obsidian 24-10-10 12:02:40

Affected files:
1.2.Regression.with.Multiple.Input.Variables.md

Author: xrxfxt

1.2.Regression.with.Multiple.Input.Variables.md

@@ -1,13 +1,10 @@
-### C1_W2: Regression with Multiple Input Variables
+# C1_W2: Regression with Multiple Input Variables
 
-This week, you'll extend linear regression to handle multiple input features.
-You'll also learn some methods for improving your model's training and performance, such as _vectorization_, _feature scaling_, _feature engineering_ and _polynomial regression_. 
-At the end of the week, you'll get to practice
-implementing linear regression in code.
+This week, you'll extend linear regression to handle multiple input features. You'll also learn some methods for improving your model's training and performance, such as _vectorization_, _feature scaling_, _feature engineering_ and _polynomial regression_.  At the end of the week, you'll get to practice implementing linear regression in code.
 
-#### C1_W2_M1 Multiple Linear Regression
+## C1_W2_M1 Multiple Linear Regression
 
-##### C1_W2_M1_1 Multiple features
+### C1_W2_M1_1 Multiple features
 
 ![](/img/1.2.1.1.multiple.features.png)
 -$\vec{x}^{(i)}$= __vector__ of 4 parameters for$i^{th}$row
@@ -27,13 +24,13 @@ implementing linear regression in code.
 - this is __multiple linear regression__
   - __Not__ _multivariate regression_
 
-###### Quiz
+#### Quiz
 
 In the training set below (see slide: C1_W2_M1_1 Multiple features), what is$x_{1}^{(4)} $?
 
 <details><summary>Ans</summary>852</details>
 
-##### C1_W2_M1_2 Vectorization part 1
+### C1_W2_M1_2 Vectorization part 1
 
 Learning to write __vectorized code__ allows you to take advantage of modern
 numberical linear algebra libraries, as well as maybe GPU hardware.
@@ -46,7 +43,7 @@ numberical linear algebra libraries, as well as maybe GPU hardware.
 - Vectorization has 2 benefits: _concise and efficient_
 - `np.dot` can use parallel hardware
 
-##### C1_W2_M1_3 Vectorization part 2
+### C1_W2_M1_3 Vectorization part 2
 
 How does vectorized algorithm works...
 
@@ -58,14 +55,14 @@ How does vectorized algorithm works...
 
 ![](/img/i1.2.1.3.gradient.descent.png)
 
-##### C1_W2_Lab01: Python Numpy Vectorization
+### C1_W2_Lab01: Python Numpy Vectorization
 
 - [Coursera](https://www.coursera.org/learn/machine-learning/ungradedLab/zadmO/optional-lab-python-numpy-and-vectorization/lab#?path=%2Fnotebooks%2FC1_W2_Lab01_Python_Numpy_Vectorization_Soln.ipynb)
 - [Local](/code/C1_W2_Lab01_Python_Numpy_Vectorization_Soln.ipynb)
   -$a \cdot b$returns a scalar
   - e.g.$[1, 2, 3, 4] \cdot [-1, 4, 3, 2] = 24 $
 
-##### C1_W2_M1_4 Gradient descent for multiple linear regression
+### C1_W2_M1_4 Gradient descent for multiple linear regression
 
 ![](/img/1.2.1.4.gradient.descent.png)
 
@@ -76,12 +73,12 @@ How does vectorized algorithm works...
 ![](/img/1.2.1.4.normal.equation.png)
 - __Normal Equation__
 
-##### C1_W2_Lab02: Muliple linear regression
+### C1_W2_Lab02: Muliple linear regression
 
 - [Optional Lab: Multiple linear regression | Coursera](https://www.coursera.org/learn/machine-learning/ungradedLab/7GEJh/optional-lab-multiple-linear-regression/lab)
 - [Local](/code/C1_W2_Lab02_Multiple_Variable_Soln.ipynb)
 
-#### Quiz: Multiple linear regression
+## Quiz: Multiple linear regression
 
 1. In the training set below, what is$x_4^{(3)} $?
 
@@ -104,9 +101,9 @@ How does vectorized algorithm works...
 
 <details><summary>Ans</summary>30, 4, F</details>
 
-### C1_W2_M2 Gradient Descent in Practice
+# C1_W2_M2 Gradient Descent in Practice
 
-#### C1_W2_M2_01 Feature scaling part 1
+## C1_W2_M2_01 Feature scaling part 1
 
 ![](/img/1.2.2.01.values.png)
 - Use __Feature Scaling__ to enable gradient descent to run faster
@@ -122,7 +119,7 @@ How does vectorized algorithm works...
 
 :bulb: We can __speed up gradient descent by scaling our features__
 
-#### C1_W2_M2_02 Feature scaling part 2
+## C1_W2_M2_02 Feature scaling part 2
 
 ![](/img/1.2.2.02.scale.png)
 - scale by dividing$x_i^{(j)} / \max_x $
@@ -138,29 +135,29 @@ How does vectorized algorithm works...
 - but the range is ok if it's relatively close
 - rescale if range is too large or too small
 
-##### Quiz:
+### Quiz:
 
 Which of the following is a valid step used during feature scaling? (see bedrooms vs size scatterplot)
 - [ ] Multiply each value by the maximum value for that feature
 - [ ] Divide each value by the maximum value for that feature
 
 <details><summary>Ans</summary>2</details>
 
-#### C1_W2_M2_03 Checking gradient descent for convergence
+## C1_W2_M2_03 Checking gradient descent for convergence
 
 ![](/img/1.2.2.03.alpha.png)
 - We can choose$\alpha $
 
 ![](/img/)
 - Want to minimize _cost function_ $\min\limits_{\vec{w}, b} J(\vec{w}, b)$
 
-#### C1_W2_M2_04 Choosing the learning rate
-#### C1_W2_M2_05 Optional Lab: Feature scaling and learning rate
-#### C1_W2_M2_06 Feature engineering
-#### C1_W2_M2_07 Polynomial regression
-#### C1_W2_M2_08 Optional lab: Feature engineering and Polynomial regression
-#### C1_W2_M2_09 Optional lab: Linear regression with scikit-learn
-#### C1_W2_M2_10 Practice quiz: Gradient descent in practice
-#### C1_W2_M2_11 Week 2 practice lab: Linear regression
+## C1_W2_M2_04 Choosing the learning rate
+## C1_W2_M2_05 Optional Lab: Feature scaling and learning rate
+## C1_W2_M2_06 Feature engineering
+## C1_W2_M2_07 Polynomial regression
+## C1_W2_M2_08 Optional lab: Feature engineering and Polynomial regression
+## C1_W2_M2_09 Optional lab: Linear regression with scikit-learn
+## C1_W2_M2_10 Practice quiz: Gradient descent in practice
+## C1_W2_M2_11 Week 2 practice lab: Linear regression
 ![](Screenshot%202024-10-09%20180220.png)
 ![](Screenshot%202024-10-09%20180317%201.png)