Week 5
Task List:
- determine which dataset to use
- read the comments on kaggle and see they posted as their started
- learn plotly to display the graphs
- look at roadmap (updated)
- read rashaim’s paper.
- later: look for code that is python based - tensorflow, keras
Challenges:
My mentors told me to look into datasets on kaggle that I was interested in and read through them to find how the currently posted solutions had performed to see if they could be used as a starting point for my research. I had some struggles remembering how to install dependecies from within the notebook and how to use bash commands to clean data, as I learned in the “Python for Data Science” course when I was practicing using what I had learned on real datasets.
I started reading the plotly documetation and went through tutorials and took notes in my jupyter notebook. The tutorial post I was following was using an old version on plotly, so I had trouble importing the libraries that had apparently been deprecated. I also couldn’t figure out how to embed the graphs into the notebook, but an html was generated. During my meeting with Dr Bein, I found out that I had attempted to learn plotly in more detail than was required of me. I was trying to understand all the different types of graphs instead of just the ones that would be useful to me during my research.
Resources used this week:
- kaggle datasets
- plotly documentation