Week 8
Hello again.
This week has been another fairly productive week.
For the dataset, I completely forgot last week to finalize the splitting of my dataset into the training and testing dataset, so I went ahead and did that with the 90% and 10% split I decided prior. I also reran all my code to place the datasets in the designated files.
The interval dataset also completely slipped my mind so I also put in some work in that, creating the dictionary needed for my code to convert the bigrams I have already created.
I mostly finished up the code for the gated recurrent neural network, and everything should be working fine although I will have to see when I get started actually testing the model with the datasets.
So I plan on starting my testing next week; we will see how much of it I can run. I do foresee many, many random bugs popping up that I will have to fix so who knows if I can do anything. I will probably just start with the simple 1 order markov model because that most likely will not have any problems and then the variable order markov model, then the neural network last since that will probably take the most work and fixing.
I started writing up the revisions for the literature review in the paper and the presentation; I am just trying to make sure that the sufficient word count and the sufficient content is still present in the paper and presentation.
I also researched some sources to better understand a gated recurrent neural network and also to write about in my paper and presentation.
A decent amount of work to do this week, so hopefully it goes well…