compute throughput correctly

Author: maumueller

datasets.py

@@ -1,3 +1,4 @@
+import h5py
 import os 
 from urllib.request import urlretrieve
 from pathlib import Path
@@ -19,6 +20,13 @@ def prepare(dataset, task):
     download(url, fn)
     download(gt_url, gt_fn)
 
+def get_query_count(dataset, task):
+    fn, _ = get_fn(dataset, task) 
+    f = h5py.File(fn)
+    qn = len(DATASETS[dataset][task]['queries'](f))
+    f.close()
+    return qn
+
 DATASETS = {
     'ccnews-small': {
         'task1': {

show_operating_points.py

@@ -1,6 +1,8 @@
 import argparse
 import pandas as pd
 
+from datasets import get_query_count
+
 # show best performing parameters exceeding threshold
 
 if __name__ == "__main__":