Exercises for the laboratory classes can be found under this git repository.
You can find e-portal course page under this link. Password to sign up will be handed out during the classes.
Grading rules
- exercises are graded during classes after which students receives a grade. This grade accounts both for the realised exercise as well as answering questions,
- after each laboratory class a report must be submitted,
- final grade is a weighted average calculated for all grades (mini-project grade is counted double),
- one absence is admissible, however an exercise for given laboratory class has to be submitted. It can be submitted in any next classes or during the last one. In justified cases, a greater number of absences is allowed in special cases, which are considered individually,
- last laboratory class is dedicated for redoing any miss laboratories,
- short test or so-called entry tests that allow you to participate in laboratories can take place,
- plagiarism as well as non-independent work are unacceptable. In the case of their occurrence, the laboratory exercise is graded with 2.0. In addition, it is possible to fail the course in the event of a gross violation of this point. The work can be verified with the help of anti-plagiarism systems,
- completed tasks should be submitted via the e-portal. Uploading of all tasks via eportal is mandatory. Failing to do so will lead to failing the laboratory exercise.
Eportal
TBD
Report
After each laboratory class a report should be prepared. The report should include following in order to ensure maximum mark:
- short description of the exercise at hand,
- essential code fragments which show how the exercise was solved,
- plots and graphs for conducted exercise showing achieved results, e.g. confusion matrix, test samples,
- solution of the homework (essential code, visualisation, simulation results),
- missing homework will result with penalty points for the report,
- late submission will result in penalty points for the report.
If laboratory class included 3 exercises for each exercise above description must be provided
Additional materials
Some additional material can be found below
The dataset comes as CSV file containing numerical data. It contains five columns. First four columns are considered to be features while the last, fifth column is considered to be a label (one of two classes).
Archive with sample images for training an image classifier.