DEEP-LEARNING-FUNDAMENTALS4 MIN READ
Sort the Training Signals
Classify core training-loop terms into their role in neural network training.
Sort each term by its role in the training loop. Data Parameters Activations Gradients Metrics Input batch x Ground-truth label y Layer weight matrix Bias term ReLU output from hidden layer Raw logits before final probability conversion Gradient of loss with respect to weights Gradient norm used for clipping Validation accuracy F1 score reported in review
Read the full lesson
Sign up free — one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in