Command
vraven analyse
Run a common analysis with beginner-friendly defaults
Aliases: analyze, quick
Minimal form
vraven analyse <model>Arguments and options
| Option | Meaning | Default |
|---|---|---|
model | PyTorch model file | None |
--model-format | Model storage format. VRAVEN instrumentation supports state_dict and trusted pickle models; TorchScript can be loaded but cannot expose PyTorch hooks. | None |
--model-factory, --factory | Import path such as my_project.models:Classifier | None |
--factory-kwargs | JSON object passed to the model factory | None |
--state-dict-key | — | None |
--strip-prefix | — | None |
--device | — | None |
--non-strict | — | False |
--trusted-model | Allow Python-pickle model loading | False |
-i, --input | Input tensor file; optional for a VRAVEN bundle directory | None |
-t, --task | — | None |
-p, --preset | Trade runtime for analysis depth | None |
--target | Optional class target | None |
--target-class | Required for counterfactual search | None |
--labels | Optional label file for dataset/adversarial tasks | None |
--class-names | Optional comma-separated class names | None |
--format | — | None |
-o, --output | Output path; a task-specific name is used by default | None |
Exact local helpRun
vraven analyse --help against your installed version before automating a release workflow.