NN-Predict - Student Performance Prediction with Neural Networks
The Student Performance Prediction project is aimed at developing a deep learning model that can predict student performance based on multiple factors, such as hours studied, previous scores, extracurricular activities, sleep hours, and sample question papers practiced. By analyzing these features, the model can provide insights into a student's expected performance.
- Predicts student performance using a deep learning neural network.
- Utilizes features like hours studied, previous scores, and extracurricular activities.
- Provides predictions for educational insights and support.
- Numpy
- Pandas
- Tensorflow (Keras)
- Joblib
- Pickle
I just used this to master regression using Neural Networks. Heres what i learned:
- The activation function for the output layer in regression problems is 'linear' because the primary objective in regression tasks is to predict a continuous numerical value (e.g., real numbers) rather than class labels or probabilities.
- Input Shape = Number of features considered for prediction. In this case, 5 (Hours Studied, Previous Scores, Extracurricular Activities, Sleep Hours, Sample Question Papers Practiced).
- Hidden layer = 'relu'. ReLU is commonly used in hidden layers of neural networks for regression tasks due to its non-linearity, computational efficiency, sparsity, and reduced vanishing gradient problem.
Gideon Ogunbanjo