How cseptesting/floatcsep works
floatCSEP deploys reproducible prospective earthquake-forecasting experiments. A user runs a configuration through the CLI; the experiment coordinates model execution, catalog and forecast access, evaluations, then reports and plots results. The separate view command serves an exported dashboard. The graph preserves configuration-driven model, evaluation, infrastructure, and web-reporting branches. Relationships are limited to README statements and source-index references; unsampled modules are represented only where the tree supports their subsystem.
Main entry points: experiment.py Experiment orchestration · main.py CLI commands · model.py Forecast models · repositories.py Catalog and result repositories
Components 17
User entry points
- CLI commands
floatcsep/commands/main.py
Experiment execution
- Experiment orchestration
floatcsep/experiment.py - Forecast models
floatcsep/model.py - Evaluation tests
floatcsep/evaluation.py
Execution infrastructure
- Task execution
floatcsep/infrastructure/engine.py - Model environments
floatcsep/infrastructure/environments.py - Run registries
floatcsep/infrastructure/registries.py - Catalog and result repositories
floatcsep/infrastructure/repositories.py - Remote model access
floatcsep/utils/accessors.py - Forecast file parsing
floatcsep/utils/file_io.py - Configuration helpers
floatcsep/utils/helpers.py
Results and exploration
- Forecast and catalog plots
floatcsep/postprocess/plot_handler.py - Reports
floatcsep/postprocess/reporting.py - Dashboard export
floatcsep/postprocess/web/export.py - Dashboard server
floatcsep/postprocess/web/serve.py - Dashboard interface
floatcsep/postprocess/web/static/js/app.js
Other components
- Experiment user
Connections 26
- Experiment user → CLI commands: runs or views
README.md - CLI commands → Experiment orchestration: configures and runs
floatcsep/commands/main.py - CLI commands → Reports: generates report
floatcsep/commands/main.py - CLI commands → Forecast and catalog plots: generates plots
floatcsep/commands/main.py - CLI commands → Dashboard server: serves dashboard
floatcsep/commands/main.py - Experiment orchestration → Forecast models: configures models
floatcsep/experiment.py - Experiment orchestration → Evaluation tests: configures tests
floatcsep/experiment.py - Experiment orchestration → Task execution: builds task graph
floatcsep/experiment.py - Experiment orchestration → Run registries: creates registries
floatcsep/experiment.py - Experiment orchestration → Catalog and result repositories: sets repositories
floatcsep/experiment.py - Forecast models → Model environments: creates environment
floatcsep/model.py - Forecast models → Run registries: creates model registry
floatcsep/model.py - Forecast models → Catalog and result repositories: loads forecasts
floatcsep/model.py - Task execution → Model environments: uses Docker manager
floatcsep/infrastructure/engine.py - Evaluation tests → Run registries: uses registries
floatcsep/evaluation.py - Evaluation tests → Configuration helpers: uses helpers
floatcsep/evaluation.py - Catalog and result repositories → Run registries: uses registry paths
floatcsep/infrastructure/repositories.py - Catalog and result repositories → Forecast file parsing: loads forecast files
floatcsep/infrastructure/repositories.py - Catalog and result repositories → Configuration helpers: uses helpers
floatcsep/infrastructure/repositories.py - Forecast models → Remote model access: retrieves model artifacts
floatcsep/model.py - Reports → Experiment orchestration: reads experiment state
floatcsep/postprocess/reporting.py - Reports → Forecast and catalog plots: uses plot functions
floatcsep/postprocess/reporting.py - CLI commands → Dashboard export: exports dashboard data
- Dashboard server → Dashboard interface: serves interface
- Dashboard export → Run registries: reads run metadata
floatcsep/postprocess/web/export.py - Dashboard interface → Dashboard export: loads dashboard data