YAML Configuration Structure¶
This page explains the common structure of a workflow configuration file and points to the full template for reference.
Start with the template¶
The repository ships a full template:
Copy the template and edit only the fields you need for your run.
Common sections¶
Most workflows use a subset of these sections:
paths: input data locations and output directoriesdata: dataset-specific settings (ranges, filtering, batching)model: model name, parameters, and boundsfitting: solver selection, stopping criteria, and tolerancesmultistart: global search options (LHS, Sobol, Halton)hybrid_streaming: streaming optimizer and defense layer configuration (v0.3.6+)resources: memory and device controlslogging: verbosity and log file destinations
Minimal example¶
paths:
input: ./data/experiment_01.csv
output_dir: ./runs/experiment_01
model:
name: exponential_decay
parameters:
p0: [2.0, 0.5]
fitting:
solver: auto
max_nfev: 200
Defense layers example (v0.3.6+)¶
Configure the 4-Layer Defense Strategy for warm-start refinement:
hybrid_streaming:
normalize: true
warmup_iterations: 300
defense_layers:
preset: strict # or "relaxed", "scientific", "disabled"
# Fine-tune individual layers (optional)
layer1_warm_start:
enabled: true
threshold: 0.01
layer2_adaptive_lr:
enabled: true
lr_refinement: 1.0e-6
lr_careful: 1.0e-5
lr_exploration: 0.001
layer3_cost_guard:
enabled: true
tolerance: 0.05
layer4_step_clipping:
enabled: true
max_step_size: 0.1
telemetry:
enabled: true
export_format: prometheus
See Configuration Reference for the complete configuration reference.
Workflow options¶
For user-level options like loss functions, callbacks, and solver choices, see Configuration Reference.
Advanced customization¶
If you need programmatic workflow construction or custom models, start with Developer Documentation.