Data Science II
Course Description
Catalog Description: This course offers the opportunity to learn advanced modeling and analysis techniques, and is intended for students who completed the course on Data Science I. A distinctive feature of this course is the opportunity to apply data-science skills on a semester long project based on real-world data. The demand for data scientists is increasing enormously these days, to the point that companies struggle finding candidates with the required skill set. After this course, students will be familiar with the most popular data science and machine learning techniques, ready to apply for a variety of data-science jobs in industry.
Extended Description: This course is the terminal design course for the undergraduate computer science degree and is the second semester of a two-semester course sequence. Students will work in project groups and apply their data analytics knowledge to the solution of a problem, providing a validated deliverable at the end of the course sequence. The intent of the second semester of the course is to give students an intensive experience in the process of: critical and technical analysis of designs subject to specifications and constraints, executing and validating a realization of the designed solution, communicating technical design details and project progress through written papers and oral presentation. A strong background in Python programming will be helpful in this course.
Prerequisites: COSC 3337 with a grade of C- or better.
Software
You can use the following link to install TensorFlow 2. However, I strongly recommend using Google Colab, which is also mentioned in this link. Google Colab is a Jupyter notebook environment that requires no setup to use and runs entirely in the cloud. Think Google Docs but for Jupyter notebooks. What's awesome about this is that it has everything installed for you (you need to use something that doesn't exist (rare), you can just download it by running pip install <package name here> in a cell. This will be helpful for your group projects so that everyone can work with and modify the same notebok and do things like leave comments, etc. When you have all of your code ready you can just download your .ipynb, open it in jupyter notebbok / anaconda as usualy, and convert it to .html or pdf. Also, Google Colab lets you use a free GPU so that your models can train faster if you decide to use a really large dataset and architecture for your project. Please feel free to email me if you have trouble using colab or the link above
Here's a YouTube video for further help in installing TensorFlow.