【Method】Batch Size in Deep Learning Dec 26, 2018 https://blog.csdn.net/qq_27923041/article/details/74925057 Full Batch Learning epoch: 300 earlystopping: 20 shuffle=True training samples : model parameters = 10:1 kernel_regularizer Look at over-fitting val_loss ↓ and val_acc ↓ Please enable JavaScript to view the comments powered by Disqus.