A Multilayer Perceptron (MLP) Neural Network is trained using Pytorch LIghtning library. CIFAR dataset is used to trian the neural network.
Different Experiments are performed and results are observed. Type of experiments and validation accuracy of network is given below:
Version 1: B_SIze: 32 H_Layers: 1 H Neurons: 512 Optim: SGD Sigmoid Val_Acc: 0.4706
Version 2: B_SIze: 32 H_Layers: 1 H Neurons:1512 Optim: SGD Sigmoid Val_Acc: 0.4626
Version 3: B_SIze: 32 H_Layers: 1 H Neurons:1512 Optim: SGD RELU Val_Acc: 0.5089
Version 4: B_SIze: 32 H_Layers: 1 H Neurons:1512 Optim: ADAM RELU Val_Acc: 0.5114
Version 5: B_SIze: 32 H_Layers: 1 H Neurons:1512 Optim: ADAM RELU Val_Acc: 0.5389 Augmentation
Version 6: B_SIze: 32 H_Layers: 2 H Neurons:1512 Optim: ADAM RELU Val_Acc: 0.5389 Augmentation