This repository consists of an effective deep learning model to predict a student's grade point average of a particular institution A deep learning model, Bi-directional long short-term memory networks are the most effecient way to do the task of predicting a student's grade. The recurrent neural networks gave greater accuracy as compared to feed forward neural networks, as they have memory and take into consideration the consistency of the student performance. To predict the accuracy of the model, a mathematical method-root mean square error was used. The file - example.csv consists of marks and grades of 2000 students. The file - stud_pred_deep.py is the model that is developed to predict the grade point averages. Running the stud_pred_deep.py gives the following details : 1. number of student marks predicted with less than or equal to 5% error 2. no of student marks predicted with less than or equal to 10% error 3. no of student marks predicted with less than or equal to 20% error