cs4486-amp-pd-checkpoints / code /ampd /retained_models.py
fryabcfry's picture
Publish AMP-PD checkpoints, portable loaders and verified metrics (part 3)
04ae9bb verified
Raw History Blame Contribute Delete
1.71 kB
"""Load the three retained current-visit models without discarded branches."""
from pathlib import Path
import numpy as np
from ampd.ae_residual import AEResidualPredictor, FrozenMonthMLP, correct
from ampd.omics_boost import transform_pls
from ampd.simple_mlp_baseline import SimpleMLPPredictor
class PLSPredictor:
def __init__(self, bundle):
bundle = Path(bundle)
self.baseline = FrozenMonthMLP(bundle / "baseline")
self.spec = self.baseline.spec
with np.load(bundle / "pls.npz", allow_pickle=False) as data:
self.params = dict(data)
def predict(self, events):
for event in events:
if set(event) != {"visit_id", "visit_month", "proteins"}:
raise ValueError("Only current visit_id, visit_month and proteins are accepted")
if len(event["proteins"]) != len(self.spec["proteins"]):
raise ValueError("Protein vocabulary mismatch")
proteins = np.asarray([e["proteins"] for e in events], dtype=float)
if np.any(proteins < 0):
raise ValueError("Protein abundances must be nonnegative")
baseline = self.baseline.predict([e["visit_month"] for e in events])
residual = transform_pls(proteins, self.params) @ self.params["coef"].T
return {"baseline": baseline, "pls": correct(baseline, residual, self.params["weights"])}
def load_retained(kind, bundle):
if kind == "baseline_mlp":
return SimpleMLPPredictor(bundle, feature_sets=("month_protein",))
if kind == "selected_ae":
return AEResidualPredictor(bundle, arms=("selected_latent_r",))
if kind == "pls":
return PLSPredictor(bundle)
raise ValueError(kind)