fix: allow processing multiple files at a time

kwyk/cli.py

@@ -1,7 +1,9 @@
+#! /usr/bin/env python
 from pathlib import Path
 import subprocess
 import tempfile
 import json
+import os
 
 import click
 import nibabel as nib
@@ -21,16 +23,19 @@
     'bvwn_multi_prior': _here.parent / 'saved_models' / 'all_50_bvwn_multi_prior' / '1556816070',
 }
 
+
 @click.command()
-@click.argument('infile')
+@click.argument('infiles', nargs=-1)
 @click.argument('outprefix')
 @click.option('-m', '--model', type=click.Choice(_models.keys()), default="bwn_multi", required=True, help='Model to use for prediction.')
 @click.option('-n', '--n-samples', type=int, default=1, help='Number of samples to predict.')
 @click.option('-b', '--batch-size', type=int, default=8, help='Batch size during prediction.')
 @click.option('--save-variance', is_flag=True, help='Save volume with variance across `n-samples` predictions.')
 @click.option('--save-entropy', is_flag=True, help='Save volume of entropy values.')
+@click.option('--overwrite', type=click.Choice(['yes', 'skip'], case_sensitive=False), help='Overwrite existing output or skip')
+@click.option('--atlocation', is_flag=True, help='Save output in the same location as input')
 @click.version_option(version=__version__)
-def predict(*, infile, outprefix, model, n_samples, batch_size, save_variance, save_entropy):
+def predict(*, infiles, outprefix, model, n_samples, batch_size, save_variance, save_entropy, overwrite, atlocation):
     """Predict labels from features using a trained model.
 
     The predictions are saved to OUTPREFIX_* with the same extension as the input file.
@@ -46,6 +51,13 @@ def predict(*, infile, outprefix, model, n_samples, batch_size, save_variance, s
             print("Your version: {0} Latest version: {1}".format(__version__,
                                                                  latest["version"]))
 
+    savedmodel_path = _models[model]
+    predictor = _get_predictor(savedmodel_path)
+    for infile in infiles:
+        _predict(infile, outprefix, predictor, n_samples, batch_size, save_variance, save_entropy, overwrite, atlocation)
+
+
+def _predict(infile, outprefix, predictor, n_samples, batch_size, save_variance, save_entropy, overwrite, atlocation):
     _orig_infile = infile
 
     # Are there other neuroimaging file extensions with multiple periods?
@@ -55,14 +67,22 @@ def predict(*, infile, outprefix, model, n_samples, batch_size, save_variance, s
         outfile_ext = Path(infile).suffix
     outfile_stem = outprefix
 
+    if atlocation:
+        outfile_stem = Path(infile).parent / outfile_stem
+
     outfile_means = "{}_means{}".format(outfile_stem, outfile_ext)
     outfile_variance = "{}_variance{}".format(outfile_stem, outfile_ext)
     outfile_entropy = "{}_entropy{}".format(outfile_stem, outfile_ext)
     outfile_uncertainty = "{}_uncertainty{}".format(outfile_stem, '.json')
 
     for ff in [outfile_means, outfile_variance, outfile_entropy, outfile_uncertainty]:
         if Path(ff).exists():
-            raise FileExistsError("file exists: {}".format(ff))
+            if overwrite == "skip":
+                return
+            elif overwrite == "yes":
+                pass
+            else:
+                raise FileExistsError("file exists: {}".format(ff))
 
     required_shape = (256, 256, 256)
     block_shape = (32, 32, 32)
@@ -79,10 +99,7 @@ def predict(*, infile, outprefix, model, n_samples, batch_size, save_variance, s
     else:
         tmp = None
 
-    savedmodel_path = _models[model]
-
     print("++ Running forward pass of model.")
-    predictor = _get_predictor(savedmodel_path)
     outputs = predict_from_filepath(
         infile,
         predictor=predictor,
@@ -125,12 +142,12 @@ def predict(*, infile, outprefix, model, n_samples, batch_size, save_variance, s
             json.dump(average_uncertainty, fp, indent=4)
 
 
-
 def _conform(input, output):
     """Conform volume using FreeSurfer."""
     subprocess.run(['mri_convert', '--conform', input, output], check=True)
     return output
 
+
 def _reslice(input, output, reference, labels=False):
     """Conform volume using FreeSurfer."""
     if labels:
@@ -140,3 +157,8 @@ def _reslice(input, output, reference, labels=False):
     else:
         subprocess.run(['mri_convert', '-rl', reference, input, output], check=True)
     return output
+
+
+if __name__ == '__main__':
+    predict()
+