Warning
This page is located in archive. Go to the latest version of this course pages. Go the latest version of this page.

Labs

The goal of the labs is to exercise the topics presented on lectures.

On some labs you will receive homework assignment, which are implementation of an algorithm or a method solving some interesting combinatorial optimisation problem. In all cases, the solutions to homework assignments are submitted to UploadSystem where they are automatically checked and evaluated. There are no strict deadlines, however, you will be penalised by -1 points for each week after deadline. Completing all homeworks successfully is a mandatory requirement for the assessment. Moreover, we encourage you to solve the homeworks since in the practical test you will use algorithms implemented for the homeworks.

Plan of the Labs

Week No. Title Notes  Handouts Other materials
1 Introduction, Gurobi installation - 01_gurobi.pdf 01_grading_and_rules.pdf gurobi_examples.zip
2 Semester Project, LP and ILP basics - Cocontest assignment Linear Programming, Duality (in Czech)
Convex optimisation, Linear Programming, Duality (in English)
Example source codes
3 ILP 1 HW1, deadline for choosing semester project 03_ilp.pdf Notes and issues with big-M,
Program skeletons
HW1 Public instances
Example source codes
4 ILP 2 - Unexpected applications - 04_ilp.pdf Game of Fivers,
Optimal adversarial examples to fool a DNN via ILP,
Solving Rubik's cube,
lab_codes_04.zip
5 Shortest Paths Consultation 05_stp.pdf Content-aware image resizing via STP,
czech_republic.txt,
Czech republic approximation (skeleton)
6 Network Flows HW2 06_flows.pdf Examples to practice,
hw2_public_instances.zip,
Pseudopolynomial time complexity
7 Minimum Cost Flows HW3 07_mcf.pdf hw3_public_instances.zip
8 Dynamic Programming CoContest consultation 08_dynamic.pdf -
9 Practical Test - - -
10 Traveling Salesman Problem HW4 10_tsp.pdf hw4_public_instances.zip, circle_approximation.py
11 consultation -
12 Tuesday Bratley's problem HW5 12_bratley.pdf hw5_public_instances.zip, Scheduling ZOO
12 Thursday class is canceled, Wednesday schedule -
13 Tuesday class is canceled, Wednesday schedule - -
13 Thursday Bratley's problem HW5 12_bratley.pdf hw5_public_instances.zip, Scheduling ZOO
14 Presentation of the semester project solutions Attendance is mandatory. Cocontest presentation -

Classroom computers

OS: Debian Linux 64b, select “DCE PXE” (without number) during booting

Login: username + main CTU password (the one you use for KOS)

Development environments: CLion (C++), IntelliJ (Java), PyCharm (Python), Visual Studio Code, GVim, Eclipse, Geany are installed. CLion, IntelliJ and PyCharm are installed in /opt and their license have to be activated via creating JetBrains account with faculty email. For interactive Python, you can use IPython.

courses/ko/labs.txt · Last modified: 2019/05/24 10:19 by novakan9