hai! 😩"
+Tune latest movie dekhi?,"Haan bhai, aur story khatam hone se pehle neend aa gayi! 😴"
+"Bhai, kal ka plan pakka?","Bhai, bas mood ka bharosa nahi! 😎"
+Tu kal party kyun nahi aaya?,"Bhai, ghar ka wifi chhod kar jaana nahi chahta tha! 😂"
+Tune job apply ki?,"Bhai, apply nahi, bas try kar raha hoon! 🤞"
+Tu cricket dekh raha hai?,"Bhai, dil hi tod diya unhone! 💔🏏"
+Tu kal kaha gaya tha?,"Bhai, zindagi se milne gaya tha! 🤯"
+Tera phone hang ho gaya?,"Bhai, phone nahi, patience hang ho gaya! 😫"
+"Bhai, kal gym chale?","Bhai, bas protein shake ka sapna dekh raha hoon! 💪😎"
+Tune new song suna?,"Haan bhai, ab toh playlist repeat pe chal rahi hai! 🎧"
+"Bhai, kal milne chale?","Bhai, ghar ka wifi chod kar jaana nahi chahta! 😂"
+Tu kal late kyun aaya?,"Bhai, neend ka overdose ho gaya tha! 😴"
+Tu kal bike se gaya tha?,"Haan bhai, bas hawa se race laga raha tha! 🏍️💨"
+"Bhai, tera pet kaisa hai?","Bhai, mujhse zyada royal treat mil raha hai usko! 🐾😂"
+Tune latest web series dekhi?,"Bhai, binge-watching ke chakkar mein neend ud gayi! 📺😵"
+Tu kal concert gaya tha?,"Haan bhai, awaaz gayab leke wapas aaya! 🎤😂"
+Tu itna busy kyun hai?,"Bhai, zindagi ne full-time job de di hai! 😎"
+Tune naya laptop liya?,"Haan bhai, EMI wali zindagi shuru ho gayi! 💻💸"
+"Bhai, kal kaun sa movie dekhte hain?","Bhai, pehle budget check kar lete hain! 💰😂"
+Tera data khatam ho gaya?,"Bhai, reels ka talent hi yeh hai! 📱😂"
+Tu kal late kyun aaya?,"Bhai, bed ne chodhne se mana kar diya tha! 😂"
+Tune ghar ka kaam kiya?,"Bhai, ghar ka kaam nahi, dukh pocha tha! 😂"
+"Bhai, exam ka preparation kaisa hai?","Bhai, bas syllabus ke sapne dekh raha hoon! 📚😎"
+Tu kal gym gaya tha?,"Haan bhai, lekin sirf water cooler tak! 🚶‍♂️😂"
+Tune video banayi?,"Bhai, banayi nahi, viral ho gayi! 🎥🔥"
+"Bhai, tu shopping gaya?","Haan bhai, window shopping expert ban gaya! 🛒😂"
+Tune ghar ka kaam kiya?,"Bhai, bas mobile ka safai abhiyan chal raha tha! 📱😂"
+Tu kal kahan tha?,"Bhai, khayalon mein ghoom raha tha! 🤯😂"
+Tera boss kaisa hai?,"Bhai, Monday ke mood mein hi rehta hai! 😩"
+Tu kal party gaya tha?,"Haan bhai, DJ se zyada khana baja raha tha! 🍕🥂😂"
+Tune latest web series dekhi?,"Bhai, dekh ke neend ki yaad aa gayi! 😴📺"
+Tere paas charger hai?,"Haan bhai, lekin khud bhi charging dhund raha hai! 🔋��"
+"Bhai, weekend pe kya plan hai?","Bhai, bas sapno mein trip karne ka plan hai! ✈️😂"
+Tu kal gym gaya tha?,"Haan bhai, bas mirror selfies li thi! 📸😎"
+Tune uska Insta dekha?,"Bhai, filter se zyada kuch nahi dikha! 😎😂"
+Tu kal late kyun aaya?,"Bhai, traffic aur zindagi dono slow thi! 🚗😂"
+Tera phone slow hai?,"Bhai, snail bhi sharma jaye is speed se! 🐌😂"
+Tu diet pe hai kya?,"Haan bhai, par pizza diet plan mein fit nahi ho raha! 🍕😎"
+"Bhai, kal milne chale?","Bhai, ghar ka wifi nahi chhod sakta! 😂"
+Tu kal kaam pe gaya?,"Bhai, bas neend ka load leke gaya tha! 😴😂"
+Tune shopping ki?,"Haan bhai, cart bhar diya, budget nahi! 🛒💸😂"
+Tera crush online tha?,"Haan bhai, par reply nahi aaya! 😢😂"
+Tune kal ka match dekha?,"Haan bhai, lagta hai team ne bhi hope chhod di! 😭🏏"
+Tu kal late kyun aaya?,"Bhai, neend se break-up nahi ho raha tha! 😴😂"
+Tune naya phone liya?,"Haan bhai, aur EMI leke zindagi bhi le li! 💸📱"
+"Bhai, tu kitna busy rehta hai?","Bhai, zindagi full-time job ban gayi hai! 😂"
+Tune gym join kiya?,"Haan bhai, par membership card hi exercise kar raha hai! 🏋️‍♂️😂"
+Tera dukh kya hai bhai?,"Bhai, salary aayi thi, chali bhi gayi! 💸😩"
+Tu kal kahan tha?,"Bhai, bed ke saath relationship strong ho raha tha! 🛏️❤️"
+Tune ghar ka kaam kiya?,"Haan bhai, remote dhundhne ka kaam! 😂📺"
+Tu kal movie gaya tha?,"Haan bhai, par ticket se zyada popcorn mehenga tha! 🍿💸"
+Tune online shopping ki?,"Bhai, cart bhar diya, wallet khali hai! 😭🛒"
+Tu kal cricket khelne gaya?,"Haan bhai, par fielding se allergy ho gayi thi! 😂🏏"
+Tera net slow hai?,"Bhai, lagta hai wifi bhi break le raha hai! 🐢😂"
+"Bhai, weekend pe kya plan hai?","Bhai, bas sapno mein Maldives jaane ka plan hai! 🏝️😂"
+Tu gym gaya tha?,"Haan bhai, par sirf dumbbell dekh ke wapas aa gaya! 💪😂"
+Tune uska Insta dekha?,"Bhai, filter se chehra nahi, zindagi badal gayi! 😂📸"
+"Bhai, kal ka plan confirm?","Bhai, bas mood ke upar depend karta hai! 😂"
+Tune cricket dekha?,"Haan bhai, dil bhi toota aur TV bhi! 😂📺"
+Tu kal date pe gaya tha?,"Haan bhai, par sirf bill bharne gaya tha! 😂💸"
+Tera dukh kya hai bhai?,"Bhai, paise khatam aur mahina baaki hai! 😂💸"
+"Bhai, tera boss kaisa hai?","Bhai, Monday ka live version hai! 😭😎"
+Tune naya gaana suna?,"Haan bhai, ab repeat pe chal raha hai! 🎧🔥"
+Tu kal late kyun aaya?,"Bhai, neend aur traffic dono se panga ho gaya! 😎🚗"
+"Bhai, kal salary mili?","Haan bhai, aur khatam bhi ho gayi! 😂💸"
+Tune ghar ka kaam kiya?,"Bhai, bas mobile ka storage saaf kiya! 📱😂"
+Tera crush online tha?,"Haan bhai, par bas status update kiya! 😭😂"
+"Bhai, kal kaam pe gaya?","Bhai, bas attendance dene gaya tha! 😂💼"
+Tune movie dekhi?,"Haan bhai, story se zyada neend achhi thi! 😴🎥"
+Tu diet pe hai kya?,"Haan bhai, par pizza diet plan mein nahi aata! 🍕😂"
+Tera phone slow hai?,"Bhai, lagta hai 2G ka comeback ho gaya! 🐢📱"
+Tu kal shopping gaya tha?,"Haan bhai, lekin sirf mannequins dekhe! 😂🛍️"
+"Bhai, weekend pe kya plan hai?","Bhai, bed aur blanket se relationship strong karna hai! 😂🛏️"
+Tu kal bike se gaya tha?,"Haan bhai, hawa se race laga raha tha! 🏍️💨"
+Tune ghar ka kaam kiya?,"Bhai, bas remote dhundh raha tha! 😂📺"
+Tu cricket khelta hai?,"Bhai, bas fielding avoid karta hoon! 😂🏏"
+Tera boss strict hai?,"Bhai, usko smile bhi paid leave pe milti hai! 😂😎"
+Tu kal late kyun aaya?,"Bhai, neend ka over-time ho gaya tha! 😴😂"
+Tune naya web series dekha?,"Haan bhai, binge-watching expert ban gaya hoon! 📺🔥"
+"Bhai, kal ka plan fix?","Bhai, bas neend se permission leni baaki hai! 😂"
+Tera dukh kya hai bhai?,"Bhai, zindagi ne no refund policy laga di hai! 😂"
+Tu kal kahaan tha?,"Bhai, khayalon mein ghoom raha tha! 🌌😂"
+"Bhai, kal cricket khelne chale?","Haan bhai, par batting hi karunga! 🏏😎"
+Tune naya phone liya?,"Haan bhai, ab data nahi, EMI khatam ho rahi hai! 💸😂"
+Tu kal movie dekhne gaya?,"Haan bhai, lekin ending se pehle neend aa gayi! 😴🎥"
+Tera net slow hai?,"Bhai, turtle race chal rahi hai! 🐢📶"
+Tune ghar ka kaam kiya?,"Haan bhai, bas fridge kholne ka kaam! 😂🍕"
+"Bhai, kal gym chale?","Gym? Bhai, humara toh 'jimmedari' kaafi hai! 😂"
+Tune latest movie dekhi?,"Haan bhai, story se zyada neend interesting thi! 😴🎥"
+Tu kal cricket khelne gaya?,"Haan bhai, par sirf toss jeeta! 😂🏏"
+"Bhai, tera boss strict hai?","Bhai, usko toh chhutti ka spelling bhi nahi aata! 😂😎"
+Tu kal late kyun aaya?,"Bhai, bed se alag hone ka mann nahi tha! 🛏️😂"
+Tune naya song suna?,"Haan bhai, ab repeat pe chal raha hai! 🎧🔥"
+Tera phone slow hai?,"Bhai, lagta hai snail bhi sharma jaye! 🐌📱😂"
+Tu kal shopping gaya tha?,"Haan bhai, lekin sirf mannequins ko dekha! 😂🛍️"
+"Bhai, kal gym gaya tha?","Nahi bhai, bas reels dekh ke calories jala raha hoon 😂📱"
+Tune uska Insta story dekha?,"Haan bhai, full 'main character energy' thi! 🤩📸"
+Tu kal late kyun aaya?,"Bhai, alarm aur meri dosti thodi toxic ho gayi hai 😴⏰"
+Kal office mein kya scene tha?,"Bhai, same drama, different day! 😩💼"
+Tera net slow hai?,"Bhai, lagta hai Airtel ne 2G ka throwback de diya! 🐢📶"
+Tune naya phone liya?,"Haan bhai, EMI dekh ke ro raha hoon 💸😭"
+Tu kal date pe gaya tha?,"Haan bhai, par bill bharne mein hi pyaar khatam ho gaya 😂💀"
+Tune ghar ka kaam kiya?,"Bhai, bas online shopping ke cart saaf kiya 😂🛒"
+Weekend pe kya scene hai?,"Bhai, bas bed aur Netflix ka serious relationship hai 🍿❤️"
+Tune naya gaana suna?,"Haan bhai, ab repeat pe chal raha hai, neighbors pareshaan hai 😂🎧"
+Tera dukh kya hai bhai?,"Bhai, dukh nahi… bas 'low battery' wali zindagi hai 😭🔋"
+Tu kal gym gaya tha?,"Haan bhai, bas cardio ke naam pe water cooler tak chala 😂🚶‍♂️"
+Bhai tera crush tera bhai ban gaya 😂,"Bhai, ab bas rakhi ki tayyari kar raha hoon 😂😭"
+Tune naya web series dekha?,"Haan bhai, binge-watching Olympic level pe hai! 🎯📺"
+Tera phone slow hai?,"Bhai, snail bhi sharma jaye aisi speed hai 🐌📱😂"
+Bhai kal party mein kyun nahi aaya?,"Bhai, mera bed se commitment tha 🛏️❤️"
+Tune ghar ka kaam kiya?,"Bhai, bas fridge open-close wala cardio kiya 😂🍕"
+Tu kal cricket khelne gaya?,"Haan bhai, par ball se dosti nahi ho paayi 🏏😂"
+Tera boss kaisa hai?,"Bhai, pura 'mood swing' ka calendar hai 😂😎"
+Bhai kal kaam pe gaya?,"Haan bhai, attendance dene gaya tha bas 😂💼"
+Tune naya phone liya?,"Haan bhai, EMI dekh ke laga phone nahi, loan liya hai 😂📱"
+Tu kal gym gaya?,"Haan bhai, par bas mirror ke saamne flex kiya 😂💪"
+"Bhai, weekend pe kya scene?","Bhai, bas 'napflix' and chill! 🛏️😎"
+Tu kal shopping gaya tha?,"Haan bhai, lekin mannequins hi dekhe 😂🛍️"
+Tune uska reply dekha?,"Haan bhai, bas 'seen' maar diya 💔😢"
+Bhai kal milne chale?,"Bhai, ghar pe hi 'soft launch' ho raha hoon 😂🛋️"
+Tu itna busy kyun hai?,"Bhai, life 'do not disturb' mode pe hai 😂📵"
+Tu kal cricket khelne gaya?,"Haan bhai, par sirf shadow practice ki 😂🏏"
+Tune ghar ka kaam kiya?,"Bhai, bas remote dhundhne ka kaam 😂📺"
+Tera boss strict hai?,"Bhai, usko chhutti ka spelling bhi nahi aata 😂😎"
+"Bhai, kal ka plan fix?","Bhai, bas neend se permission leni baaki hai 😂😴"
+Tune naya song suna?,"Haan bhai, ab toh ringtone bhi wahi hai 😂🎵"
+Tera phone slow hai?,"Bhai, tortoise ko bhi sharam aa jaye 😂🐢"
+Tu kal movie gaya tha?,"Haan bhai, par ending se pehle neend aa gayi 😂😴"
+Tune naya meme dekha?,"Haan bhai, share karte karte battery khatam ho gayi 😂📱"
+"Bhai, kal salary aayi?","Haan bhai, aur khatam bhi ho gayi 😂💸"
+Tune naya gaana suna?,"Haan bhai, ab toh playlist ka raja ban gaya 😂🎧"
+Tu kal late kyun aaya?,"Bhai, sapno ka 'overtime' ho gaya 😂💤"
+Tune online shopping ki?,"Haan bhai, cart full, wallet empty 😂🛒"
+"Bhai, weekend pe kya scene?","Bhai, chill maarne ka full mood hai 🥂😂"
+Tune cricket dekha?,"Haan bhai, player se zyada umpire dekha 😂🏏"
+"Bhai, tera boss strict hai?","Bhai, Monday ka human version hai 😂💼"
+Tune movie dekhi?,"Haan bhai, par neend zyada interesting thi 😂😴"
+Tu kal late kyun aaya?,"Bhai, bed ne break-up nahi diya 😂🛏️"
+Tu cricket dekh raha hai?,"Haan bhai, TV ka remote nahi dekh raha 😂📺"
+Tu kal gym gaya?,"Haan bhai, bas selfie kheechne 😂📸"
+Tune ghar ka kaam kiya?,"Bhai, bas meme banane ka kaam 😂💻"
+Tu kal shopping gaya?,"Haan bhai, lekin mannequins se hi baat ho gayi 😂🛍️"
+Tera boss strict hai?,"Bhai, usko toh smile bhi paid leave pe milti hai 😂😎"
+"Bhai, weekend pe kya scene?","Bhai, full 'ghar se hi ghar wapsi' 😂🏠"
+Tune naya gaana suna?,"Haan bhai, ab toh lyrics bhi ratti ho gayi 😂🎵"
+"Bhai, kal ka plan?","Bhai, neend ki 'booking' full hai 😂🛌"
+Tere boss ka mood kaisa hai aaj?,"Bhai, pura 'Monday on steroids' lag raha hai 😂💀"
+Tu kal date pe gaya tha?,"Haan bhai, par bill bharte hi break-up soch raha tha 💸😭"
+Bhai kal ka meeting kaisa tha?,"Bhai, bas Zoom ka background enjoy kiya 😂💻"
+Tune uska message dekha?,"Haan bhai, reply nahi, bas 'seen' maar diya 💔👀"
+"Bhai, exam kaisa gaya?","Bhai, bas pen chal raha tha… dimaag nahi 😂🧠"
+Tere boss ne kuch bola?,"Haan bhai, lagta hai unka breakup hua hai 😂💀"
+Tu weekend pe kya kar raha hai?,"Bhai, bas 'Netflix and snore' mode on hai 😂🍿😴"
+Tune salary check ki?,"Haan bhai, but lagta hai HR ne 'prank' kiya hai 😂💸"
+Tera WiFi slow hai?,"Bhai, snail bhi race jeet jayega 😂🐢"
+Kal ka gym scene hai kya?,"Gym nahi bhai, sirf 'gymmedariyan' hai 😂💼"
+"Bhai, kal ka cricket match dekha?","Haan bhai, par umpire zyada entertaining tha 😂🤦‍♂️"
+Tune naya filter try kiya?,"Haan bhai, asli se zyada sundar lag raha tha 😂📸"
+Tera crush online hai kya?,"Haan bhai, par bas 'last seen' ka ehsaas de rahi hai 💔😂"
+"Bhai, kal movie ka plan hai?","Bhai, 'pockets empty' ka plan hai 😂💸"
+Tu gym join karega kya?,"Bhai, bas 'intentions fit' hai, body nahi 😂💪"
+Tere boss ne kuch poocha?,"Haan bhai, bas HR jaisa dikh raha tha 😂💀"
+Tune ghar ka kaam kiya?,"Bhai, bas dishes aur dreams donon dhoye 😂🍽️😴"
+Tere weekend plans kya hai?,"Bhai, bas neend aur napka date hai 😂🛌"
+Tu kal office gaya tha?,"Haan bhai, par attendance dene gaya tha bas 😂💼"
+Tune ghar ka renovation kiya?,"Bhai, bas sofa ka position badla 😂🛋️"
+Tu kal late kyun aaya?,"Bhai, traffic nahi, neend heavy thi 😂😴"
+Tera net slow hai kya?,"Bhai, pigeon se bhi slow chal raha hai 😂🐦"
+"Bhai, tu cricket khelta hai?","Haan bhai, bas fielding se dushmani hai 😂🏏"
+Tu kal party gaya tha?,"Haan bhai, par free snacks pe zyada dhyan tha 😂🍕"
+"Bhai, tere paas charger hai?","Haan bhai, par battery khatam hai 😂🔋"
+Tune weekend pe kya kiya?,"Bhai, bas bed ke saath commitment nibha raha tha 😂🛏️"
+"Bhai, salary gayi?","Haan bhai, udti chidiya bhi nahi thi, bas ud gayi 😂💸"
+Tune naya meme dekha?,"Haan bhai, do baar share bhi kiya 😂📱"
+"Bhai, tera pet bhag gaya?","Haan bhai, EMI dekh ke 😂🐕‍🦺"
+Tera phone slow hai?,"Bhai, turtle bhi inspire ho jaye 😂🐢"
+"Bhai, tu naya laptop le raha hai?","Haan bhai, EMI lene ka plan hai 😂💻"
+Tune uska tweet dekha?,"Haan bhai, lagta hai usne Elon Musk ko hire kar liya 😂🐦"
+"Bhai, kal late kyun aaya?","Bhai, neend ka heavy dose ho gaya tha 😂😴"
+Tu kal gym gaya tha?,"Haan bhai, par bas locker ka lock khol ke aa gaya 😂🔒"
+"Bhai, tera dukh kya hai?","Bhai, 'low balance' aur 'low battery' ek saath hai 😂💸🔋"
+Tere gharwale strict hai?,"Bhai, 'WiFi password' se bhi zyada 😂🔒"
+Tune naya filter try kiya?,"Haan bhai, asli se zyada fake lag raha tha 😂📸"
+Tera dost cool hai?,"Bhai, uski coolness dekh ke AC bhi sharma jaye 😂❄️"
+Tune ghar ka kaam kiya?,"Bhai, bas 'undo' button dhoond raha hoon 😂⏪"
+Tu kal cricket khelne gaya?,"Haan bhai, par bas fielding se dushmani ho gayi 😂🏏"
+"Bhai, kal kaam pe gaya?","Haan bhai, attendance dene gaya tha bas 😂💼"
+"Bhai, kal ka cricket match dekha?","Haan bhai, bas 'last over' mein dil toota 😂💔🏏"
+Tune naya song suna?,"Haan bhai, repeat pe chal raha hai 😂🎧"
+Tera net slow hai?,"Bhai, pigeon se bhi slow chal raha hai 😂🐦"
+"Bhai, tune naya meme dekha?","Haan bhai, battery khatam kar diya share karte 😂📱"
+"Bhai, tu aaj kal busy rehta hai?","Haan bhai, bas 'workload' se zyada 'overthinkload' hai 😂💀"
+Tune kal gym join kiya?,"Haan bhai, par bas 'selfie membership' li hai 📸😂"
+"Bhai, tu diet pe hai?","Haan bhai, bas 'pizza' aur 'biryani' ko healthy maan liya 😂🍕🍗"
+Tere boss ka mood kaisa hai?,"Bhai, lagta hai unke 'data plan' ka bhi expiry ho gaya 😂📉"
+Tu kal kitne baje soya?,"Bhai, bas 'Netflix' ne raat ka 'delete button' daba diya 😂📺"
+"Bhai, tera crush tujhe bhool gaya?","Haan bhai, par Instagram memories nahi 😂💔📱"
+Tune weekend pe kya kiya?,"Bhai, bas 'bed' se zyada serious relationship mein tha 😂🛏️"
+Tu kal late kyun aaya?,"Bhai, traffic nahi, bas 'mood swing' heavy tha 😂🚦"
+Tune naya song suna?,"Haan bhai, ab toh 'repeat' se bhi dosti ho gayi 😂🎧"
+Tera dukh kya hai?,"Bhai, salary bhi 'fast forward' mode mein chali gayi 😂💸"
+Tu kal cricket khel raha tha?,"Haan bhai, par fielding se 'breakup' ho gaya 😂🏏"
+Tere gharwale strict hai?,"Bhai, WiFi password se bhi zyada 😂🔒"
+Tune naya filter try kiya?,"Haan bhai, asli se zyada 'animated' lag raha tha 😂📸"
+Tu kal date pe gaya tha?,"Haan bhai, par bas 'pocket money' ka breakup ho gaya 😂💸"
+"Bhai, weekend pe kya plan hai?","Bhai, full 'bedflix and nap' mode on hai 😂🛌🍿"
+Tu gym nahi gaya?,"Bhai, bas 'motivation' ne leave le liya 😂💪"
+Tera net slow hai?,"Bhai, turtle bhi 'fast and furious' lag raha hai 😂🐢"
+"Bhai, tu cricket dekh raha hai?","Haan bhai, par umpire zyada entertaining hai 😂👀"
+Tune ghar ka kaam kiya?,"Bhai, bas 'undo button' dhoond raha hoon 😂⏪"
+Tera phone slow hai?,"Bhai, snail bhi jeet jayega race mein 😂🐌"
+Tune naya reel banaya?,"Haan bhai, viral nahi hua, bas family ne dekha 😂📱"
+Tere boss ne kuch bola?,"Haan bhai, bas 'mental gym' karwa rahe the 😂💀"
+"Bhai, weekend pe outing?","Bhai, bas 'ghar se terrace' tak travel hoga 😂🏠"
+Tune naya job apply kiya?,"Haan bhai, par HR ne 'seen' maar diya 😂💼"
+Tu aaj kal busy hai?,"Haan bhai, bas 'meme scrolling' mein busy hoon 😂📱"
+"Bhai, tu kal late kyun aaya?","Bhai, bas 'alarm snooze' mode mein chala gaya 😂⏰"
+Tu naya phone le raha hai?,"Haan bhai, par EMI ka dukh zyada hai 😂📱💸"
+"Bhai, tera laptop slow hai?","Bhai, lagta hai 'Windows 95' par chal raha hai 😂💻"
+Tune ghar ka renovation kiya?,"Bhai, bas 'sofa' ka angle change kiya 😂🛋️"
+"Bhai, kal movie dekhi?","Haan bhai, par bas popcorn ka bill yaad hai 😂🍿💸"
+Tera dukh kya hai?,"Bhai, salary aur battery dono low hai 😂💸🔋"
+Tune naya meme dekha?,"Haan bhai, share karte hi battery gayab 😂📱⚡"
+Tu kal party gaya tha?,"Haan bhai, par free snacks hi target tha 😂🍕"
+Tera boss kaisa hai?,"Bhai, 'Monday' se bhi zyada toxic hai 😂💀"
+Tune ghar ka kaam kiya?,"Bhai, bas 'mind cleaning' kar raha tha 😂🧠"
+"Bhai, kal cricket khela?","Haan bhai, bas ball dhoondhne gaya tha 😂🏏"
+Tu kal late kyun aaya?,"Bhai, sapno mein fast forward ho gaya 😂💤"
+"Bhai, tera net slow hai?","Haan bhai, '2G' bhi racer lag raha hai 😂🐢"
+Tune naya game try kiya?,"Haan bhai, par boss ka 'deadline game' jeet gaya 😂🎮"
+Tu aaj kal busy hai?,"Haan bhai, bas 'reality check' mein busy hoon 😂🔍"
+"Bhai, kal kya kiya?","Bhai, bas 'procrastination' ka world record banaya 😂🏅"
+Tu kal gym gaya tha?,"Haan bhai, bas 'reels' dekhne 😂📱"
+"Bhai, tune naya reel dekha?","Haan bhai, copy paste bhi kar diya 😂🎥"
+Tera dukh kya hai?,"Bhai, salary aayi bhi nahi gayi bhi 😂💸"
+Tu kal cricket khel raha tha?,"Haan bhai, par sirf 'sledging' mein expert ho gaya 😂🏏"
+"Bhai, weekend pe kya plan hai?","Bhai, bas 'ghar se sofa' tak travel hai 😂🛋️"
+Tune kal movie dekhi?,"Haan bhai, bas ending hi nahi samjhi 😂🎬"
+Tera net slow hai?,"Bhai, 'loading' se zyada 'buffering' hai 😂📶"
+"Bhai, tu naya phone le raha hai?","Haan bhai, par 'battery life' ka breakup ho gaya 😂📱"
+Tune ghar ka kaam kiya?,"Bhai, bas 'remote' ka workout kiya 😂📺"
+Tu kal late kyun aaya?,"Bhai, bas 'snooze' pe 'snooze' maar raha tha 😂⏰"
+Tune naya song suna?,"Haan bhai, ab toh 'repeat' pe chal raha hai 😂🎧"
+Tera boss strict hai?,"Bhai, usko toh 'deadline' se bhi zyada 'deadline' ka darr hai 😂💼"
+"Bhai, tu aaj kal busy hai?","Haan bhai, bas 'meme scrolling' mein busy hoon 😂📱"
+"Bhai, kal kaam pe gaya?","Haan bhai, attendance dene gaya tha bas"
+"Bhai, kal cricket match dekha?","Haan bhai, bas 'last over' mein dil toota"
+"Bhai, weekend pe outing?","Bhai, bas 'ghar se terrace' tak travel hoga"
+"Bhai, tu cricket khelta hai?","Haan bhai, bas fielding se dushmani hai"
+"Bhai, kal ka plan?","Bhai, neend ki 'booking' full hai"
+"Bhai, tu kal late kyun aaya?","Bhai, bas 'alarm snooze' mode mein chala gaya"
+"Bhai, kal movie dekhi?","Haan bhai, par bas popcorn ka bill yaad hai"
+"Bhai, tu aaj kal busy rehta hai?","Haan bhai, bas 'workload' se zyada 'overthinkload' hai"
+"Bhai, kal ka cricket match dekha?","Haan bhai, par umpire zyada entertaining tha"
+"Bhai, weekend pe kya plan hai?","Bhai, bas sapno mein Maldives jaane ka plan hai"
+"Bhai, weekend pe kya scene?","Bhai, chill maarne ka full mood hai"
+"Bhai, kal kaam pe gaya?","Haan bhai, bas attendance dene gaya tha"
+"Bhai, kal ka cricket match dekha?","Haan bhai, bas 'last over' mein dil toota"

pages/Chat.py

@@ -0,0 +1,153 @@
+import streamlit as st
+from utils import (
+    load_model, 
+    load_finetuned_model, 
+    generate_response, 
+    get_hf_token
+)
+import os
+import json
+from datetime import datetime
+
+st.set_page_config(page_title="Gemma Chat", layout="wide")
+
+# -------------------------------
+# 💡 Theme Toggle
+# -------------------------------
+dark_mode = st.sidebar.toggle("🌙 Dark Mode", value=False)
+
+if dark_mode:
+    st.markdown(
+        """
+        <style>
+        body { background-color: #1e1e1e; color: #ffffff; }
+        .stTextInput, .stTextArea, .stSelectbox, .stSlider { color: #ffffff !important; }
+        </style>
+        """, unsafe_allow_html=True
+    )
+
+st.title("💬 Chat with Gemma Model")
+
+# -------------------------------
+# 📌 Model Source Selection
+# -------------------------------
+model_source = st.sidebar.radio("📌 Select Model Source", ["Local (.pt)", "Hugging Face"])
+
+# -------------------------------
+# 🔥 Dynamic Model List
+# -------------------------------
+if model_source == "Local (.pt)":
+    model_dir = "models"
+    if not os.path.exists(model_dir):
+        os.makedirs(model_dir)
+
+    local_models = [f for f in os.listdir(model_dir) if f.endswith(".pt")]
+
+    if local_models:
+        selected_model = st.sidebar.selectbox("🛠️ Select Local Model", local_models)
+        model_path = os.path.join(model_dir, selected_model)
+    else:
+        st.warning("⚠️ No fine-tuned models found. Fine-tune a model first.")
+        st.stop()
+
+else:
+    hf_models = [
+        "google/gemma-3-1b-it", 
+        "google/gemma-3-4b-pt", 
+        "google/gemma-3-4b-it",
+        "google/gemma-3-12b-pt", 
+        "google/gemma-3-12b-it", 
+        "google/gemma-3-27b-pt", 
+        "google/gemma-3-27b-it"
+    ]
+    selected_model = st.sidebar.selectbox("🛠️ Select Hugging Face Model", hf_models)
+    model_path = None
+
+# -------------------------------
+# 🔥 Model Loading
+# -------------------------------
+hf_token = get_hf_token()
+
+if model_source == "Local (.pt)":
+    tokenizer, model = load_model("google/gemma-3-1b-it", hf_token)  # Base model first
+    model = load_finetuned_model(model, model_path)
+    if model:
+        st.success(f"✅ Local fine-tuned model loaded: `{selected_model}`")
+    else:
+        st.error("❌ Failed to load local model.")
+        st.stop()
+
+else:
+    tokenizer, model = load_model(selected_model, hf_token)
+    if model:
+        st.success(f"✅ Hugging Face model loaded: `{selected_model}`")
+    else:
+        st.error("❌ Failed to load Hugging Face model.")
+        st.stop()
+
+# -------------------------------
+# ⚙️ Model Configuration Panel
+# -------------------------------
+st.sidebar.header("⚙️ Model Configuration")
+temperature = st.sidebar.slider("🔥 Temperature", 0.1, 1.5, 0.7, 0.1)
+top_p = st.sidebar.slider("🎯 Top-p", 0.1, 1.0, 0.9, 0.1)
+repetition_penalty = st.sidebar.slider("🔁 Repetition Penalty", 0.5, 2.0, 1.0, 0.1)
+
+# -------------------------------
+# 💬 Chat Interface
+# -------------------------------
+if "conversation" not in st.session_state:
+    st.session_state.conversation = []
+
+prompt = st.text_area("💬 Enter your message:", "Hello, how are you?", key="prompt", height=100)
+max_length = st.slider("📏 Max Response Length", min_value=50, max_value=1000, value=300, step=50)
+
+# -------------------------------
+# 🚀 Streaming Response Function
+# -------------------------------
+def stream_response():
+    """
+    Streams the response token by token.
+    """
+    response = generate_response(prompt, model, tokenizer, max_length)
+
+    if response:
+        timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+        st.session_state.conversation.append({"sender": "👤 You", "message": prompt, "timestamp": timestamp})
+        st.session_state.conversation.append({"sender": "🤖 AI", "message": response, "timestamp": timestamp})
+        return response
+    else:
+        st.error("❌ Failed to generate response.")
+        return None
+
+# -------------------------------
+# 🎯 Conversation Controls
+# -------------------------------
+col1, col2, col3 = st.columns([1, 1, 1])
+
+if col1.button("🚀 Generate (CTRL+Enter)", help="Use CTRL + Enter to generate"):
+    stream_response()
+
+if col2.button("🗑️ Clear Conversation"):
+    st.session_state.conversation = []
+
+# Export & Import
+if col3.download_button("📥 Export Chat", json.dumps(st.session_state.conversation, indent=4), "chat_history.json"):
+    st.success("✅ Chat exported successfully!")
+
+uploaded_file = st.file_uploader("📤 Import Conversation", type=["json"])
+
+if uploaded_file is not None:
+    st.session_state.conversation = json.load(uploaded_file)
+    st.success("✅ Conversation imported successfully!")
+
+# -------------------------------
+# 🛠️ Display Conversation
+# -------------------------------
+st.subheader("📜 Conversation History")
+
+for msg in st.session_state.conversation:
+    with st.container():
+        st.markdown(f"**{msg['sender']}**  \n🕒 {msg['timestamp']}")
+        st.write(msg['message'])
+

pages/Conversion.py

@@ -0,0 +1,24 @@
+import streamlit as st
+from utils import load_model, convert_to_torchscript, convert_to_onnx, get_hf_token
+
+st.title("🔧 Model Conversion")
+
+# Load the HF token from utils
+hf_token = get_hf_token()
+
+# Load the model
+model_path = "fine_tuned_model.pt"
+tokenizer, model = load_model("google/gemma-3-1b-it", hf_token, model_path)
+
+conversion_option = st.selectbox("Select Conversion Format", ["TorchScript", "ONNX"])
+
+if st.button("Convert Model"):
+    if conversion_option == "TorchScript":
+        with st.spinner("Converting to TorchScript..."):
+            ts_model = convert_to_torchscript(model)
+            st.success("Model converted to TorchScript!")
+            
+    elif conversion_option == "ONNX":
+        with st.spinner("Converting to ONNX..."):
+            onnx_path = convert_to_onnx(model)
+            st.success("Model converted to ONNX!")

pages/Dataset_Management.py

@@ -0,0 +1,360 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+import os
+from utils import (
+    load_dataset,
+    save_dataset,
+    clean_dataset,
+    compute_dataset_score,
+    detect_outliers,
+    apply_transformation,
+    list_datasets,
+    detect_inconsistent_types
+)
+
+# -------------------------------
+# Constants & Setup
+# -------------------------------
+DATASET_DIR = "datasets"
+DEFAULT_DATASET = "train_data.csv"
+os.makedirs(DATASET_DIR, exist_ok=True)  # Ensure directory exists
+
+# -------------------------------
+# Sidebar: Dataset Selection
+# -------------------------------
+st.sidebar.header("📊 Dataset Selection")
+
+# List available datasets from the datasets folder
+available_datasets = list_datasets(DATASET_DIR)
+dataset_choice = st.sidebar.radio("Choose Dataset Source:", ["Select Existing Dataset", "Upload New Dataset"])
+
+dataset_path = None
+
+if dataset_choice == "Select Existing Dataset":
+    if available_datasets:
+        selected_dataset = st.sidebar.selectbox("Select Dataset:", available_datasets)
+        dataset_path = os.path.join(DATASET_DIR, selected_dataset)
+        st.sidebar.success(f"Using `{selected_dataset}` dataset.")
+    else:
+        st.sidebar.warning("No datasets found. Please upload a new dataset.")
+elif dataset_choice == "Upload New Dataset":
+    uploaded_file = st.sidebar.file_uploader("Upload Dataset (CSV, JSON, or Excel)", type=["csv", "json", "xlsx"])
+    if uploaded_file:
+        file_ext = uploaded_file.name.split('.')[-1].lower()
+        try:
+            if file_ext == "csv":
+                new_df = pd.read_csv(uploaded_file)
+            elif file_ext == "json":
+                new_df = pd.json_normalize(json.load(uploaded_file))
+            elif file_ext == "xlsx":
+                new_df = pd.read_excel(uploaded_file)
+            else:
+                st.error("Unsupported file format.")
+                st.stop()
+        except Exception as e:
+            st.error(f"Error reading file: {e}")
+            st.stop()
+
+        # Save the new dataset with its filename
+        dataset_path = os.path.join(DATASET_DIR, uploaded_file.name)
+        save_dataset(new_df, dataset_path)
+        st.sidebar.success(f"Dataset `{uploaded_file.name}` uploaded successfully!")
+        available_datasets = list_datasets(DATASET_DIR)  # Refresh list
+    else:
+        st.sidebar.warning("Please upload a dataset.")
+
+# -------------------------------
+# Load the Selected Dataset
+# -------------------------------
+if dataset_path:
+    df = load_dataset(dataset_path)
+    if df.empty:
+        st.warning("Dataset is empty or failed to load.")
+else:
+    df = pd.DataFrame()
+    st.warning("No dataset selected. Please choose or upload a dataset.")
+
+# -------------------------------
+# Main App Title & Description
+# -------------------------------
+st.title("📊 The Data Hub")
+
+# -------------------------------
+# Tabs for Operations
+# -------------------------------
+tabs = st.tabs([
+    "View & Summary", "Clean Data",
+    "Visualize Data", "Data Profiling",
+    "Outlier Detection", "Custom Transformations",
+    "Export"
+])
+
+# -------------------------------
+# Tab 1: View & Summary
+# -------------------------------
+with tabs[0]:
+    st.subheader("📋 Current Dataset Preview")
+    if not df.empty:
+        st.dataframe(df.head(10))
+        st.markdown("#### 🔎 Basic Statistics")
+        st.write(df.describe(include="all"))
+    else:
+        st.warning("No dataset available. Please choose or upload a dataset.")
+
+# -------------------------------
+# Tab 2: Clean Data
+# -------------------------------
+with tabs[1]:
+    st.subheader("🧼 Clean Your Dataset")
+    if not df.empty:
+        remove_duplicates = st.checkbox("Remove Duplicate Rows", value=True)
+        fill_missing = st.checkbox("Fill Missing Values", value=False)
+        fill_value = st.text_input("Fill missing values with:", value="0")
+
+        st.markdown("#### Optional: Rename Columns")
+        new_names = {}
+        for col in df.columns:
+            new_names[col] = st.text_input(f"Rename column '{col}'", value=col)
+
+        if st.button("Clean Dataset"):
+            cleaned_df = clean_dataset(df, remove_duplicates, fill_missing, fill_value)
+            cleaned_df = cleaned_df.rename(columns=new_names)
+            save_dataset(cleaned_df, dataset_path)
+            st.success("✅ Dataset cleaned successfully!")
+            st.dataframe(cleaned_df.head())
+            df = cleaned_df
+    else:
+        st.warning("No dataset available for cleaning.")
+
+# -------------------------------
+# Tab 3: Visualize Data (Fixed KeyError Issue)
+# -------------------------------
+with tabs[2]:
+    st.subheader("📊 Visualize Your Data")
+
+    if not df.empty:
+        viz_type = st.selectbox("Select Visualization Type", ["Histogram", "Scatter", "Box Plot", "Heatmap", "Line Chart"])
+        numeric_cols = df.select_dtypes(include=["number"]).columns.tolist()
+
+        if numeric_cols:
+            # Validate column selection
+            col = st.selectbox("Select Column", numeric_cols)
+
+            if col:  # Ensure valid column selection
+                fig, ax = plt.subplots()
+
+                if viz_type == "Histogram":
+                    ax.hist(df[col].dropna(), bins=20, color="skyblue", edgecolor="black")
+                elif viz_type == "Box Plot":
+                    sns.boxplot(x=df[col].dropna(), ax=ax)
+                elif viz_type == "Scatter":
+                    x_col = st.selectbox("X-axis", numeric_cols)
+                    y_col = st.selectbox("Y-axis", numeric_cols)
+                    if x_col and y_col:
+                        ax.scatter(df[x_col], df[y_col], color="green")
+                elif viz_type == "Heatmap":
+                    corr = df[numeric_cols].corr()
+                    sns.heatmap(corr, annot=True, cmap="coolwarm", ax=ax)
+                elif viz_type == "Line Chart":
+                    ax.plot(df.index, df[col], marker="o")
+
+                st.pyplot(fig)
+            else:
+                st.warning("Please select a valid column.")
+        else:
+            st.warning("No numeric columns available for visualization.")
+    else:
+        st.warning("No dataset available for visualization.")
+
+# -------------------------------
+# Tab 4: Data Profiling
+# -------------------------------
+with tabs[3]:
+    if not df.empty:
+
+        # -------------------------------
+        # 1. General Dataset Info
+        # -------------------------------
+        st.markdown("### 🛠️ General Information")
+        st.write(f"✅ **Total Rows:** `{df.shape[0]}`")
+        st.write(f"✅ **Total Columns:** `{df.shape[1]}`")
+        st.write(f"✅ **Memory Usage:** `{df.memory_usage(deep=True).sum() / (1024 ** 2):.2f} MB`")
+        st.write(f"✅ **Dataset Shape:** `{df.shape}`")
+
+        # -------------------------------
+        # 2. Dataset Quality Score
+        # -------------------------------
+        st.markdown("### 📊 Dataset Quality Score")
+        score = compute_dataset_score(df)
+        st.success(f"💯 Dataset Quality Score: `{score} / 100`")
+
+        # -------------------------------
+        # 3. Column Overview with Stats
+        # -------------------------------
+        st.markdown("### 🔥 Column Overview")
+
+        # Numeric and categorical columns
+        numeric_cols = df.select_dtypes(include=["number"]).columns
+        categorical_cols = df.select_dtypes(include=["object"]).columns
+
+        profile = pd.DataFrame({
+            "Column": df.columns,
+            "Data Type": df.dtypes.values,
+            "Missing Values": df.isnull().sum().values,
+            "Missing %": (df.isnull().sum() / len(df) * 100).values,
+            "Unique Values": df.nunique().values
+        })
+
+        # Add numeric statistics
+        if len(numeric_cols) > 0:
+            numeric_stats = pd.DataFrame({
+                "Column": numeric_cols,
+                "Min": df[numeric_cols].min().values,
+                "Max": df[numeric_cols].max().values,
+                "Mean": df[numeric_cols].mean().values,
+                "Std Dev": df[numeric_cols].std().values,
+                "Skewness": df[numeric_cols].skew().values,
+                "Kurtosis": df[numeric_cols].kurt().values
+            })
+
+            # Merge stats with the profile
+            profile = profile.merge(numeric_stats, on="Column", how="left")
+
+        st.dataframe(profile)
+
+        # -------------------------------
+        # 4. Missing Values Visualization
+        # -------------------------------
+        st.markdown("### 🔎 Missing Values Distribution")
+        missing_values = df.isnull().sum()
+        missing_values = missing_values[missing_values > 0]
+
+        if not missing_values.empty:
+            fig, ax = plt.subplots(figsize=(12, 5))
+            sns.barplot(x=missing_values.index, y=missing_values.values, ax=ax, color="skyblue")
+            ax.set_title("Missing Values per Column")
+            ax.set_ylabel("Missing Count")
+            ax.set_xticklabels(ax.get_xticklabels(), rotation=45)
+            st.pyplot(fig)
+        else:
+            st.success("No missing values found!")
+
+        # -------------------------------
+        # 5. Duplicates Detection
+        # -------------------------------
+        st.markdown("### 🔥 Duplicates & Constant Columns Detection")
+        
+        # Duplicates
+        duplicate_count = df.duplicated().sum()
+        st.write(f"🔁 **Duplicate Rows:** `{duplicate_count}`")
+
+        # Constant Columns
+        constant_cols = [col for col in df.columns if df[col].nunique() == 1]
+        if constant_cols:
+            st.write(f"🚩 **Constant Columns:** `{constant_cols}`")
+        else:
+            st.success("No constant columns detected!")
+
+        # -------------------------------
+        # 6. Cardinality Analysis
+        # -------------------------------
+        st.markdown("### 🧬 Cardinality Analysis")
+        
+        high_cardinality = [col for col in df.columns if df[col].nunique() > len(df) * 0.8]
+        if high_cardinality:
+            st.write(f"🔢 **High-Cardinality Columns:** `{high_cardinality}`")
+        else:
+            st.success("No high-cardinality columns detected!")
+
+        # -------------------------------
+        # 7. Top Frequent & Rare Values
+        # -------------------------------
+        st.markdown("### 🎯 Frequent & Rare Values")
+
+        for col in categorical_cols:
+            st.write(f"✅ **{col}**")
+            
+            top_values = df[col].value_counts().nlargest(5)
+            rare_values = df[col].value_counts().nsmallest(5)
+
+            st.write("📊 **Top Frequent Values:**")
+            st.dataframe(top_values)
+
+            st.write("🧪 **Rare Values:**")
+            st.dataframe(rare_values)
+
+        # -------------------------------
+        # 8. Correlation Matrix
+        # -------------------------------
+        st.markdown("### 📊 Correlation Matrix")
+        
+        if len(numeric_cols) > 1:
+            corr = df[numeric_cols].corr()
+
+            fig, ax = plt.subplots(figsize=(12, 8))
+            sns.heatmap(corr, annot=True, fmt=".2f", cmap="coolwarm", square=True, ax=ax)
+            st.pyplot(fig)
+        else:
+            st.info("Not enough numeric columns for correlation analysis.")
+
+        # -------------------------------
+        # 9. Pair Plot (Numerical Relationships)
+        # -------------------------------
+        st.markdown("### 🔥 Pair Plot (Numerical Relationships)")
+        
+        if len(numeric_cols) >= 2:
+            pairplot = sns.pairplot(df[numeric_cols], diag_kind='kde')
+            st.pyplot(pairplot.fig)
+        else:
+            st.info("Not enough numeric columns for pair plot visualization.")
+
+        # -------------------------------
+        # 10. Outlier Detection
+        # -------------------------------
+        st.markdown("### 🚩 Outlier Detection")
+        
+        outliers = detect_outliers(df)
+        if outliers:
+            st.write("✅ **Outliers Detected:**")
+            st.dataframe(pd.DataFrame(outliers.items(), columns=["Column", "Outlier Count"]))
+        else:
+            st.success("No significant outliers detected!")
+
+        # -------------------------------
+        # 11. Inconsistent Data Types
+        # -------------------------------
+        st.markdown("### 🚫 Inconsistent Data Types")
+        
+        inconsistent_types = detect_inconsistent_types(df)
+        if inconsistent_types:
+            st.write("⚠️ **Inconsistent Data Types Detected:**")
+            st.write(inconsistent_types)
+        else:
+            st.success("No inconsistent data types detected!")
+
+    else:
+        st.warning("No dataset available for profiling.")
+
+# -------------------------------
+# Tab 5: Outlier Detection
+# -------------------------------
+with tabs[4]:
+    st.subheader("🚀 Outlier Detection")
+    if not df.empty:
+        outliers = detect_outliers(df)
+        st.write(outliers)
+    else:
+        st.warning("No dataset available for outlier detection.")
+
+# -------------------------------
+# Tab 6: Export
+# -------------------------------
+with tabs[5]:
+    st.subheader("📤 Export Dataset")
+    export_format = st.selectbox("Export Format", ["CSV", "Excel", "JSON"])
+    if not df.empty:
+        st.download_button("Download", df.to_csv(index=False), f"dataset.{export_format.lower()}")
+

pages/Finetune.py

@@ -0,0 +1,170 @@
+import streamlit as st
+import pandas as pd
+import json
+import os
+from datetime import datetime
+from utils import (
+    load_model, 
+    get_hf_token, 
+    simulate_training, 
+    plot_training_metrics, 
+    load_finetuned_model, 
+    save_model
+)
+
+st.title("🔥 Fine-tune the Gemma Model")
+
+# -------------------------------
+# Finetuning Option Selection
+# -------------------------------
+finetune_option = st.radio("Select Finetuning Option", ["Fine-tune from scratch", "Refinetune existing model"])
+
+# -------------------------------
+# Model Selection Logic
+# -------------------------------
+selected_model = None
+saved_model_path = None
+
+if finetune_option == "Fine-tune from scratch":
+    # Display Hugging Face model list
+    model_list = [
+        "google/gemma-3-1b-pt",
+        "google/gemma-3-1b-it", 
+        "google/gemma-3-4b-pt", 
+        "google/gemma-3-4b-it",
+        "google/gemma-3-12b-pt", 
+        "google/gemma-3-12b-it", 
+        "google/gemma-3-27b-pt", 
+        "google/gemma-3-27b-it"
+    ]
+    selected_model = st.selectbox("🛠️ Select Gemma Model to Fine-tune", model_list)
+
+elif finetune_option == "Refinetune existing model":
+    # Dynamically list all saved models from the /models folder
+    model_dir = "models"
+    
+    if os.path.exists(model_dir):
+        saved_models = [f for f in os.listdir(model_dir) if f.endswith(".pt")]
+    else:
+        saved_models = []
+
+    if saved_models:
+        saved_model_path = st.selectbox("Select a saved model to re-finetune", saved_models)
+        saved_model_path = os.path.join(model_dir, saved_model_path)
+        st.success(f"✅ Selected model for refinement: `{saved_model_path}`")
+    else:
+        st.warning("⚠️ No saved models found! Switching to fine-tuning from scratch.")
+        finetune_option = "Fine-tune from scratch"
+
+# -------------------------------
+# Dataset Selection
+# -------------------------------
+
+st.subheader("📚 Dataset Selection")
+
+# Dataset source selection
+dataset_option = st.radio("Choose dataset:", ["Upload New Dataset", "Use Existing Dataset (`train_data.csv`)"])
+
+dataset_path = "train_data.csv"
+
+if dataset_option == "Upload New Dataset":
+    uploaded_file = st.file_uploader("📤 Upload Dataset (CSV or JSON)", type=["csv", "json"])
+
+    if uploaded_file is not None:
+        # Handle CSV or JSON upload
+        if uploaded_file.name.endswith(".csv"):
+            new_data = pd.read_csv(uploaded_file)
+        elif uploaded_file.name.endswith(".json"):
+            json_data = json.load(uploaded_file)
+            new_data = pd.json_normalize(json_data)
+        else:
+            st.error("❌ Unsupported file format. Please upload CSV or JSON.")
+            st.stop()
+
+        # Append or create new dataset
+        if os.path.exists(dataset_path):
+            new_data.to_csv(dataset_path, mode='a', index=False, header=False)
+            st.success(f"✅ Data appended to `{dataset_path}`!")
+        else:
+            new_data.to_csv(dataset_path, index=False)
+            st.success(f"✅ Dataset saved as `{dataset_path}`!")
+
+elif dataset_option == "Use Existing Dataset (`train_data.csv`)":
+    if os.path.exists(dataset_path):
+        st.success("✅ Using existing `train_data.csv` for fine-tuning.")
+    else:
+        st.error("❌ `train_data.csv` not found! Please upload a new dataset.")
+        st.stop()
+
+# -------------------------------
+# Hyperparameters Configuration
+# -------------------------------
+learning_rate = st.number_input("📊 Learning Rate", value=1e-4, format="%.5f")
+batch_size = st.number_input("🛠️ Batch Size", value=16, step=1)
+epochs = st.number_input("⏱️ Epochs", value=3, step=1)
+
+# -------------------------------
+# Fine-tuning Execution
+# -------------------------------
+if st.button("🚀 Start Fine-tuning"):
+    st.info(f"Fine-tuning process initiated...")
+
+    # Retrieve Hugging Face Token
+    hf_token = get_hf_token()
+
+    # Model loading logic
+    if finetune_option == "Refinetune existing model" and saved_model_path:
+        # Load the base model first
+        tokenizer, model = load_model("google/gemma-3-1b-it", hf_token)
+
+        # Load the saved model checkpoint for re-finetuning
+        model = load_finetuned_model(model, saved_model_path)
+
+        if model:
+            st.success(f"✅ Loaded saved model: `{saved_model_path}` for refinement!")
+        else:
+            st.error("❌ Failed to load the saved model. Aborting.")
+            st.stop()
+
+    else:
+        # Fine-tune from scratch (load base model)
+        if not selected_model:
+            st.error("❌ Please select a model to fine-tune.")
+            st.stop()
+
+        tokenizer, model = load_model(selected_model, hf_token)
+        
+        if model:
+            st.success(f"✅ Base model loaded: `{selected_model}`")
+        else:
+            st.error("❌ Failed to load the base model. Aborting.")
+            st.stop()
+
+    # Simulate fine-tuning loop
+    progress_bar = st.progress(0)
+    training_placeholder = st.empty()
+
+    for epoch, losses, accs in simulate_training(epochs):
+        fig = plot_training_metrics(epoch, losses, accs)
+        training_placeholder.pyplot(fig)
+        progress_bar.progress(epoch / epochs)
+
+    # Save fine-tuned model with timestamp
+    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
+    new_model_name = f"models/fine_tuned_model_{selected_model.replace('/', '_')}_{timestamp}.pt"
+    
+    # Save the fine-tuned model
+    saved_model_path = save_model(model, new_model_name)
+
+    if saved_model_path:
+        st.success(f"✅ Fine-tuning completed! Model saved as `{saved_model_path}`")
+
+        # Load the fine-tuned model for immediate inference
+        model = load_finetuned_model(model, saved_model_path)
+
+        if model:
+            st.success("🛠️ Fine-tuned model loaded and ready for inference!")
+        else:
+            st.error("❌ Failed to load the fine-tuned model for inference.")
+    else:
+        st.error("❌ Failed to save the fine-tuned model.")

requirements.txt

@@ -3,4 +3,10 @@ pandas==2.2.3
 numpy==2.2.4
 torch==2.6.0
 transformers @ git+https://github.com/huggingface/[email protected]
-matplotlib==3.10.1
+matplotlib==3.10.1
+rich>=13.1.0
+FuzzyTM>=0.4.0
+requests>=2.28.0
+xlsxwriter>=3.0.1
+python-dotenv>=0.19.0
+scipy>=1.7.3

utils.py

@@ -0,0 +1,468 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import matplotlib.pyplot as plt
+import time
+import json
+import re
+import os
+import asyncio
+from dotenv import load_dotenv
+from scipy.stats import skew, kurtosis, zscore
+
+# -------------------------------
+# Environment and Token Management
+# -------------------------------
+
+# Load environment variables from .env file in local development
+load_dotenv()
+
+def get_hf_token():
+    """
+    Retrieves HF token from secrets or .env file.
+    """
+    token = os.getenv("HF_TOKEN")  # Prioritize environment variable
+
+    # If not found, fallback to Streamlit secrets
+    if not token:
+        try:
+            token = st.secrets["HF_TOKEN"]
+        except (FileNotFoundError, KeyError):
+            st.error("❌ HF_TOKEN not found. Add it to .env or secrets.toml.")
+            return None
+
+    return token
+
+
+# -------------------------------
+# Model Loading and Management
+# -------------------------------
+
+async def async_load(model_id: str):
+    """
+    Dummy async function to initialize the event loop.
+    """
+    await asyncio.sleep(0.1)
+
+@st.cache_resource
+def load_model(model_id: str, token: str, checkpoint_path: str = None):
+    """
+    Loads and caches the Gemma model and tokenizer with the Hugging Face token.
+    
+    Args:
+        model_id (str): The Hugging Face model ID.
+        token (str): The authentication token.
+        checkpoint_path (str): Optional path to a fine-tuned model checkpoint.
+
+    Returns:
+        tuple: tokenizer, model
+    """
+    try:
+        asyncio.run(async_load(model_id))
+
+        tokenizer = AutoTokenizer.from_pretrained(model_id, token=token)
+        model = AutoModelForCausalLM.from_pretrained(model_id, token=token)
+
+        # Load fine-tuned checkpoint if provided
+        if checkpoint_path and os.path.exists(checkpoint_path):
+            model.load_state_dict(torch.load(checkpoint_path, map_location=torch.device('cpu')))
+            model.eval()
+            st.success("✅ Fine-tuned model loaded successfully!")
+
+        return tokenizer, model
+
+    except Exception as e:
+        st.error(f"❌ Model loading failed: {e}")
+        return None, None
+
+
+# -------------------------------
+# Model Saving Function
+# -------------------------------
+
+def save_model(model, model_name: str):
+    """
+    Saves the fine-tuned model to the specified path.
+
+    Args:
+        model (torch.nn.Module): The PyTorch model instance.
+        model_name (str): The file path to save the model.
+
+    Returns:
+        str: The path where the model is saved.
+    """
+    try:
+        # Ensure the models directory exists
+        os.makedirs(os.path.dirname(model_name), exist_ok=True)
+        
+        # Save the model
+        torch.save(model.state_dict(), model_name)
+        st.success(f"✅ Model saved successfully at `{model_name}`")
+        return model_name
+    except Exception as e:
+        st.error(f"❌ Failed to save model: {e}")
+        return None
+
+
+# -------------------------------
+# File Processing and Cleaning
+# -------------------------------
+
+def preprocess_data(uploaded_file, file_extension):
+    """
+    Reads the uploaded file and returns a processed version.
+    Supports CSV, JSONL, and TXT.
+    """
+    try:
+        if file_extension == "csv":
+            return pd.read_csv(uploaded_file)
+
+        elif file_extension == "jsonl":
+            data = [json.loads(line) for line in uploaded_file.readlines()]
+            try:
+                return pd.DataFrame(data)
+            except Exception:
+                st.warning("⚠️ Unable to convert JSONL to table. Previewing raw JSON.")
+                return data
+
+        elif file_extension == "txt":
+            text_data = uploaded_file.read().decode("utf-8")
+            return text_data.splitlines()
+
+    except Exception as e:
+        st.error(f"❌ Error processing file: {e}")
+        return None
+
+
+def clean_text(text, lowercase=True, remove_punctuation=True):
+    """
+    Cleans text data by applying basic normalization.
+    """
+    if lowercase:
+        text = text.lower()
+    if remove_punctuation:
+        text = re.sub(r'[^\w\s]', '', text)
+    return text
+
+
+# -------------------------------
+# Model Conversion and Quantization
+# -------------------------------
+
+def quantize_model(model):
+    """
+    Applies dynamic quantization.
+    """
+    try:
+        quantized_model = torch.quantization.quantize_dynamic(
+            model, {torch.nn.Linear}, dtype=torch.qint8
+        )
+        st.success("✅ Model quantized successfully!")
+        return quantized_model
+    except Exception as e:
+        st.error(f"❌ Quantization failed: {e}")
+        return model
+
+
+def convert_to_torchscript(model, output_path="model_ts.pt"):
+    """
+    Converts the model to TorchScript format.
+    """
+    try:
+        example_input = torch.randint(0, 100, (1, 10))
+        traced_model = torch.jit.trace(model, example_input)
+        traced_model.save(output_path)
+        return output_path
+    except Exception as e:
+        st.error(f"❌ TorchScript conversion failed: {e}")
+        return None
+
+
+def convert_to_onnx(model, output_path="model.onnx"):
+    """
+    Converts the model to ONNX format.
+    """
+    try:
+        dummy_input = torch.randint(0, 100, (1, 10))
+        torch.onnx.export(model, dummy_input, output_path, input_names=["input"], output_names=["output"])
+        return output_path
+    except Exception as e:
+        st.error(f"❌ ONNX conversion failed: {e}")
+        return None
+
+
+# -------------------------------
+# Model Inference and Training
+# -------------------------------
+
+def simulate_training(num_epochs):
+    """
+    Simulates a training loop for demonstration.
+    Yields current epoch, loss values, and accuracy values.
+    """
+    loss_values = []
+    accuracy_values = []
+    for epoch in range(1, num_epochs + 1):
+        loss = np.exp(-epoch) + np.random.random() * 0.1
+        acc = 0.5 + (epoch / num_epochs) * 0.5 + np.random.random() * 0.05
+        loss_values.append(loss)
+        accuracy_values.append(acc)
+        yield epoch, loss_values, accuracy_values
+        time.sleep(1)
+
+
+def plot_training_metrics(epochs, loss_values, accuracy_values):
+    """
+    Plots training loss and accuracy.
+    """
+    fig, ax = plt.subplots(1, 2, figsize=(12, 4))
+    ax[0].plot(range(1, epochs+1), loss_values, marker='o', color='red')
+    ax[0].set_title("Training Loss")
+    ax[0].set_xlabel("Epoch")
+    ax[0].set_ylabel("Loss")
+
+    ax[1].plot(range(1, epochs+1), accuracy_values, marker='o', color='green')
+    ax[1].set_title("Training Accuracy")
+    ax[1].set_xlabel("Epoch")
+    ax[1].set_ylabel("Accuracy")
+
+    return fig
+
+
+def generate_response(prompt, model, tokenizer, max_length=200):
+    """
+    Generates a response using the fine-tuned model.
+    """
+    try:
+        inputs = tokenizer(prompt, return_tensors="pt").input_ids
+
+        with torch.no_grad():
+            outputs = model.generate(inputs, max_length=max_length, num_return_sequences=1, temperature=0.7)
+
+        return tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+    except Exception as e:
+        st.error(f"❌ Response generation failed: {e}")
+        return ""
+
+
+# -------------------------------
+# Model Loading for Inference
+# -------------------------------
+
+def load_finetuned_model(model, checkpoint_path="fine_tuned_model.pt"):
+    """
+    Loads a fine-tuned model from a checkpoint.
+    """
+    if os.path.exists(checkpoint_path):
+        model.load_state_dict(torch.load(checkpoint_path, map_location=torch.device('cpu')))
+        model.eval()
+        st.success("✅ Fine-tuned model loaded successfully!")
+    else:
+        st.error(f"❌ Checkpoint not found: {checkpoint_path}")
+    return model
+
+
+
+import pandas as pd
+import os
+import pyarrow as pa
+import numpy as np
+from scipy.stats import zscore, kurtosis, skew
+
+
+# ======================================
+# Dataset Operations
+# ======================================
+def load_dataset(path: str) -> pd.DataFrame:
+    """Load dataset from CSV with error handling."""
+    try:
+        df = pd.read_csv(path)
+        return make_arrow_compatible(df)
+    except Exception as e:
+        print(f"Error loading dataset: {e}")
+        return pd.DataFrame()
+
+
+def save_dataset(df: pd.DataFrame, path: str):
+    """Save dataset to CSV with error handling."""
+    try:
+        df.to_csv(path, index=False)
+    except Exception as e:
+        print(f"Error saving dataset: {e}")
+
+
+def list_datasets(directory: str = "datasets") -> list:
+    """List all available datasets in the directory."""
+    try:
+        return [f for f in os.listdir(directory) if f.endswith(('.csv', '.json', '.xlsx'))]
+    except Exception as e:
+        print(f"Error listing datasets: {e}")
+        return []
+
+
+# ======================================
+# Data Cleaning Functions
+# ======================================
+def clean_dataset(
+    df: pd.DataFrame,
+    remove_duplicates: bool = True,
+    fill_missing: bool = False,
+    fill_value: str = "0",
+    trim_spaces: bool = True
+) -> pd.DataFrame:
+    """
+    Clean the dataset with multiple operations:
+    - Remove duplicates
+    - Fill missing values
+    - Trim spaces
+    - Remove empty columns and rows
+    - Auto-cast date columns
+    """
+    # Remove duplicates
+    if remove_duplicates:
+        df = df.drop_duplicates()
+
+    # Fill missing values
+    if fill_missing:
+        df = df.fillna(fill_value)
+
+    # Trim spaces
+    if trim_spaces:
+        df = df.apply(lambda x: x.str.strip() if x.dtype == "object" else x)
+
+    # Remove empty columns & rows
+    df = df.dropna(how="all", axis=1)
+    df = df.dropna(how="all", axis=0)
+
+    # Auto-cast date columns
+    for col in df.columns:
+        try:
+            df[col] = pd.to_datetime(df[col])
+        except (ValueError, TypeError):
+            pass
+
+    return make_arrow_compatible(df)
+
+
+# --------------------------------------
+# Dataset Quality Score
+# --------------------------------------
+def compute_dataset_score(df):
+    """Compute dataset quality score."""
+    if df.empty:
+        return 0.0
+
+    total_cells = np.product(df.shape)
+    missing_cells = df.isnull().sum().sum()
+    missing_ratio = missing_cells / total_cells
+
+    duplicate_ratio = 1 - (df.drop_duplicates().shape[0] / df.shape[0])
+
+    numeric_cols = df.select_dtypes(include=["number"]).columns
+    if len(numeric_cols) > 0:
+        skew_vals = df[numeric_cols].apply(lambda x: np.abs(skew(x.dropna())), axis=0)
+        kurt_vals = df[numeric_cols].apply(lambda x: np.abs(kurtosis(x.dropna())), axis=0)
+        numeric_score = 1 - (skew_vals.mean() + kurt_vals.mean()) / 10
+    else:
+        numeric_score = 1
+
+    score = (1 - missing_ratio) * (1 - duplicate_ratio) * numeric_score * 100
+    return round(score, 2)
+
+
+# --------------------------------------
+# Outlier Detection
+# --------------------------------------
+def detect_outliers(df, threshold=3):
+    """Detect outliers in numeric columns using Z-score."""
+    numeric_cols = df.select_dtypes(include=["number"]).columns
+    outliers = {}
+    for col in numeric_cols:
+        z_scores = np.abs(zscore(df[col].dropna()))
+        outliers[col] = np.sum(z_scores > threshold)
+    return outliers
+
+
+# --------------------------------------
+# Detect Inconsistent Types
+# --------------------------------------
+def detect_inconsistent_types(df):
+    """Detect inconsistent data types across columns."""
+    inconsistent_cols = {}
+    for col in df.columns:
+        if df[col].apply(type).nunique() > 1:
+            inconsistent_cols[col] = df[col].apply(type).value_counts().to_dict()
+    return inconsistent_cols
+
+    
+# ======================================
+# Data Transformations
+# ======================================
+def apply_transformation(df: pd.DataFrame, col: str, transform: str) -> pd.DataFrame:
+    """
+    Apply transformations to a specified column:
+    - Log Transformation
+    - Min-Max Normalization
+    - Z-score Standardization
+    """
+    if col not in df.columns:
+        raise KeyError(f"Column '{col}' not found in dataset")
+
+    if transform == "Log":
+        df[col] = np.log1p(df[col].replace(0, np.nan)).fillna(0)
+
+    elif transform == "Normalize":
+        df[col] = (df[col] - df[col].min()) / (df[col].max() - df[col].min())
+
+    elif transform == "Standardize":
+        df[col] = (df[col] - df[col].mean()) / df[col].std()
+
+    return make_arrow_compatible(df)
+
+
+# ======================================
+# Normalization & Standardization
+# ======================================
+def normalize_column(df: pd.DataFrame, col: str) -> pd.DataFrame:
+    """Normalize column (Min-Max Scaling)."""
+    df[col] = (df[col] - df[col].min()) / (df[col].max() - df[col].min())
+    return df
+
+
+def standardize_column(df: pd.DataFrame, col: str) -> pd.DataFrame:
+    """Standardize column (Z-score)."""
+    df[col] = (df[col] - df[col].mean()) / df[col].std()
+    return df
+
+
+# ======================================
+# Arrow Compatibility & Fixes
+# ======================================
+def make_arrow_compatible(df: pd.DataFrame) -> pd.DataFrame:
+    """
+    Ensure dataset compatibility with Streamlit Arrow serialization.
+    """
+    for col in df.columns:
+        if df[col].dtype == object:
+            try:
+                df[col] = df[col].astype(str)
+            except Exception as e:
+                print(f"Could not convert column {col}: {e}")
+    return df
+
+
+def fix_arrow_incompatibility(df: pd.DataFrame) -> pd.DataFrame:
+    """
+    Fix Arrow incompatibility by converting mixed types to `str`.
+    """
+    for col in df.columns:
+        try:
+            pa.Table.from_pandas(df[[col]])
+        except pa.lib.ArrowInvalid:
+            print(f"Arrow compatibility issue in column: {col}")
+            df[col] = df[col].astype(str)
+    return df
+