Transfer and Deep Learning-Based Gurmukhi Handwritten Word Classification Model

Table 7

Training parameters’ results of VGG16 at different epochs and batch sizes.

Results	Training parameters
	Epoch = 100		Epoch = 40
	Batch size 20	Batch size 40	Batch size 20	Batch size 40
Training accuracy	0.7925	0.7832	0.7265	0.7212
Validation accuracy	0.8138	0.8056	0.7758	0.7563
Training loss	0.6274	0.6515	0.8382	0.8525
Validation loss	0.5484	0.5720	0.6908	0.7279