Update index.md

index.md

@@ -43,7 +43,7 @@ Figure 4 shows the facial emotion distribution for training and testing datasets
 <p align="center"> <img src="Fig/5.png" alt="hi" class="inline"/> </p>
 <div align="center"> Figure 5. The evolution of accuracy and loss for training and validation dataset. </div>
 
-The Python software was used with the TensorFlow library to build the CNN model. The model was trained on a local machine using GPU for 100 epochs. The history of accuracy and loss are shown in Fig. 5. The accuracy of the two groups initially increases with the epoch and then reaches a plateau. The accuracy of the two groups is close to each other until the 10th epochs, where the accuracy of the validation dataset becomes stable while the accuracy of the training dataset keeps increasing. Similarly, the loss of the two groups decreases and reaches the bottom at the 10th epoch. Training and validation dataset have a similar magnitude of loss until the 10th epochs,  where the loss of the training dataset keeps decreasing while the loss of the validation dataset suddenly increases. These two observations indicate the model is overfitted.
+\The Python software was used with the TensorFlow library to build the CNN model. The model was trained on a local machine using GPU for 100 epochs. The history of accuracy and loss are shown in Fig. 5. The accuracy of the two groups initially increases with the epoch and then reaches a plateau. The accuracy of the two groups is close to each other until the 10th epochs, where the accuracy of the validation dataset becomes stable while the accuracy of the training dataset keeps increasing. Similarly, the loss of the two groups decreases and reaches the bottom at the 10th epoch. Training and validation dataset have a similar magnitude of loss until the 10th epochs,  where the loss of the training dataset keeps decreasing while the loss of the validation dataset suddenly increases. These two observations indicate the model is overfitted.
 
 <p align="center"> <img src="Fig/6.png" alt="hi" class="inline"/> </p>
 <div align="center"> Figure 6. Confusion matrix of the trained CNN model. (0=Angry, 1=Disgust, 2=Fear, 3=Happy, 4=Sad, 5=Surprise, 6=Neutral） </div>