numin 1.2.8

numin package

numin Package

numin is a Python package designed for algorithmic trading and backtesting providing an API called NuminAPI.

Features

• Data Retrieval: Download training, round, and validation data.
• Prediction Submission: Upload prediction CSVs to the server with built-in validation.
• Real-Time Round Management: Fetch the current trading round from the server.
• Backtesting: Run backtests on user-defined strategies with support for additional features and discretization.
• File Management: Upload and deploy files via Anvil's server.
• Returns Summary: Retrieve and format live (or simulation) trading returns.

Supported Methods

• Data Download:

  • get_data(data_type: str)
    • Fetches data from the server based on the type (training, round, or validation).

• Prediction Submission:

  • submit_predictions(file_path: str)
    • Submits a CSV file of predictions to the server.
    • Note: The file must include mandatory columns ["id", "predictions", "round_no"] and optional columns ["stop", "target", "tLimit"] (if provided, they must be integers between 1 and 100).

• Round and Validation Data:

  • get_current_round()
    • Retrieves the current round number from the server.
  • fetch_validation_data(date: str)
    • Downloads validation data for a given date.
  • get_validation_dates()
    • Lists available validation dates on the server.

• Backtesting:

  • run_backtest(user_strategy : str, date:str, val_data:str, val_df:DataFrame,result_type="results" or "returns")
    • Executes a backtest using a user-provided strategy function on the given date.
  • display_results(backtest_results: results as above, validation_dataframe: dataFrame, indicators: list of indicators to show e.g. used for entry by strategy )
    • Displays results of a backtest in a readable format with entry, exit prices, p/l etc.
    • Input can be merely the date, in which case the data is downloade; if a csv file path is provided in val_data then that file is taken; if a dataframe is provided in val_df that is taken. (e.g. using the dataframe returned by fetch_validation_data.)

• File Upload and Deployment:

  • upload_file(file, user_id, filename)
    • Uploads a file to remote storage via the Anvil server.
  • deploy_file(filename: str, user_id: str)
    • Deploys a file for a given user.

• Live Returns Summary:

  • show_returns(user_id, mode="live"/"sim")
    • Retrieves and displays a formatted summary of returns for the strategy associated with the given user ID

Installation

Install numin using pip:

pip install numin