# How to interpret a drastic accuracy loss while training a neuronal net (CNN)?

How can one interpret a drastic accuracy loss after ~38 epochs? Maybe more dropout should be added to the CNN network?

(x-axis shows the number of epochs)

• Numeric stability issues + overfitting. Some neural activation, or some weights, continue to increase/decrease monotonically until some function (e.g. softmax) encounters a numerical stability issue. This could happen when your network is memorizing instead of properly learning.
For the above two reasons, you may want to print the min and max values for the activations and weights in each layer during your training. If something looks like exploding (i.e. reaching something like 10E+38) you need to check on that. Also monitor if any values become NaN or infinity.