updates

Author: AlexandreAndr

notebooks/Notebook_1/Notebook_1a_Overview_and_Data_Representation.ipynb

@@ -359,6 +359,7 @@
     "- which values are rare\n",
     "\n",
     "You can think of a distribution as:\n",
+    "\n",
     "> A *model* of uncertainty.\n",
     "\n",
     "Later in this course we will:\n",

notebooks/Notebook_1/Notebook_1a_Overview_and_Data_Representation.py

@@ -211,6 +211,7 @@
 # - which values are rare
 #
 # You can think of a distribution as:
+#
 # > A *model* of uncertainty.
 #
 # Later in this course we will:

notebooks/Notebook_1/Notebook_1b_Noise_and_Modeling_Data.ipynb

@@ -136,6 +136,7 @@
     "- **Clustering**: group similar data points together (e.g. k-means).\n",
     "\n",
     "You can think of unsupervised learning as:\n",
+    "\n",
     "> “Let the data speak; find structure in $\\mathbf{x}$ itself.”\n",
     "\n",
     "In this module, we focus on **supervised learning and classification** to get familiar with modeling in `scikit-learn`.\n"

notebooks/Notebook_1/Notebook_1b_Noise_and_Modeling_Data.py

@@ -105,6 +105,7 @@
 # - **Clustering**: group similar data points together (e.g. k-means).
 #
 # You can think of unsupervised learning as:
+#
 # > “Let the data speak; find structure in $\mathbf{x}$ itself.”
 #
 # In this module, we focus on **supervised learning and classification** to get familiar with modeling in `scikit-learn`.

notebooks/Notebook_2/Notebook_2a_Building_a_Model.ipynb

@@ -790,6 +790,7 @@
     "  - sample size needs?\n",
     "\n",
     "Key takeaway:\n",
+    "\n",
     "> Noise does not \"erase\" signal, but it makes it harder to see. Uncertainty increases, p-values increase, and confidence intervals widen — so the chance of a **Type II error** (missing a real effect) goes up and power goes down. A non-significant result in a noisy, small study is not evidence that nothing is there.\n"
    ]
   },
@@ -861,6 +862,7 @@
     "   - Could some low BMI patients still have worse outcomes?\n",
     "\n",
     "Reflection:\n",
+    "\n",
     "> What question does the t-test answer in this context?\n",
     "\n",
     "---\n",
@@ -876,6 +878,7 @@
     "3. Does ANOVA tell us which groups differ?\n",
     "\n",
     "Reflection:\n",
+    "\n",
     "> Why might ANOVA say \"significant\" even if distributions overlap?\n",
     "\n",
     "---\n",

notebooks/Notebook_2/Notebook_2a_Building_a_Model.py

@@ -437,6 +437,7 @@ def power_widget(widgets):
 #   - sample size needs?
 #
 # Key takeaway:
+#
 # > Noise does not "erase" signal, but it makes it harder to see. Uncertainty increases, p-values increase, and confidence intervals widen — so the chance of a **Type II error** (missing a real effect) goes up and power goes down. A non-significant result in a noisy, small study is not evidence that nothing is there.
 #
 
@@ -490,6 +491,7 @@ def power_widget(widgets):
 #    - Could some low BMI patients still have worse outcomes?
 #
 # Reflection:
+#
 # > What question does the t-test answer in this context?
 #
 # ---
@@ -505,6 +507,7 @@ def power_widget(widgets):
 # 3. Does ANOVA tell us which groups differ?
 #
 # Reflection:
+#
 # > Why might ANOVA say "significant" even if distributions overlap?
 #
 # ---