This project is a simple convolution neural network. We are training this neural network to classify CIFAR images.
We have imported tensorflow and matplotlib libraries and downloaded the dataset. Now we have added the convolution layers and given relu activation to it. We got paramaters based on our input. To know the accuracy, adam optimizer is provided and epochs count is set to 10. We can see it as epoch goes down further the accuracy improves and machine started learning efficiently over time.
We ran into challenges of what activation layer and optimizer to be choosen to get the best accuracy.
Glad to have got accuracy over 70%
I have learning how to distribute train and test data and how it can optimized well.
I can work on more advanced projects to explore artifical intelligence deeply.
Log in or sign up for Devpost to join the conversation.