loop over runs in dc, dc-closure-test, and plot-dc to prevent using too much memory

1 files changed, 8 insertions, 1 deletions

diff --git a/utils/dc-closure-test b/utils/dc-closure-test
index ba5daf8..e23f554 100755
--- a/utils/dc-closure-test
+++ b/utils/dc-closure-test
@@ -425,13 +425,20 @@ if __name__ == '__main__':
     parser.add_argument("--save", action="store_true", default=False, help="save plots")
     parser.add_argument("--mc", nargs='+', required=True, help="atmospheric MC files")
     parser.add_argument("-n", type=int, default=10, help="number of fits to run")
+    parser.add_argument("--nhit-thresh", type=int, default=None, help="nhit threshold to apply to events before processing (should only be used for testing to speed things up)")
     args = parser.parse_args()
 
     setup_matplotlib(args.save)
 
     import matplotlib.pyplot as plt
 
-    ev = get_events(args.filenames,merge_fits=True)
+    # Loop over runs to prevent using too much memory
+    evs = []
+    rhdr = pd.concat([read_hdf(filename, "rhdr").assign(filename=filename) for filename in args.filenames],ignore_index=True)
+    for run, df in rhdr.groupby('run'):
+        evs.append(get_events(df.filename.values, merge_fits=True, nhit_thresh=args.nhit_thresh))
+    ev = pd.concat(evs)
+
     ev = ev[ev.prompt]
     ev = ev[ev.nhit_cal > 100]