feature-engineering
Build trading model features without future data.
What it does
Guides construction, validation, and selection of market-data features for crypto trading models.
When to use it
Use when engineering features for classification or regression models targeting crypto and Solana token trading.
How to use it
It changes how the agent constructs and selects trading-model features, requiring rolling calculations, forward-looking targets, and temporal train/test splits.
What you provide
- An existing project
Uses
What you need · 7
The bundled feature-building script runs with Python 3.
The bundled scripts import numpy for numerical feature computation.
The bundled scripts import pandas to process market data and feature tables.
The feature-importance script imports adfuller from scipy.stats as written.
Access · 0
No additional actions listed.