How meet-vasita/cryptocurrency-price-prediction-using-lstm works
Main Components
• Model & Prediction Engine:
– File: Latest_bit_coin_model.keras holds the pre-trained Keras model.
– Code files: main.py and main.ipynb indicate where data processing, model prediction, and potential training/experimentation take place.
• Data Visualization & Static Assets:
– The static directory includes numerous HTML files showing candlestick, line, grouped bar, and other types of plots for various cryptocurrencies.
– Additional assets like script.js and style.css are used for interactive behavior and styling.
– There’s also a CSV file (e.g., btc-usd_stock_data.csv) that might be used to feed data into these visualizations.
• Web Presentation / UI Layer:
– The views folder contains multiple HTML templates (e.g., home, cryptocurrency, liveprices, marketnews, result) used for the user interface.
– These pages likely combine dynamic data (from backend predictions and visualizations) with static text/images.
Relationships and Interactions
• Data Flow:
– Data (stock/cryptocurrency pricing data) is either pre-loaded (CSV in static folder) or fetched from an external source by the prediction engine.
– The prediction engine (in main.py/main.ipynb) loads the pre-trained LSTM model, processes the data, and outputs predictions.
– The results of these predictions are visualized via automatically generated HTML plots (in static folder).
• Web Interface Integration:
– The main.py file could be serving as the application’s controller layer (using a web framework such as Flask or Django) to render views.
– The views use static assets to present data dynamically and might include endpoints that trigger predictions or refresh live data.
• Architectural Pattern:
– The project uses a form of MVC (Model-View-Controller) architecture:
▪ Model: The LSTM model and prediction engine.
▪ View: HTML templates in the views folder and static visualization files.
▪ Controller: The main.py file orchestrates the interaction between the model and view.
Important Architectural Patterns & Principles
• Separation of Concerns: By clearly dividing prediction logic (model), static visual assets (charts, CSV), and HTML templates (views), the project maintains well-separated responsibilities.
Relevant Technologies / Frameworks
• Python (for main.py and main.ipynb)
• Keras/TensorFlow (for the LSTM model stored as Latest_bit_coin_model.keras)
• HTML, CSS, and JavaScript (for static visualization pages and UI)
• Potential web frameworks (Flask or Django) inferred by the structure of the views folder.