add nn interpolator script

bin/rvs_train_nn_interpolator

@@ -0,0 +1,7 @@
+#!/usr/bin/env python
+
+import sys
+import rvspecfit.nn.train_interpolator as rvt
+
+if __name__ == '__main__':
+    rvt.main(sys.argv[1:])

py/rvspecfit/nn/train_interpolator.py

@@ -1,16 +1,15 @@
-import rvspecfit
+import pickle
+import os
+import sys
+import argparse
 import time
+import rvspecfit
 import torch.utils.data as toda
-# import matplotlib.pyplot as plt
 import torch.nn.functional as tofu
 import sklearn.decomposition as skde
-import pickle
 import numpy as np
 import torch
-import os
-from idlsave import idlsave
 from NNInterpolator import Mapper, NNInterpolator
-import argparse
 
 git_rev = rvspecfit.__version__
 
@@ -46,7 +45,7 @@ def getSchedOptim(optim):
     return sched
 
 
-def get_predictions(myint, Tvecs0, dev):
+def get_predictions(myint, Tvecs0, dev, batch):
     alldat = toda.TensorDataset(Tvecs0)
     pred = []
     batchdatF = toda.DataLoader(
@@ -64,7 +63,7 @@ def get_predictions(myint, Tvecs0, dev):
     return pred
 
 
-if __name__ == '__main__':
+def main(args):
 
     parser = argparse.ArgumentParser()
     parser.add_argument('--cpu', action='store_true', default=False)
@@ -89,7 +88,7 @@ def get_predictions(myint, Tvecs0, dev):
     parser.add_argument('--mask_ids', type=str, default=None)
     parser.add_argument('--setup', type=str)
     parser.add_argument('--batch', type=int, default=100)
-    args = parser.parse_args()
+    args = parser.parse_args(args)
     log_ids = [int(_) for _ in (args.log_ids).split(',')]
     setup = args.setup
     revision = args.revision
@@ -304,10 +303,14 @@ def get_predictions(myint, Tvecs0, dev):
     myint.pc_layer.bias.data[:] = tD_0[:] + myint.pc_layer.bias.data[:] * tSD_0
 
     torch.save(myint.state_dict(), finalfile_path)
-    pred = get_predictions(myint, Tvecs0, train_dev)
-    idlsave.save(f'{directory}/pred_{setup}.psav',
-                 'pred,vecs,dats,mapper,vecs_orig,mask_ids', pred, vecs, dats,
-                 mapper, vecs_orig, mask_ids)
+    pred = get_predictions(myint, Tvecs0, train_dev, batch)
+    with open(f'{directory}/pred_{setup}.psav', 'wb') as fp:
+        DD = {}
+        DD['pred'] = pred,
+        DD['vecs'] = vecs
+        DD['dats'] = dats,
+        DD['mapper'] = mapper
+        DD['vecs_orig'] = vecs_orig
     if os.path.exists(statefile_path):
         os.unlink(statefile_path)
     import rvspecfit.nn.RVSInterpolator
@@ -331,3 +334,7 @@ def get_predictions(myint, Tvecs0, dev):
             'interpolation_type': 'generic'
         }
         pickle.dump(D, fp)
+
+
+if __name__ == '__main__':
+    main(sys.argv[1:])