Merge pull request #249 from nipreps/fix/dwi_experiment_improvements

ENH: Improve CLI of the dwi/error experiment

scripts/dwi_gp_estimation_error_analysis.py

@@ -34,7 +34,7 @@
 
 import numpy as np
 import pandas as pd
-from sklearn.model_selection import RepeatedKFold, cross_val_score
+from sklearn.model_selection import KFold, RepeatedKFold, cross_val_predict, cross_val_score
 
 from eddymotion.model._sklearn import (
     EddyMotionGPR,
@@ -92,22 +92,52 @@ def _build_arg_parser() -> argparse.ArgumentParser:
         description=__doc__, formatter_class=argparse.RawTextHelpFormatter
     )
     parser.add_argument(
-        "hsph_dirs",
+        "--bval-shell",
+        help="Shell b-value",
+        type=float,
+        default=1000,
+    )
+    parser.add_argument("--S0", help="S0 value", type=float, default=100)
+    parser.add_argument(
+        "--hsph-dirs",
         help="Number of diffusion gradient-encoding directions in the half sphere",
         type=int,
+        default=60,
     )
-    parser.add_argument("bval_shell", help="Shell b-value", type=int)
-    parser.add_argument("S0", help="S0 value", type=float)
     parser.add_argument(
-        "error_data_fname",
+        "--output-scores",
         help="Filename of TSV file containing the data to plot",
         type=Path,
+        default=Path() / "scores.tsv",
+    )
+    parser.add_argument(
+        "-n",
+        "--n-voxels",
+        help="Number of diffusion gradient-encoding directions in the half sphere",
+        type=int,
+        default=100,
+    )
+    parser.add_argument(
+        "--write-inputs",
+        help="Filename of NIfTI file containing the generated DWI signal",
+        type=Path,
+        default=None,
+    )
+    parser.add_argument(
+        "--output-predicted",
+        help="Filename of NIfTI file containing the predicted DWI signal",
+        type=Path,
+        default=None,
     )
     parser.add_argument("--evals", help="Eigenvalues of the tensor", nargs="+", type=float)
     parser.add_argument("--snr", help="Signal to noise ratio", type=float)
     parser.add_argument("--repeats", help="Number of repeats", type=int, default=5)
     parser.add_argument(
-        "--kfold", help="Number of directions to leave out/predict", nargs="+", type=int
+        "--kfold",
+        help="Number of folds in repeated-k-fold cross-validation",
+        nargs="+",
+        type=int,
+        default=None,
     )
     return parser
 
@@ -134,28 +164,36 @@ def main() -> None:
     parser = _build_arg_parser()
     args = _parse_args(parser)
 
-    n_voxels = 100
-
     data, gtab = testsims.simulate_voxels(
         args.S0,
         args.hsph_dirs,
         bval_shell=args.bval_shell,
         snr=args.snr,
-        n_voxels=n_voxels,
+        n_voxels=args.n_voxels,
         evals=args.evals,
         seed=None,
     )
 
+    # Save the generated signal and gradient table
+    if args.write_inputs:
+        testsims.serialize_dmri(
+            data,
+            gtab,
+            args.write_inputs,
+            args.write_inputs.with_suffix(".bval"),
+            args.write_inputs.with_suffix(".bvec"),
+        )
+
     X = gtab[~gtab.b0s_mask].bvecs
     y = data[:, ~gtab.b0s_mask]
 
     snr_str = args.snr if args.snr is not None else "None"
 
     a = 1.15
     lambda_s = 120
-    alpha = 100
+    alpha = 1
     gpr = EddyMotionGPR(
-        kernel=SphericalKriging(a=a, lambda_s=lambda_s),
+        kernel=SphericalKriging(beta_a=a, beta_l=lambda_s),
         alpha=alpha,
         optimizer=None,
         # optimizer="Nelder-Mead",
@@ -164,25 +202,35 @@ def main() -> None:
         # max_iter=2e5,
     )
 
-    # Use Scikit-learn cross validation
-    scores = defaultdict(list, {})
-    for n in args.kfold:
-        for i in range(args.repeats):
-            cv_scores = -1.0 * cross_validate(X, y.T, n, np.max(args.kfold) // n, gpr)
-            scores["rmse"] += cv_scores.tolist()
-            scores["repeat"] += [i] * len(cv_scores)
-            scores["n_folds"] += [n] * len(cv_scores)
-            scores["bval"] += [args.bval_shell] * len(cv_scores)
-            scores["snr"] += [snr_str] * len(cv_scores)
-
-        print(f"Finished {n}-fold cross-validation")
-
-    scores_df = pd.DataFrame(scores)
-    scores_df.to_csv(args.error_data_fname, sep="\t", index=None, na_rep="n/a")
-
-    grouped = scores_df.groupby(["n_folds"])
-    print(grouped[["rmse"]].mean())
-    print(grouped[["rmse"]].std())
+    if args.kfold:
+        # Use Scikit-learn cross validation
+        scores = defaultdict(list, {})
+        for n in args.kfold:
+            for i in range(args.repeats):
+                cv_scores = -1.0 * cross_validate(X, y.T, n, gpr)
+                scores["rmse"] += cv_scores.tolist()
+                scores["repeat"] += [i] * len(cv_scores)
+                scores["n_folds"] += [n] * len(cv_scores)
+                scores["bval"] += [args.bval-shell] * len(cv_scores)
+                scores["snr"] += [snr_str] * len(cv_scores)
+
+            print(f"Finished {n}-fold cross-validation")
+
+        scores_df = pd.DataFrame(scores)
+        scores_df.to_csv(args.output_scores, sep="\t", index=None, na_rep="n/a")
+
+        grouped = scores_df.groupby(["n_folds"])
+        print(grouped[["rmse"]].mean())
+        print(grouped[["rmse"]].std())
+    else:
+        gpr.fit(X, y.T)
+        print(gpr.kernel_)
+
+    if args.output_predicted:
+        cv = KFold(n_splits=3, shuffle=False, random_state=None)
+        predictions = cross_val_predict(gpr, X, y.T, cv=cv)
+
+        testsims.serialize_dwi(predictions.T, args.output_predicted)
 
 
 if __name__ == "__main__":