buckyroberts/ella

Ella is a small Python program that reads through a dataset and makes decisions based on what it finds. Think of it as a simple automated rule-maker: you give it data, and it creates conditions and actions to respond to that data. The core idea is that it builds a "brain" of decision rules, then applies them row by row. The program builds its decision-making system out of a few layers. It generates conditions from your data (like "volume is greater than 3700"), pairs each condition with an action (like "BUY" or "SELL"), and calls each condition-action pair a "neuron." The full collection of neurons is the "brain." Once the brain is assembled, Ella walks through your dataset one row at a time. For each row, it checks whether any conditions are met, and if so, it triggers the matching action. The README's main example uses stock-market-style data with price and volume columns. Starting with a hypothetical $1,000, Ella applies its rules across five days of data, buying on certain days based on volume and ending with a final value of $1,008.65. This makes it most relevant for someone exploring automated trading or rule-based decision systems over structured data. The author notes this was a small project written in a morning, likely not actively maintained, though open to community contributions. It requires Python 3.6 or newer. The README doesn't go into detail on how conditions are chosen or optimized beyond mentioning they start as random expressions, so anyone expecting sophisticated learning should temper their expectations.