Output Modes¶
Bayaml supports multiple output rendering modes, allowing results to be displayed in different formats depending on workflow, integration needs, or presentation style.
These modes control how model results, metrics, and execution details are formatted — without affecting the underlying execution.
Available Output Modes¶
pretty– Human-readable formatted outputoriginal– Raw native Bayaml execution resultsklearn– scikit-learn compatible output formatpandas– Pandas DataFrame structured outputnumpy– NumPy array structured outputjson– Machine-readable JSON outputtable– Clean tabular display formatmarkdown– Markdown-rendered outputlatex– LaTeX formatted outputdiagnostic– Detailed execution diagnostics
Example Usage¶
result = p.auto(
"use iris.csv treat species as target train classification model",
mode="pretty"
)
result = p.auto(
"use iris.csv treat species as target train classification model",
mode="json"
)
Each mode adapts the output structure while keeping the execution deterministic and reproducible.
Why This Matters¶
These output modes allow Bayaml to integrate seamlessly with:
Research workflows (LaTeX, Markdown)
Production systems (JSON, NumPy)
Data pipelines (Pandas, sklearn)
Debugging and diagnostics
Human-readable reporting
This flexibility bridges experimentation, reporting, and deployment.