Weekly plans and update for week 45
Dear All, we hope all is well with project 2. The deadline is approaching and as we wrote in a previous post on Piazza, devilry does not allow us to set a soft deadline. It means that we would like to propose the following procedure:
1) upload to devilry.ifi.uio.no your GitHub/GitLab or similar link where you have your report, codes and data files. Make sure you have a Readme file which explains where we find the material. Structuring the repo in folders like Report, Codes etc also helps with respect to the readability. Do this by the deadline Friday 8 at 23.59!!
2) Then, in case things go wrong or you need some more time to brush up things, make sure you upload (Git gives a timestamp) your report and codes not later than Monday the 11th of November at noon (12pm) to your Github/Gitlab or similar repository.
3) For those of you who have been sick or have other legal and approved reasons, we want you also to upload your GitHub/Gitlab or similar link by the deadline November 8 at 23.59. The final deadline for uploading your report to your repository is in this case Wednesday the 13th at 23.59.
If something is unclear, please let us know asap.
Else, this Thursday we will wrap up our discussion of Trees and Forests by looking at Boosting methods, a cool topic with methods like AdaBoost (adapative boosting), Gradient Boosting and XGBoost (extreme gradient boosting, I love these names!). The textbook by Hastie is pretty good at this, see chapter 10.2-10.10.
On Friday we start with one of our final supervised learning methods, Support Vector Machines, covered by the slides and Hastie et al chapter 12.1-12.4. Geron's text chapter 5 is also pretty good, with lots of examples.
Finally, and sorry for this long mail, we are gradually punching in the feedback on project 1.
Next week, project 3 will be available on Monday (the 11th). We though then of having a small workshop on Friday the 15th where you could present possible data cases you would like to explore in project 3.
Project three will have some proposed data sets as well as additional topics (solving eigenvalue problems and partial differential equations with neural networks) where you are free to use existing libraries like tensorflow and scikit-learn.
The last two weeks of the semester will most likely be devoted to Bayesian statistics and Bayesian Neural Networks, a hot topic.
Best wishes,
Hanna, Lucas, Morten, Stian and ?yvind.