From 19ca89c6b02dfaa3efbb6e94ffd1df52345e4ebd Mon Sep 17 00:00:00 2001
From: Patrick Nicodemus <patrick.nicodemus@pennmedicine.upenn.edu>
Date: Fri, 3 Jan 2025 13:00:47 -0500
Subject: [PATCH] Set an appropriate rho value.

---
 docs/notebooks/Example_5.ipynb | 41 ++++++++++++++++++++++++++--------
 1 file changed, 32 insertions(+), 9 deletions(-)

diff --git a/docs/notebooks/Example_5.ipynb b/docs/notebooks/Example_5.ipynb
index 6628f75..1581ca6 100644
--- a/docs/notebooks/Example_5.ipynb
+++ b/docs/notebooks/Example_5.ipynb
@@ -24,7 +24,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 1,
    "id": "b6b7e64d",
    "metadata": {},
    "outputs": [],
@@ -52,16 +52,39 @@
    "execution_count": 2,
    "id": "36765c5a",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(645, 100, 100)\n",
+      "10865.144685592724\n",
+      "Done first pass.\n",
+      "CPU times: user 2d 1h 5min 41s, sys: 50.3 s, total: 2d 1h 6min 32s\n",
+      "Wall time: 2h 28min 12s\n"
+     ]
+    }
+   ],
    "source": [
     "eps = 100.0\n",
     "UGW_dmat = UGW_multicore.ugw_armijo_pairwise(\n",
-    "    mass_kept = 0.90,\n",
+    "    mass_kept = 0.80,\n",
     "    eps=eps,\n",
-    "    dmats=join(bd,\"geodesic_100_icdm_50cells.csv\")\n",
+    "    dmats=join(bd,\"geodesic_100_icdm.csv\")\n",
     ")"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "0a14b977-2b08-44f5-8bed-6ab2822a899a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "np.save(\"UGW_dmat_mass_80_eps_100.npy\", UGW_dmat)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "f09b79aa",
@@ -73,7 +96,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 5,
    "id": "78e048c4",
    "metadata": {},
    "outputs": [
@@ -81,8 +104,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Accuracy: 0.7021276595744681\n",
-      "MCC:  0.4171682921141799\n"
+      "Accuracy: 0.5510204081632653\n",
+      "MCC:  0.4520340022810735\n"
      ]
     }
    ],
@@ -92,7 +115,7 @@
     "from sklearn.model_selection import LeaveOneOut, cross_val_score\n",
     "from cajal.utilities import cell_iterator_csv\n",
     "\n",
-    "cells, idcms = zip(*cell_iterator_csv(intracell_csv_loc=join(bd,\"geodesic_100_icdm_50cells.csv\")))\n",
+    "cells, idcms = zip(*cell_iterator_csv(intracell_csv_loc=join(bd,\"geodesic_100_icdm.csv\")))\n",
     "metadata = pd.read_csv(join(bd,'m1_patchseq_meta_data.csv'),sep='\\t',index_col='Cell').loc[pd.Series(cells)]\n",
     "RNA_family = metadata['RNA family']\n",
     "hq = RNA_family != 'low quality' # Filter down to the cells that have a well-defined RNA family.\n",
@@ -126,7 +149,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 6,
    "id": "46a4c5d8",
    "metadata": {},
    "outputs": [