Solana Price Data Analysis Repository

This repository contains a comprehensive analysis of Solana price data collected from 01-01-2021 to 29-09-2024 in OHLCV format. The project demonstrates the process of fetching raw data, adding technical indicators, performing data cleaning, feature engineering, and preprocessing to ultimately generate an enhanced dataset with 44 columns.

---

Repository Structure

• Datasets:

  • solana_price_dataset.csv
    The original OHLCV dataset containing basic open, high, low, close, and volume data.
  • solana_price_dataset_with_indicators.csv
    The dataset after technical indicators have been added using the ta library.

• Notebooks:

  • add_indicators.ipynb
    Jupyter Notebook for fetching the OHLCV data and adding technical indicators.
  • data_analysis.ipynb
    Jupyter Notebook for performing data cleaning, feature engineering, and preprocessing.
  • crypto-prediction.ipynb
    Notebook for training a model to predict Solana price based on historical data.
  • Cypto_data_processing_v3.ipynb
    Complete end-to-end data pipeline: ingestion, cleaning, merging, and feature engineering for Solana data.
  • live-dashboard.py
    Python script for a real-time Solana price dashboard using Binance API.

• Figures:

  • ./images/solana_price_chart.png
    A sample visualization of the Solana price data with added indicators.
  • ./images/live_dashboard.png
    Screenshot of the real-time Solana price dashboard using Binance API.
  • ./images/terminal_output.png
    Terminal output showcasing the successful execution of data pipeline scripts.
  • ./images/solana_prediction_vs_actual.png
    Visualization comparing actual vs predicted Solana prices, with MAE, RMSE, and MAPE metrics.

---

Notebooks Overview

add_indicators.ipynb

• Purpose:
  This notebook demonstrates how to:

  • Fetch Solana price data in OHLCV format.
  • Add various technical indicators using the ta library.
  • Save the resulting dataset as solana_price_dataset_with_indicators.csv.

• Key Library:
  ta

data_analysis.ipynb

• Purpose:
  This notebook is used to:
  • Clean the dataset.
  • Perform feature engineering.
  • Execute preprocessing steps in an orderly manner.
  • Produce the final, enhanced dataset with 44 columns.

Cypto_data_processing_v3.ipynb

• Purpose:
  This notebook presents an end-to-end data pipeline for Solana price data:

  • Ingest raw data.
  • Clean, merge, and transform it.
  • Apply feature engineering to prepare for modeling.

• Output:
  Cleaned and feature-rich dataset ready for model training.

crypto-prediction.ipynb

• Purpose:
  This notebook is designed to:

  • Train a machine learning model for Solana price prediction.
  • Evaluate model performance using test data.

• Modeling Techniques:
  Includes time-series model training steps and evaluation metrics.

live-dashboard.py

• Purpose:
  This Python script runs a real-time dashboard:

  • Fetches live Solana price data using the Binance API.
  • Displays price movement and key metrics dynamically.

• Tech Stack:
  Built using streamlit for interactive visualization.

---

Getting Started

Clone the Repository

git clone https://github.com/your_username/solana-price-data-analysis.git

go to project directory

cd solana-price-data-analysis

create a API key file in .streamlit directory

mkdir .streamlit/secrets.toml

In .streamlit/secrets.toml file

API_KEY = "put_your_api_key_here"
API_SECRET = "put_your_api_secret_here"

Run the File

streamlit run live-dashboard.py

Screenshots

1. Solana Price Chart

2. Key_Features_Visualisation

3. Z-Score_Outlier_Detection

4. Feature_Importance_Plotting

5. solana_prediction_vs_actual_Plotting

6. terminal_output

7. live_dashboard