Fix typo in LaTeX equation of loss function.

silverlining21 · Dec 27, 2016 · 194570b · 194570b
1 parent b7e729f
commit 194570b
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/cifar10_classical_vs_maxmin_baseline.ipynb b/cifar10_classical_vs_maxmin_baseline.ipynb
@@ -494,7 +494,7 @@
     "If you're unfamiliar with Keras, one needs to \"compile\" the model after its declaration. Here's where we set the target **Loss Function** and the **Optimizer** to use for Gradient Descent. The paper mentions usage of Stochastic Gradient Descent, but for the sake of faster convergence, we'll use the **Adam Optimizer**. Also for categorical targets, we'll use **Categorical Cross Entropy**. Both of these are pre implemented and provided by Keras.\n",
     "\n",
     "* **Categorical Cross Entropy: **\n",
-    "$$C=−\\frac{1}{n} \\sum [y ln(\\hat{y}) + (1 − y) ln(1 − \\hat{y})]$$  \n",
+    "$$C = - \\frac{1}{n} \\sum [y ln(\\hat{y}) + (1 - y) ln(1 - \\hat{y})]$$  \n",
     "where $y$ and $\\hat{y}$ would be one hot encoded vectors.  \n",
     "\n",
     "\n",