What is this for?

[jgwill]

it's a public exploration of exploring data sets for trading. It involves what ?

• a centralized dataset of the learning
• a periodical refresh of the dataset (onNewTimeframe, onSignalDetected,onPatternChanged)

what else ?

• Notes on Mastering the existing Pipelines such as Github Action and HuggingFace Spaces.

[jgwill]

Summary of Observations

(warn:copilot assisted)

• HuggingFace Spaces: When new content is pushed, it rebuilds and restarts the space.
• Prototype Details:
  • Platform: JupyterLab
  • Data Refresh: Utilizes the jgtml package and its dependencies (jgtpy, jgtfxcon).
  • Next Action Step: Generating variations of the 'MLF' class.
  • Functionality: Creates datasets of lagging features based on settings.
  • Tools: CLI pncli installed with jgtml.
  • Settings: Defined in .jgt/settings.json in multiple possible locations (current directory, user home, system /etc/jgt/settings.json, or JGTSETTINGS environment variable).

Next Steps

• Generate Variations of the 'MLF' Class:
  • Ensure the settings are correctly defined in one of the recognized locations.
  • Use the CLI pncli to generate the desired datasets.

[jgwill]

@jgwill MLF Data exploration experiment 240907 could end up (partially or entirely in this repo) and refreshed from the github action (or triggered by a push/pull or any changes we desire). One of the requirements would be for it to communicate with its sister repo space on HuggingFace (of the same namespace dataset:jgwill/xjgtml ...??