In this work the paper "Facial Expression Recognition Using Weighted Mixture Deep Neural Network Based on Double-Channel Facial Images" is replicated. In this paper facial expression recognition problem (which is a classification problem) is solved using a deep learning model. As a start the previous works on this problem was searched. The typical common approach to this problem is using some machine learning models with hand crafted features. In this work a weighted deep neural net model is used. The model has two parts. First part is a VGG-16(similar to original VGG) like model that takes a gray-scale facial image as an input. Second part is a CNN with 2 layers that takes the LBP encoded version of the facial image that is the input of the VGG-16 part.After these two parts take their own inputs and after some additional layers their input is weighted. The model outputs its prediction for the facial image. There are 6 categories in the data set which are sad,happy,angry and etc.