Docker seminar

Please note that this session will be given a lecture in Japanese.


In this seminar, we will learn how to run a job in OCTOPUS by using Docker, which has been widely used in recent years, in a hands-on format.
We will learn MLP and CNN models that perform image classification by using Python framework (Tensorflow, PyTorch).
You can learn how to run jobs for training such models using Docker, without having to manually build environments such as Python frameworks, libraries, and CUDA (cudnn) which considering a version.



13:30 - 13:50 How to run a job at OCTOPUS by using Docker
Lecturer: Yasuhiro Watashiba, Associate Professor, Faculty of Engineering Division of Engineering Information Science, University of Fukui
13:50 - 14:30 Running the image classification model by Tensorflow, PyTorch by using Docker
Lecturer: Chonho Lee, Faculty of Information Science and Engineering, Okayama University of Science


Highly recommended for…

  • Interested in using Docker
  • Those who perform calculation using publicly available containers
  • Those who do not like troublesome preprocessing to build the environment although interested in deep learning


Important notice with regard to taking this seminar

Because of COVID‑19, this seminar is live-streamed online using online meeting tools.
Please make sure you have an internet access. We will contact you about how to connect separately.


The materials for the seminar

How to run a job at OCTOPUS by using Docker(lecture)
Tried to use Docker by OCTOPUS (hands-on)

Date : Dec 13, p.m. 1:30 - p.m. 2:30
Instructor: Chonho Lee (Professor, Okayama University of Science)
     Yasuhiro Watashiba (Associate Professor, University of Fukui)
Venue: Online
Type : Lecture
Quota: 20
Application deadline: Dec 12, noon