Added csv file as performance_test argument

notebook/batching_performance.ipynb

@@ -69,14 +69,14 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "def performance_test(batch_size, disp=True):\n",
+    "def performance_test(csv_file, batch_size, disp=True):\n",
     "    print(\"-----------------------------------\")\n",
     "    print(\"Starting performance test for batch size\", batch_size)\n",
     "    # create t0 timestamp\n",
     "    t0 = time.time()\n",
     "\n",
     "    # import files from csv file\n",
-    "    email_list = pd.read_csv(\"../mailcom/test/data/mails_lb_sg_copy2.csv\")\n",
+    "    email_list = pd.read_csv(csv_file)\n",
     "    #print(email_list)\n",
     "\n",
     "    t_csv_read = time.time()\n",
@@ -194,6 +194,7 @@
     "# batching_sizes = [1]\n",
     "n_samples = 5\n",
     "outputs = []\n",
+    "csv_file = \"../mailcom/test/data/mails_lb_sg_copy2.csv\"\n",
     "\n",
     "# first make a dummy run since there seem to be some inconsitencies when loading for the first time\n",
     "_ = performance_test(-1, disp=False)\n",
@@ -203,7 +204,7 @@
     "for bs in batching_sizes:\n",
     "    average_email_time = 0\n",
     "    for sid in range(n_samples):\n",
-    "        t, out = performance_test(bs, disp=False)\n",
+    "        t, out = performance_test(csv_file, bs, disp=False)\n",
     "        average_email_time += t\n",
     "        test_result_dict = {\n",
     "            \"batch_size\": bs,\n",