Update pages/ML vs DL.py

pages/ML vs DL.py

@@ -13,13 +13,13 @@ st.markdown("""
     .title {
         text-align: center;
         font-size: 2.5rem;
-        color: #4CAF50;
+        color: black;
         margin-bottom: 10px;
     }
     .subtitle {
         text-align: center;
         font-size: 1.2rem;
-        color: #333;
+        color: violet;
         margin-bottom: 30px;
     }
     .table-container {
@@ -66,40 +66,45 @@ html_table = """
         <thead>
             <tr>
                 <th>Aspect</th>
-                <th>Machine Learning</th>
-                <th>Deep Learning</th>
+                <th>**Machine Learning**</th>
+                <th>**Deep Learning**</th>
             </tr>
         </thead>
         <tbody>
             <tr>
                 <td>Definition</td>
-                <td>Uses algorithms to parse data, learn from it, and make predictions.</td>
-                <td>A subset of ML focused on neural networks with many layers.</td>
+                <td>Machine Learning is a tool which needs statistical concepts to copy / mimic the learning ability in natural intelligence </td>
+                <td>Deep Learning is a tool which needs logical structure known as neural network to copy / mimic the learning ability in natural intelligence</td>
             </tr>
             <tr>
                 <td>Data Dependency</td>
-                <td>Performs well with structured and smaller datasets.</td>
-                <td>Requires large amounts of unstructured data to perform well.</td>
+                <td>ML performs well with structured data (**tabular data**) and smaller datasets</td>
+                <td>DL is hungry of data as it requires large amounts of unstructured data and also structured data to perform well</td>
             </tr>
             <tr>
-                <td>Feature Engineering</td>
-                <td>Requires manual feature extraction by domain experts.</td>
-                <td>Automates feature extraction using neural networks.</td>
+                <td>Performance</td>
+                <td>ML have treshold as the data increases the performance becomes stable</td>
+                <td>DL performance increases as the data increases because DL is hungry of data/td>
+            </tr>
+            <tr>
+                <td>Memory Management</td>
+                <td>ML memory uasage is less as it uses less data</td>
+                <td>DL memory usage is large as it has huge data</td>
             </tr>
             <tr>
                 <td>Hardware Requirements</td>
-                <td>Works on standard CPUs; lower hardware demands.</td>
-                <td>Requires GPUs/TPUs for efficient computation.</td>
+                <td>ML works on standard CPUs; lower hardware demands</td>
+                <td>DL requires GPUs for efficient computation.</td>
             </tr>
             <tr>
                 <td>Interpretability</td>
-                <td>More interpretable and explainable results.</td>
-                <td>Less interpretable due to complex neural networks.</td>
+                <td>ML is more interpretable as it works on smaller datasets </td>
+                <td>DL is less interpretable as it works on complex neural networks</td>
             </tr>
             <tr>
                 <td>Training Time</td>
-                <td>Relatively faster to train models.</td>
-                <td>Training can take significantly longer.</td>
+                <td>ML is relatively faster to train models as it uses less data</td>
+                <td>DL training can take significantly longer</td>
             </tr>
         </tbody>
     </table>