Partially Trained Neural Network Updates

Week 9 Progress Update     This week I spent a large amount of time preparing my code to be partially rewritten. I discovered that I actually have to use a slightly different input pipeline and that implementing this pipeline will require me to change a large part of the code. One of the things I often try to do before making big changes to code is clean it up so it is more readable and optimized. In cleaning up my code I changed my data processing code for the fifth time, this time including code that allowed me to make sure the images were valid as well  as processing the binary as python lists before converting it to tensors. I also add some checkpoints in my code to make sure everything was running properly. After all this I am about ready to edit my code and be done with this segment of the research, however I am a little behind task on what I was planning to accomplish, because of this I am planning on implementing only one or two visualization methods inste...

Week 8 Progress Update Progress is relatively slow, I debugged to the point where it has an output, unfortunately the output is erroneous. Along with this, the program outputs an error after running the optimization loop after only one run. I have been stuck on this problem for the past few days and if I can't seem to fix it by the end of next week I'm going to start looking at visualization algorithms and coming back to the actual CNN later.  Error given after first optimization (above), and output from first optimization (below)

Week 7 Progress Update         This week I debugged much of the data input system as well as the creation of the neural net architecture. The new slightly modified code is shown below, I just basically created a placeholder for x_batch so that it could initialize before I actually run the network. Other than that I made a few minor corrections across the code. The NN architecture

Week 6 Progress Update This week has again been very slow, I fixed my code for importing Cifar-10 which involved switching from a prebuilt TensorFlow library which handled Cifar-10 to custom code which imported the data using the TensorFlow data module (shown below). As with any module that I have not worked with before, it took a significant amount of time to learn and implement. I have moved on to debugging where I will probably be stuck for the remainder of this week and possibly further if any major problems arise, which they most likely will. The functions I used to import the Cifar 10 dataset

Week 5 Progress Update I spent much of this week making my CNN with Cifar-10 an image set that offers a lot of training data. I didn't get a whole lot done code wise this week because this was my main focus and I am not very well acquainted with the data pipeline system in Tensorflow. However I do have the beginning of processing code and I completed the code for optimizing the CNN which should work pretty well with my code for the rest of the model. Over the next week I will probably finish up my data processing code and start debugging the CNN. Data processing functions and variables (above) and optimization code (below)

Week 4 Progress Update         This week I spent the first few days doing more research into how to build a CNN in TensorFlow. I primarily used the TensorFlow example ( https://www.tensorflow.org/tutorials/images/deep_cnn ) and the corresponding documentation for this. Once I found I had a solid understanding of the basics of TensorFlow I began programming my own CNN, loosely based off of the TensorFlow example. I have not yet used it or chosen a dataset to test it with but I have created all of my helper functions and built the framework for the model as shown below.

Week 3 progress update         Most of the progress I made this week was just research, I focused on getting familiar with the TensorFlow framework. I made several neural networks that work in slightly different ways and using different data sets. I have not completely decided on a data set that I want to use but I am probably going to use Image Net as it provides a relatively large sample for each class as well as plenty of classes. During my research I also designed a small script for processing "pickled" data, which has been stored for use in python as well as a script for processing some particular data sets such as Mnist, both of these are shown below. I also did a bit of research on making a CNN in TensorFlow which I will start actually building over the course of the next week or two.