The official implementation of the paper "CryptoPulse: Short-Term Cryptocurrency Forecasting with Dual-Prediction and Cross-Correlated Market Indicators". Portions of the code have been adapted from the DLinear implementation.
- Short-Term Forecasting: Predict cryptocurrency prices with a model designed for short-term predictions.
- Batch Processing: Run predictions for multiple cryptocurrencies at once.
- Flexible & Easy to Use: Just install, run, and watch the results roll in!
You can install the package directly from GitHub with pip
:
pip install git+https://github.com/Sequential-Analytics-Lab/SAL-CryptoPulse.git
Alternatively, you can install it manually:
-
Clone the repo:
git clone https://github.com/Sequential-Analytics-Lab/SAL-CryptoPulse.git cd SAL-Cryptopulse
-
Install dependencies in a virtual environment:
python -m venv env_crypto env_crypto\Scripts\activate pip install -e .
After installation, you can run the model directly using the package's command line interface if you had installed the package directly from GitHub with `pip :
For a single run:
cryptopulse --data <crypto-ticker-symbol> --train-epochs 10 --batch-size 32
For batch processing (multiple cryptos at once):
cryptopulse_batch
Alternatively, you can run the model using the Python command if you have cloned the repo in your local
For a single run:
python -m cryptopulse.main --data BTC-USD --train-epochs 10 --batch-size 32
For batch processing (multiple cryptos at once):
python -m cryptopulse.batch_processor
All results are saved in the results/cryptopulse_results directory.
Feel free to use this research work. Let’s build a smarter CryptoPulse together! If you have any questions, or suggestions, or want to collaborate on future developments, don’t hesitate to reach out.
You can connect with me on LinkedIn: Amit - https://www.linkedin.com/in/aamit-datascientist/
Update once I will get the citation details @article{salcryptopulse2024, author = {Amit Kumar and Dr. Taoran Ji}, title = {CryptoPulse: Short-Term Cryptocurrency Forecasting with Dual-Prediction and Cross-Correlated Market Indicators}, journal = {Conference of IEE BigData 2024}, year = {2024}, volume = {5}, pages = {}, }