The stopping criteria contains parameters that control when to end/stop training procedure.
- Training Goal: Training will be regarded as successful if the Loss value falls below this value during the duration of the training.
- The training goal is measured by cost function value for given Neural Network' weights and given training data inputs. If this value < training goal specified in stopping criteria (0.0001 for example), the training goal is reached.
- Training Epoch: Number of trials the Neural Network will try to find best solution
- When all training data inputs are fed into the Neural Network, it propagates through Neural Network layers to calculate Neural Network outputs; these outputs are then compared to training data targets to calculate cost function and errors that will be used to back propagate through Neural Network layers to adjust its weights to optimize the cost function. Each round of back propagation will increase the training epoch by 1.
- Batch Size: number of samples the system will load for training
- When training data-set becomes too big/large, feeding all training data into Neural Network during training process will cause performance issue due to restriction in computer speed and internal memory. To avoid that issue, the training data-set will be broken down into smaller sets that has the size defined in Batch Size.
- Max Fail Count: Number of full set of trials to run to try to find best solution
- When training goal, gradient goal are both not reached, but the training epoch reached limit, the Max Fails will increase by 1. The training will stop if the Max Fails value > the value specified in the stopping criteria settings (1 for example).
In short, the Neural Network training is stopped when:
- Training goal is reached.
- Max Fail Count is reached.
* Note: you can stop the training at any time, and proceed if you think the performance of the model is good enough. Stop criteria is just settings to automatically stop training.