Asset Management

Platform for multi-strategy multi-asset, systematic trading strategies. Built with Python, fast api, PostgreSQL, htmx, Alpine.js and Daisy UI

Pipeline from data -> modeling -> execution.
admin_dashboard: UI tasks: Tasks that run on vm. common: Common utilities notebooks: Jupyter notebooks for any user case. Names should have format YYYY-mm-dd-whatever-you-want.ipynb*

TODO:

• Move backtest_results to tmp
• Add main test
• Further combine trading and allocation.
  • Take paralelization out and into both
  • Make inputs more uniform
  • Make trading have oscilators and indicators
• Take data out of BacktestResult
• Make sure strategy name drop down across models tabs is synced
• Try to send "with DatabaseConnection() as [conn, cur]:" to the bottom db layer in all cases in this case BaseTable and others
• Shouldnt have strategy_args and backtest_args
• fix RuntimeWarning: invalid value encountered in divide np.divide
  • Probably comming from math_utils.py
• Standardize all urls
  • api should start with /api
  • Also routes
• Add loging
• Add tab in models-portfolios to be able to combine results results with a checkbox with the mean
• Add signal level metrics to Assets like Streamlit had
• Add functionality to see stopped portfolios and also delete them

gcloud config set project aglaparadim

deployment

cd platform/terraform/chatborg/staging terraform taint module.container_registry_container.null_resource.image && terraform apply --auto-approve