module documentation
Models for the riskpredictor Django app.
The model CheckpointModel holds an upload of parameter samples that were produced
during an inference run of the lymph-model. The samples should be fetched from the
DVC remote storage and used to compute the prior risk matrices.
Given a specific diagnosis, as entered via the forms.RiskpredictorForm, the
lymph-model package and the precomputed risk matrices in the CheckpointModel
instances, the personalized risk estimates can be computed.
| Class | |
Results of a round of parameter sampling for one of the lymph.models. |
| Function | cached |
Compute the prior state dists for the given model, samples, and t_stage. |
| Function | cached |
Construct the lymph model and add the distributions to it. |
| Function | cached |
Fetch and merge the YAML configuration files from the repo. |
| Function | cached |
Fetch the model samples from the HDF5 file in the DVC repo. |
| Function | validate |
Validate the pydantic configs necessary for constructing the model. |
| Type Alias | |
Undocumented |
| Variable | logger |
Undocumented |
@JOBLIB_MEMORY.cache
def cached_compute_priors(model: Model, samples: np.ndarray, t_stage: str | int) -> np.ndarray:
¶
Compute the prior state dists for the given model, samples, and t_stage.
@JOBLIB_MEMORY.cache
def cached_construct_model_and_add_dists(graph_config: GraphConfig, model_config: ModelConfig, dist_configs: dict[, version: int) -> Model:
¶
Construct the lymph model and add the distributions to it.
@JOBLIB_MEMORY.cache
def cached_fetch_and_merge_yaml(repo_name: str, ref: str, graph_config_path: str, model_config_path: str, dist_configs_path: str) -> dict[:
¶
Fetch and merge the YAML configuration files from the repo.
@JOBLIB_MEMORY.cache
def cached_fetch_model_samples(repo_name: str, ref: str, samples_path: str, num_samples: int, remote: str | None = None, seed: int = 42) -> np.ndarray:
¶
Fetch the model samples from the HDF5 file in the DVC repo.
ConfigAndVersionTupleType =
¶
Undocumented
| Value |
|