diff --git a/dashboard/__pycache__/filtered.cpython-312.pyc b/dashboard/__pycache__/filtered.cpython-312.pyc
index 9657b97517da14facffd3aea1e733af5a73f1f1b..1c3f806dd16e292cf1216e6b6f536e97ea0e5287 100644
Binary files a/dashboard/__pycache__/filtered.cpython-312.pyc and b/dashboard/__pycache__/filtered.cpython-312.pyc differ
diff --git a/dashboard/__pycache__/overview.cpython-312.pyc b/dashboard/__pycache__/overview.cpython-312.pyc
index dc04ca1fb621572c768fa61d9b0981e5ad646759..6a3c8a7fa4f9d18dc0887a8e0e1c917e88f13173 100644
Binary files a/dashboard/__pycache__/overview.cpython-312.pyc and b/dashboard/__pycache__/overview.cpython-312.pyc differ
diff --git a/dashboard/__pycache__/visuals.cpython-312.pyc b/dashboard/__pycache__/visuals.cpython-312.pyc
index 73ef74535482f388e96a20731e3f7e66e4fd4f40..bb606234b188a82a8c8f031487771e7597c1fc90 100644
Binary files a/dashboard/__pycache__/visuals.cpython-312.pyc and b/dashboard/__pycache__/visuals.cpython-312.pyc differ
diff --git a/dashboard/categorical_f.png b/dashboard/categorical_f.png
new file mode 100644
index 0000000000000000000000000000000000000000..eb26e7464a30604efede8a9f305d20bff81be71f
Binary files /dev/null and b/dashboard/categorical_f.png differ
diff --git a/dashboard/filtered.py b/dashboard/filtered.py
index b04db93eaa2d4cc1462e8b343e4c14071a61b192..4026c58c8ae065e287528bb0920fb8dd40d4e448 100644
--- a/dashboard/filtered.py
+++ b/dashboard/filtered.py
@@ -4,7 +4,7 @@ import numpy as np
 import matplotlib.pyplot as plt
 import seaborn as sns
 import pickle
-from visuals import plot_corr_matrix, plot_feature_distributions_by_diagnosis
+from visuals import plot_corr_matrix
 
 df_filtered = pd.read_csv("filtered_selected_features.csv")
 df_full = pd.read_csv("alzheimers_disease_data.csv")  # assuming this is your original dataset
@@ -30,8 +30,5 @@ def selected_data():
         st.pyplot(fig)
 
     st.subheader(" Feature Distributions by Diagnosis (Filtered Dataset)")
-    with st.expander("Show all feature distribution plots"):
-        plots = plot_feature_distributions_by_diagnosis(df_filtered)
-        for fig in plots:
-            st.pyplot(fig)
-
+    st.image(image= "categorical_f.png")
+    st.image(image="numerical_f.png")
\ No newline at end of file
diff --git a/dashboard/numerical_f.png b/dashboard/numerical_f.png
new file mode 100644
index 0000000000000000000000000000000000000000..d62812212e824ad060a5ff1bd89739da7e566f3a
Binary files /dev/null and b/dashboard/numerical_f.png differ
diff --git a/dashboard/overview.py b/dashboard/overview.py
index f858942d665f33ae83afe03fd55183b34dd6d28c..993f4e695205fa89ee2b59767e486f8090071961 100644
--- a/dashboard/overview.py
+++ b/dashboard/overview.py
@@ -5,7 +5,7 @@ import numpy as np
 import matplotlib.pyplot as plt
 import seaborn as sns
 import pickle
-from visuals import load_full_data, create_grouped_features, plot_correlation_with_diagnosis, plot_distribution, plot_corr_matrix, plot_feature_distributions_by_diagnosis
+from visuals import load_full_data, create_grouped_features, plot_correlation_with_diagnosis, plot_distribution, plot_corr_matrix
 
 df_filtered = pd.read_csv("filtered_selected_features.csv")
 df_full = pd.read_csv("alzheimers_disease_data.csv")  # assuming this is your original dataset
diff --git a/dashboard/visuals.py b/dashboard/visuals.py
index 32c4acfc450a0d43fda855202c9c0394e3391e76..97a85c0aa9e3a69b101821abea062a19abd9e12c 100644
--- a/dashboard/visuals.py
+++ b/dashboard/visuals.py
@@ -50,22 +50,3 @@ def plot_corr_matrix(df):
     ax.set_title("Correlation Matrix")
     return fig
 
-def plot_feature_distributions_by_diagnosis(df):
-    # Identify features (everything except 'Diagnosis')
-    features = [col for col in df.columns if col != "Diagnosis"]
-
-    plots = []
-    for feature in features:
-        fig, ax = plt.subplots(figsize=(8, 5))
-        
-        if df[feature].nunique() > 10:
-            # Assume it's continuous — use violinplot
-            sns.violinplot(x='Diagnosis', y=feature, data=df, ax=ax, palette='Set2')
-        else:
-            # Categorical or few unique values — use countplot
-            sns.countplot(x=feature, hue='Diagnosis', data=df, ax=ax, palette='Set2')
-
-        ax.set_title(f"{feature} Distribution by Diagnosis")
-        plots.append(fig)
-    
-    return plots
\ No newline at end of file
diff --git a/models.ipynb b/models.ipynb
index 4a53da3fdf2918489445c2a258e9897b5b471f94..1f6421d8fb5e91b1edf8fab34111ef1405a5e8a8 100644
--- a/models.ipynb
+++ b/models.ipynb
@@ -832,6 +832,66 @@
     "plt.show()\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1500x500 with 3 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "numerical_cols = ['FunctionalAssessment', 'ADL', 'MMSE']\n",
+    "\n",
+    "plt.figure(figsize=(15, 5))\n",
+    "for i, col in enumerate(numerical_cols):\n",
+    "    plt.subplot(1, 3, i + 1)\n",
+    "    sns.boxplot(x= 'Diagnosis', y= col, data = df_filtered)\n",
+    "    plt.title(f'{col} by Diagnosis')\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.savefig(\"numerical_f.png\", dpi=300)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x400 with 3 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "categorical_cols = ['Gender', 'BehavioralProblems', 'MemoryComplaints']\n",
+    "\n",
+    "plt.figure(figsize=(12, 4))\n",
+    "for i, col in enumerate(categorical_cols):\n",
+    "    plt.subplot(1, 3, i+1)\n",
+    "    sns.countplot(x=col, hue='Diagnosis', data=df_filtered, palette='Set1')\n",
+    "    plt.title(f'{col} vs Diagnosis')\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.savefig(\"categorical_f.png\", dpi=300)\n",
+    "plt.show()"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -1541,6 +1601,13 @@
    "source": [
     "metrics"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {