Public repository for Sapient's Cardano Native Tokens data portal. Fungible Tokens at launch: AGIX, BOOK (now: STUFF), COPI, EMP, FLDT, GENS, HUNT, IAG, INDY, LENFI, MELD, MILK, MIN, NEWM, NMKR, NTX, SNEK, SUNDAE, WMT, WRT. Note that MELD token are no longer updated or included on the website, as the project shut down operations.
The Native Token Dataportal is now live with step-by-step introduction to open source Python backtesting and plotting, strategy execution, risk management, clean token data and API. Please watch the video below how to get the most out of the service. There is also a landing website and feedback channels via X, Telegram and Google forms. We have opened Github Discussions and will deal with submitted issues and PRs prompty. Looking forward to your feedback and contribution!
Simple SMA20 backtest for WMT token since first DEX listing used to showcase use of open source Python libraries and data API to build from simple trading strategies to more complex ones.
Optional: set up virtual environment to manage dependencies and keep separate from your own projects
python3 -m venv backtesting
source backtesting/bin/activate
To use all the files of this repo in once place (recommended):
git clone https://github.com/Sapient-Predictive-Analytics/dataportal.git
pip install requirements.txt
(1) If you are new to Python data analysis, backtesting, trading or native tokens please start with our simple World Mobile Token Casestudy.
(2) Next or if you have some background, proceed to the Backtesting section to get started with Backtrader and Python plotting.
(3) Refer to our section on the Heatmap feature next for advanced plotting and optimization of trading strategies.
(4) More feature-rich or complex backtests are dealt with in the Advanced Backtesting section.
(5) This is continued in the Advanced Heatmap optimization study.
(6) Don't miss the chapter on Risk Management before you consider actually trading native tokens!
(7) Learn about data cleaning and outlier detection here.
(8) To trade, discuss with likeminded people or test DEXes on Testnet, visit the Community section about trading venues, their communities and Dataportal collaboration
(9) Finally, download or ingest native tokens data using our API
-
Get in touch, share ideas, submit feedback, submit an issue, reach out.
-
This website and repo are [NOT FINANCIAL ADVICE](https://github.com/Sapient-Predictive-Analytics/dataportal/blob/main/disclaimer.md)!
Any feedback, request for new features, tickers, custom signals or plots can be submitted via our Google Form
Please contact us or contribute to this repository if you have ideas, requests or directions how to grow quantitative trading and better market making and risk management for Cardano's native tokens investable assets universe.
Notes on data formats, packages and utility
How the first 10 Native Tokens to be featured were chosen
Simple data manipulation on clean WMT (Minswap) data
Download folder for interactive notebook
Download folder for interactive IPython notebook and time series data.
Introduction to backtesting concepts, the Zipline package, trading strategy design with heatmaps and use of Seaborn and other plotting libraries. New token prices have been added and charts, backtests and heatmap related files (images, code and notebooks) are added regularly.
Here, we scale the axis logarithmically optimized for the entire price range to show meaningful price action across a highly volatile token like MILK (MuesliSwap's native token)
def custom_log_scale(min_val, max_val):
lower = max(0.05, np.floor(min_val * 2) / 2)
upper = np.ceil(max_val * 2) / 2
base = 2
start = np.log(lower) / np.log(base)
stop = np.log(upper) / np.log(base)
ticks = np.logspace(start, stop, num=8, base=base)
ticks[0] = lower
return lower, upper, ticks
Below is a Heatmap for parameter-optimization and risk-tuning for a trading strategies with 2 degrees of freedom
Even complex strategies with many features can be optimized using the Heatmap feature with enough compute and the right plotting library.