update

doc/src/week43/exercisesweek43.do.txt

@@ -134,12 +134,13 @@ Assume also that the bias of the output layer is zero and that the weights of th
 !et
 and multiplying with $\bm{a}_h$ gives the output
 !bt
-\bm{a}_o=\bm{w}_h^T\begin{bmatrix} 0 & 0 \\
+\bm{a}_o=\begin{bmatrix} 0 & 0 \\
                        1 & 0 \\
 		       1 & 0 \\
-		       2 & 1 \end{bmatrix}=\begin{bmatrix} 0 \\ 1 \\ 1 \\0\end{bmatrix},
+		       2 & 1 \end{bmatrix}\begin{bmatrix} 1 \\
+                       -2\end{bmatrix}=\begin{bmatrix} 0 \\ 1 \\ 1 \\0\end{bmatrix},
 !et
-the wanted result.
+the wanted result.  Pay attention to the dimensionalities as well.
 
 
 !split

doc/src/week43/week43.do.txt

@@ -61,6 +61,7 @@ cost function.
 
 
 
+
 Their inputs and outputs can be
 summarized using the following tables, first for the OR gate with
 inputs $x_1$ and $x_2$ and outputs $y$:
@@ -173,12 +174,13 @@ Assume also that the bias of the output layer is zero and that the weights of th
 !et
 and multiplying with $\bm{a}_h$ gives the output
 !bt
-\bm{a}_o=\bm{w}_h^T\begin{bmatrix} 0 & 0 \\
+\bm{a}_o=\begin{bmatrix} 0 & 0 \\
                        1 & 0 \\
 		       1 & 0 \\
-		       2 & 1 \end{bmatrix}=\begin{bmatrix} 0 \\ 1 \\ 1 \\0\end{bmatrix},
+		       2 & 1 \end{bmatrix}\begin{bmatrix} 1 \\
+                       -2\end{bmatrix}=\begin{bmatrix} 0 \\ 1 \\ 1 \\0\end{bmatrix},
 !et
-the wanted result.
+the wanted result.  Pay attention to the dimensionalities as well.
 
 
 !split
@@ -1321,8 +1323,6 @@ scores = logistic_regression.fit(X, yXOR, scheduler, epochs=1000)
 !ec
 Not bad, but the results depend strongly on the learning reate. Try different learning rates.
 
-
-
 !split
 ===== Lecture Thursday October 26 =====