feat(scripts): add workflow for updating the efficient frontier #107
Conversation
| return daily_returns.values, name | ||
|
|
||
|
|
||
| def update_returns(): |
There was a problem hiding this comment.
seems odd this is called update_returns when it really is getting and formatting data.
There was a problem hiding this comment.
good point, will change it to get_returns() or process_returns()
| MAX_DELAY = 3.0 | ||
| MAX_RETRIES = 3 | ||
|
|
||
| while True: |
There was a problem hiding this comment.
why have this code in a while loop if it just runs once?
There was a problem hiding this comment.
ah this was a dirty fix for the case when the status code != 200
There was a problem hiding this comment.
Could potentially run into an infinite loop if the status code != 200. Probably a good idea to include a max retry count
| df = df[start_date:end_date] | ||
|
|
||
| # convert to daily returns | ||
| daily_returns = (1 + df.apy) ** (1 / 365.2425) - 1 |
There was a problem hiding this comment.
why 1 / 365.2425 ? Isn't it just ** 365 ?
There was a problem hiding this comment.
ah this is because the returns from defillama are measured in yearly values, so I converted them to daily by taking the 365.2425-th root
|
|
||
| def main_chart(returns): | ||
| # aggregate to yearly values | ||
| T = 365.2425 |
There was a problem hiding this comment.
Dupe line as L66. Probably can abstract into a separate function to be shared
| frontier_chart, frontier_data = main_chart(data) | ||
|
|
||
| # login to datapane | ||
| dp.login(token=os.environ["DATAPANE_TOKEN"]) |
There was a problem hiding this comment.
Think it's good to login first before parsing the data. Had the script throwing KeyError: 'DATAPANE_TOKEN' after waiting for half an hour
adds the entrypoint file and the functions to allow automatic updates of the efficient frontier visualization.
Currently only supports the stablecoin assets, will open another PR to support the other asset pools.