1. MNIST Dataset
"Hello World" program for ML/AI/DL. MNIST is a 28x28 pixel B&W image. There are 60,000 samples for training and testing. I have attained 98% accuracy after tweaking network in CNN layers. Here is the image during excuetion on tensorflow-GPU.
2. CIFAR Dataset
The CIFAR-10 dataset consists of 60000 32x32 colour images in 10 classes, with 6000 images per class. There are 50000 training images and 10000 test images. It is available here on there official page : Dataset link. The model in the picture is trained for only 5 epochs for demo purpose, or else is trianed on 155 epochs. It took 69.3 minutes for training. To avoid overfitting keep the epochs ~170-200.
