From fbc8164f1a0bb9653cc532c25a0868d17206e735 Mon Sep 17 00:00:00 2001
From: Henry Webel <henry.webel@cpr.ku.dk>
Date: Mon, 30 Oct 2023 16:12:20 +0000
Subject: [PATCH] :bug: add dependencies specific to notebook

- pandas 2.0 requires numeric values for mean aggregation
---
 tutorial/log_reg.ipynb | 12 +++++++++++-
 tutorial/log_reg.py    |  5 ++++-
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/tutorial/log_reg.ipynb b/tutorial/log_reg.ipynb
index 6c3fb91..ba725cd 100644
--- a/tutorial/log_reg.ipynb
+++ b/tutorial/log_reg.ipynb
@@ -14,6 +14,16 @@
     "> Predict Alzheimer disease based on proteomics measurements."
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "cd90b1b6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!pip install njab heatmapz openpyxl "
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -672,7 +682,7 @@
     "    return_train_score=True,\n",
     "    # fit_params=dict(sample_weight=weights)\n",
     ")\n",
-    "cv_feat = cv_feat.groupby('n_features').agg(['mean', 'std'])\n",
+    "cv_feat = cv_feat.drop('test_case', axis=1).groupby('n_features').agg(['mean', 'std'])\n",
     "cv_feat"
    ]
   },
diff --git a/tutorial/log_reg.py b/tutorial/log_reg.py
index 59d02bc..03e0eb8 100644
--- a/tutorial/log_reg.py
+++ b/tutorial/log_reg.py
@@ -22,6 +22,9 @@
 #
 # > Predict Alzheimer disease based on proteomics measurements.
 
+# %%
+# !pip install njab heatmapz openpyxl 
+
 # %%
 import itertools
 import logging
@@ -366,7 +369,7 @@
     return_train_score=True,
     # fit_params=dict(sample_weight=weights)
 )
-cv_feat = cv_feat.groupby('n_features').agg(['mean', 'std'])
+cv_feat = cv_feat.drop('test_case', axis=1).groupby('n_features').agg(['mean', 'std'])
 cv_feat
 
 # %% [markdown]