Torchscan

TorchScan

Use the smallest API that answers the request:

• crawl_module(...): JSON-serializable module report.
• summary(...): printed table plus the same report.
• measure_flops(workload): operator FLOPs for one zero-argument workload call.
• measure_peak_memory(workload, device=...): backend-specific PyTorch peak memory.
• compare_reports(before, after): pure same-schema comparison.

Workflow

1. Reuse the project's model and representative inputs. Do not download weights without permission.
2. Prefer args and kwargs for real calls; use input_shape only for simple synthetic tensors.
3. Use strict=True when incomplete module metrics must stop automation.
4. Serialize the report directly. Never parse the summary table.
5. Check every metric's status and preserve diagnostics.
6. Ask the owner for thresholds. TorchScan measures; it does not decide whether a model fits.

Truth rules

• complete: use value with its method, unit, scope, and context.
• partial: known_value is only a lower bound; do not extrapolate.
• unavailable: report that no measurement was produced.
• Zero is valid only with status == "complete".
• Keep module FLOPs and operator FLOPs separate.
• Peak PyTorch memory is not process RSS or total device memory.
• Mocked or skipped CUDA/MPS checks are not hardware evidence.

For an uncounted operator, preserve the partial result. Rerun the equivalent workload with measure_flops(..., custom_mapping=...) only when the owner can justify that operator's counting convention; crawl_module does not accept custom mappings. Do not create a global registry, baseline store, wrapper service, or automatic budget policy.

In a repository checkout, read ../../../docs/docs/agent-quickstart.md for the full workflow and ../../../docs/docs/report-schema.md for the report contract. Outside a checkout, use the published documentation at https://frgfm.github.io/torch-scan/.