🎉 Meme Stock Insights: Modern Dashboard 🚀💎🙌

Streamlit app for exploring meme stock data with interactive visualizations and in-depth model performance analysis.

---

🌟 Overview & Purpose ✨

Meme Stock Insights is a powerful, refactored Streamlit application designed for in-depth analysis of meme stock data, leveraging data preparation and model training performed in Google Colab.

The app provides a wonderful, modern, multi-tab experience to explore datasets, visualize key trends using interactive Plotly charts, browse associated memes, and gain critical model performance insights.

---

🚀 Key Features of the Modern App 📊

Section	Feature Highlight	Description
📂 Datasets	Organized Data Views	Access raw data, mapped features, trading signals, and the pickled model contents via a clean tab interface (st.tabs).
📊 Visualizations	Interactive Plotly Charts	Explore data distributions (sentiment, price change) and time series trends using dynamic, zoomable charts for superior user experience.
🧠 Model Insights	Comprehensive Performance	View key metrics (Accuracy, F1, ROC AUC), Top Feature Importance, and a Plotly-rendered Confusion Matrix and ROC Curve for classification evaluation.
🖼️ Meme Gallery	Associated Imagery	Browse stock-related images used in the analysis via a clean, collapsible gallery, powered by Streamlit's st.expander.
⬆️ Upload & Explore	Custom Data Upload	Upload your own CSV files for quick inspection of descriptive statistics and column information.

---

💡 Technical Deep Dive 🧠

This application employs several modern Streamlit and Python techniques to ensure a seamless, high-performance experience:

• Interactive Plotting: Instead of static matplotlib images, we use the Plotly Express library (plotly.express as px) to generate all visualizations. This provides users with built-in zoom, pan, hover, and data-point visibility toggles instantly, enhancing data exploration.
• Robust Data Loading: We use the @st.cache_resource decorator to efficiently download large .pkl (model data) and .zip (image assets) files from Google Drive using the gdown library. This ensures the files are only downloaded once, drastically improving load times on subsequent runs.
• Model Probability Handling (The Fix): The "Model Insights" section includes crucial logic to robustly handle the y_proba array saved from the model. It automatically checks if the array is saved as a one-column (positive class probability only) or two-column (all class probabilities) format to prevent the common IndexError during the calculation of the ROC Curve and AUC.
• Modern UI Components: The application utilizes native Streamlit components like st.set_page_config(layout="wide"), st.tabs, and st.metric to create a clean, modern dashboard aesthetic with organized, segmented content.

📂 Folder Structure

meme-stock-insights/
├── app.py                      # 🌟 Main Streamlit dashboard
├── requirements.txt            # 📦 Dependencies (plotly, gdown, scikit-learn, etc.)
├── data/                       # 💾 All required dataset files
│   ├── mapped_meme_data.csv                # Cleaned & mapped meme data
│   ├── processed_meme_data.pkl             # Pickled ML components (model, y_test, y_proba…)
│   ├── meme_dataset_with_images.csv        # Dataset linking memes to image URLs/files
│   └── tradeable_signals.csv               # Generated trading signals
├── meme_images/                # 🎨 Folder for downloaded meme images
│   └── (e.g., GME_meme.png)
└── README.md                   # 📖 Project guide

🛠️ Setup & Local Installation ⚡

🎯 Prerequisites

• 🐍 Python 3.11 or later
• Data Files: Ensure your Google Drive IDs for the image zip and processed_meme_data.pkl are correct within app.py and publicly accessible.

💻 Installation Steps

1. Clone the repository:

   git clone [https://github.com/Lihini0202/meme-stock-insights.git](https://github.com/Lihini0202/meme-stock-insights.git)
   cd meme-stock-insights

2. Install all required dependencies (including Plotly and scikit-learn):

   pip install -r requirements.txt

3. Run the application:

   streamlit run app.py

☁️ Deployment on Streamlit Cloud 🌍

Deploying Meme Stock Insights is simple and fast:

1. Commit & Push Your Code Make sure all updates — especially app.py and requirements.txt — are pushed to your GitHub repository.

2. Connect Your Repo on Streamlit Cloud Go to https://share.streamlit.io, log in, and select your GitHub repository for deployment.

3. Automatic Environment Setup Streamlit Cloud will automatically:

   • Install dependencies from requirements.txt
   • Download meme images and dataset files using Google Drive File IDs in app.py
   • Build and launch the dashboard

4. Deploy & Share Once the build completes, Streamlit provides a public shareable link to your live dashboard.

---

🙌 Acknowledgments

Built with:

• ❤️ Streamlit for building a clean, modern dashboard UI
• 📊 Plotly for interactive, high-quality visualizations
• 🤖 Google Colab + scikit-learn for model training and data preparation
• 🚀 Inspired by the exciting, unpredictable world of meme stocks