EVA8_Assignment_2.5
Your code MUST be: well documented (via readme file on GitHub and comments in the code) must mention the data representation must mention your data generation strategy (basically the class/method you are using for random number generation) must mention how you have combined the two inputs (basically which layer you are combining) must mention how you are evaluating your results must mention "what" results you finally got and how did you evaluate your results must mention what loss function you picked and why! training MUST happen on the GPU Accuracy is not really important for the SUM
Data Representation is taking MNIST data in seaparate CNN Block Random Integer is one hot encoded and fed to the second last layer of the model architechture
Data generation strategy is to create a random integer and store the integer and the total in the training and testing set
Combining at the fully connected layer through concatenation , but then added another fully connected layer to store weights where network learns how to add
Cross Entropy for the block for MNIST Mean Squared Error for the addition block
Epoch 1: Loss = 0.1873, Accuracy = 0.9468 Epoch 2: Loss = 0.1163, Accuracy = 0.9644 Epoch 3: Loss = 0.0867, Accuracy = 0.9728 Epoch 4: Loss = 0.0786, Accuracy = 0.9751 Epoch 5: Loss = 0.0702, Accuracy = 0.9769 Epoch 6: Loss = 0.0609, Accuracy = 0.9804 Epoch 7: Loss = 0.0702, Accuracy = 0.9775 Epoch 8: Loss = 0.0484, Accuracy = 0.9829 Epoch 9: Loss = 0.0478, Accuracy = 0.9830 Epoch 10: Loss = 0.0478, Accuracy = 0.9839