Portfolio Assignment Databases
Portfolio Exercise Databases:
Introduction
- Now it is time to get some coding done again :)
- This will be our first portfolio exercise,where you can practice your new skills.
Task
- Develop a Proof-of-Concept version of an application that is querying a database to come provide an output to the user.
- This can be for example:
- Selecting observations from database, performing prediction with a (beforehand fitted) SML model.
- Perform a UML procedure on observations queried from a database.
- Perform a semantic/similarity search for an user input, retrieve most similar docs from a database.
- The data used should be non-trivial (eg.: enough observations,´maybe multiple tables, different types of data…)
- The solution has to be self-contained. This can be done:
- Within a colab using for grad.io. (Hint: An option is to save the database on github, and then load it in the colab).)
- As a streamlit app (figure out how to make it self-contained).
- … (sky is the limit.)
Possible databases:
- SQL DB (eg. SQL-lite)
- NoSQL DB
- Document (eg. tinyDB)
- Vector (Eg. Faiss, Chroma)
Delivery
- Create a github repository (or use the existing one and adapt it)
- Provide a readme.md with brief description.
- Submission can be in groups up to 3.
- Deadline 24/3.
- Submit by sending an email with link to repo to Hamid (hamidb@business.aau.dk) with Daniel & Roman in cc. (dsh@…, roman@…)