mirror of
https://github.com/harvard-edge/cs249r_book.git
synced 2026-07-21 00:23:30 -05:00
Vijay Janapa Reddi
cb0ae4082f
Run quartodoc build against the refactored package (engine/ extracted from core/, infra->infrastructure): regenerate the API reference so every stub points at the new paths. Adds engine.*/infrastructure.* stubs, removes the 41 stale core.solver.* / core.engine.* / core.scenarios.* / core.config.* / core.evaluation.* / infra*.qmd orphans, and refreshes the package pages (core now primitives-only) + index. Verified: all 47 documented symbols resolve in the new package; every index link resolves; zero stale mlsysim.core.<engine-mod> / mlsysim.infra references anywhere in docs/. (objects.json inventory is gitignored -- regenerated at build.)
35 lines
926 B
Plaintext
35 lines
926 B
Plaintext
# engine.solver.SensitivitySolver { #mlsysim.engine.solver.SensitivitySolver }
|
|
|
|
```python
|
|
engine.solver.SensitivitySolver()
|
|
```
|
|
|
|
Identifies the binding constraint via numerical sensitivity analysis (Wall 21).
|
|
|
|
This solver computes numerical partial derivatives of performance
|
|
with respect to hardware parameters to identify the 'binding constraint.'
|
|
|
|
Literature Source:
|
|
1. Williams et al. (2009), "Roofline Model."
|
|
2. Ofenbeck et al. (2014), "Applying the Roofline Model."
|
|
|
|
## Methods
|
|
|
|
| Name | Description |
|
|
| --- | --- |
|
|
| [solve](#mlsysim.engine.solver.SensitivitySolver.solve) | Solves for sensitivities and identifies the binding constraint. |
|
|
|
|
### solve { #mlsysim.engine.solver.SensitivitySolver.solve }
|
|
|
|
```python
|
|
engine.solver.SensitivitySolver.solve(
|
|
model,
|
|
hardware,
|
|
precision='fp16',
|
|
perturbation_pct=10.0,
|
|
efficiency=0.5,
|
|
)
|
|
```
|
|
|
|
Solves for sensitivities and identifies the binding constraint.
|