From 2525dcdf4c33edb6a25544653a0ecdcf06cfe97a Mon Sep 17 00:00:00 2001
From: Felix <65565033+fexfl@users.noreply.github.com>
Date: Tue, 24 Dec 2024 11:43:09 +0100
Subject: [PATCH] Added display as option

---
 notebook/batching_performance.ipynb | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/notebook/batching_performance.ipynb b/notebook/batching_performance.ipynb
index ade26e6..62c05b4 100644
--- a/notebook/batching_performance.ipynb
+++ b/notebook/batching_performance.ipynb
@@ -50,7 +50,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "def performance_test(batch_size):\n",
+    "def performance_test(batch_size, disp=True):\n",
     "    print(\"-----------------------------------\")\n",
     "    print(\"Starting performance test for batch size\", batch_size)\n",
     "    # create t0 timestamp\n",
@@ -102,11 +102,12 @@
     "        ts_list.append([ts_email_start, ts_email_ppr_done, ts_email_end])\n",
     "\n",
     "        # display the pseudonymized text\n",
-    "        display(HTML(output_text))\n",
+    "        # display(HTML(output_text))\n",
     "\n",
     "        # display original text and highlight found and replaced NEs\n",
-    "        highlighted_html = highlight_ne(text, ps.per_list, ps.org_list, ps.loc_list, ps.misc_list)\n",
-    "        display(HTML(highlighted_html))\n",
+    "        if disp:\n",
+    "            highlighted_html = highlight_ne(text, ps.per_list, ps.org_list, ps.loc_list, ps.misc_list)\n",
+    "            display(HTML(highlighted_html))\n",
     "\n",
     "    # display timestamps\n",
     "\n",
@@ -161,15 +162,15 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "batching_sizes = [-1, -1, 1, 2, 3, 6, 10] # first run is ignored since there seem to be some inconsistencies when loading for the first time\n",
+    "batching_sizes = [-1, -1, 1, 2, 3, 4, 6, 8, 10] # first run is ignored since there seem to be some inconsistencies when loading for the first time\n",
     "# batching_sizes = [1]\n",
-    "n_samples = 1\n",
+    "n_samples = 3\n",
     "\n",
     "av_email_times_for_batches = []\n",
     "for bs in batching_sizes:\n",
     "    average_email_time = 0\n",
     "    for _ in range(n_samples):\n",
-    "        t = performance_test(bs)\n",
+    "        t = performance_test(bs, disp=False)\n",
     "        average_email_time += t\n",
     "    av_email_times_for_batches.append(average_email_time/n_samples)\n"
    ]