• Completed course and assignments on ML by Andrew NG on coursera.
• Read and studied 1st four chapters on Neural Networks and Deep Learning by Michael Nielsen.
• Learned and implemented basics of python, git and lua.
• Installed and implemented torch, itorch and loaded MNIST data.
• Understood, tried and practiced MNIST tutorial provided by Andrea Ferretti on RNDuja Blog.
• Tried to train data found on http://www.ee.surrey.ac.uk/CVSSP/demos/chars74k/ only to get an accuracy of 75 percent.
• Using a better data set on https://lvdmatten.github.io/software/code/wride.tar.gz/ and tweaking some parameters we improved the accuracy to 95%.
• Studied convolutional neural networks and their implementation on http://cs231n.github.io/convolutional-networks/
• Implemented convolutional neural network architecture on the old data set after filtering garbage data, achieving accuracy up to 99.1% on capital letters and digits.
• Using character segmentation code on MATLAB by Diego Barragan, Technical University of Loja, Ecuador, available at http://www.mathworks.com/matlabcentral/fileexchange/22922-image-segmentation---extractionfacilitating accepting images and printing text in them.
• Used graph plotting tools to show graphs of loss vs time and accuracy vs time.

We are really grateful to Programming Club, IIT Kanpur for selecting such fantastic project templates and assigning us such wonderful mentors to guide us in our project. We had started out with just a faint hope that at least we will learn something and most probably we will not be able to come up with anything close to a finished code. Through our mentor's guidance and adequate reading material, we made swift progress. We are happy that finally we have something close to finished to show and hope you will like it! We hope to continue our efforts in this field. Let's see how far this goes on!