Sehr geehrte Damen und Herren,
ich soll im Auftrag meines Instituts (Institut für Satellitennavigation) einen Jupyter Kernel für Julia beantragen. Leider funktioniert der Link in Ihrem FAQ zum Beantragen eines Kernels nicht, weshalb ich Ihnen per Mail schreibe. Brauchen Sie bestimmte Informationen für die Beantragung und wie ist das weitere Vorgehen?
Mit freundlichen Grüßen
Erik Deinzer
Hi Nika Khosravi,
Handelt es sich beiden Jupyter Notebooks denn um Materialen/Übungsblätter für eine Vorlesung oder Praktikum?
In diesen Fall sollte ein neues Jupyter Profil angelegt werden, welches dann nach dem Login ausgewählt werden kann.
Den Prozess dazu haben wir hier beschrieben:
https://jupyter.pages.rwth-aachen.de/documentation/instructors/PreperationN…
Sollte es sich nur um eigene Jupyter Notebooks handeln, die Sie persönlich nutzen wollen, können Sie diese einfach per Drag & Drop oder Upload in den Datei Manager der Jupyer Lab Oberfläche hochladen.
Diese werden dort in ihrem persönlichen Home Verzeichnis dann persistent gespeichert.
Viele Grüße
Steffen Vogel
From: "Khosravi, Nika" <nika.khosravi(a)rwth-aachen.de>
Date: Monday, 5. October 2020 at 19:02
To: "jupyter(a)lists.rwth-aachen.de" <jupyter(a)lists.rwth-aachen.de>
Subject: [Jupyter] Jupyter Notebook Deployment
Liebe Jupyter-Interessierte,
Ich habe ein paar Fragen bezüglich des Hochladens von Jupyter Notebook Skripts und ich wäre Ihnen sehr dankbar, wenn Sie mir damit weiterhelfen könnten:
1. Welche Schritte sind für Hochladen von Jupyter Notebook Skripts durchzuführen?
2. Gibt es vielleicht eine ausführliche Anleitung dafür?
Auf Ihre Rückmeldung freue ich mich
Mit freundlichen Grüßen
Nika Khosravi
[HiWi an dem ACS und ITHE Institute]
Liebe Jupyter-Interessierte,
Ich habe ein paar Fragen bezüglich des Hochladens von Jupyter Notebook Skripts und ich wäre Ihnen sehr dankbar, wenn Sie mir damit weiterhelfen könnten:
1. Welche Schritte sind für Hochladen von Jupyter Notebook Skripts durchzuführen?
2. Gibt es vielleicht eine ausführliche Anleitung dafür?
Auf Ihre Rückmeldung freue ich mich
Mit freundlichen Grüßen
Nika Khosravi
[HiWi an dem ACS und ITHE Institute]
Dear all,
The grading of Jupyter Notebooks is a frequently requested feature by other lectures as well.
In our discussions with Harald Schnurbusch / CLS we identified this feature be part of a third phase for Jupyter-Moodle integration:
- Phase 1: Opening Jupyter individual Notebook via Hyperlinks from Moodle (done)
- Phase 2: Multi-tenancy / Access-control to Jupyter materials via Moodle course-membership (work-in-progress)
- Phase 3: Notebook grading. Sync points/grades back into Moodle
> Are there already recommended methods to do this or can you use something like nbgrader (https://nbgrader.readthedocs.io/en/stable/) on the RWTH JupyterHub?
There is currently no recommended method for implementing it supported by the ITC/CLS or the Jupyter Admins.
The RWTH JupyterHub cluster is currently in the progress of concluding its pilot phase.
As such, it will graduate to an official ITC service. The scope of this service is yet to be defined.
So we don’t know yet to which extend the integration of Moodle and Jupyter will be part of this initial service offering.
This will be decided by the IT center.
Unfortunately, nbgrader is not supported with the new JupyterLab frontend [1] which blocks us from using it with our cluster.
But I am optimistic that there are alternative methods to handle auto grading of notebooks.
But it is a challenging task on many layers:
- How do you define your tests?
- How do execute the unit-tests without allowing the students to fake the results?
- How do you link Moodle Test to a Jupyter Notebook?
Currently, my time and the resources of the ITC are limited to continue the development of the JupyterHub cluster in this direction.
But in my opinion we should use the Jupyter mailinglist/user-group as a place to develop such a method [2].
Once we found a common approach for auto grading and the moodle integration we can approach the ITC to incorporate it into the cluster.
There is also an active group around nbgrader currently working on JupyterLab support.
Best regards
Steffen Vogel
PS: To all the new participants in this conversation: Please consider joining the Jupyter User Mailing list [2]
[1] https://github.com/jupyter/nbgrader/issues/1006
[2] https://lists.rwth-aachen.de/postorius/lists/jupyter.lists.rwth-aachen.de/
---
Steffen Vogel
Research Associate
Tel: +49 241 80-49577
Institute for Automation of Complex Power Systems
E.ON Energy Research Center
RWTH Aachen University
Mathieustr. 10
52074 Aachen - Germany
http://www.acs.eonerc.rwth-aachen.de/
From: "Persike, Malte" <persike(a)cls.rwth-aachen.de>
Date: Tuesday, 8. September 2020 at 08:53
To: "Pieschel, Lars" <lars.pieschel(a)rwth-aachen.de>, "jupyter(a)lists.rwth-aachen.de" <jupyter(a)lists.rwth-aachen.de>
Cc: "welten(a)dbis.rwth-aachen.de" <welten(a)dbis.rwth-aachen.de>, "gleim(a)dbis.rwth-aachen.de" <gleim(a)dbis.rwth-aachen.de>, "Schnurbusch(a)lists-new.rwth-aachen.de" <Schnurbusch(a)lists-new.rwth-aachen.de>
Subject: [Jupyter] Re: Automatic evaluation of jupyter notebooks
Dear all,
FYI: my colleague, Harald Schnurbusch, has attended to the issue of auto-grading Jupyter Notebooks as part of the Moodle integration of Jupyter. I took the liberty to forward him Lars Pieschel‘s mail.
Best wishes
Malte Persike
Von: Pieschel, Lars <lars.pieschel(a)rwth-aachen.de>
Gesendet: Freitag, 4. September 2020 18:31
An: jupyter(a)lists.rwth-aachen.de
Cc: welten(a)dbis.rwth-aachen.de; gleim(a)dbis.rwth-aachen.de
Betreff: [Jupyter] Automatic evaluation of jupyter notebooks
Hello everyone,
I was tasked by the chair I5 to digitalize the Semantic Web Lecture using Jupyter notebooks but have some questions regarding the evaluation of said notebooks.
We plan on using the RWTH JupyterHub as a substitute for Exercises and would like to automatically grade the jupyter notebooks with unittest on the output. Are there already recommended methods to do this or can you use something like nbgrader (https://nbgrader.readthedocs.io/en/stable/) on the RWTH JupyterHub?
Apart from that, are there ways to integrate the submission and/or grading of jupyter notebooks into RWTHmoodle?
The workflow should look like this:
The students get their assignments in form of a jupyter notebook with gaps they have to fill.
After they are done working on their notebooks, they can submit them for grading.
The notebook is evaluated using unittests.
Ideally the notebooks can only be submitted before a certain deadline and after a notebook gets graded, the grade is automatically transmitted to RWTHmoodle.
Best regards and thank you in advance,
Lars Pieschel
Hello everyone,
I was tasked by the chair I5 to digitalize the Semantic Web Lecture using Jupyter notebooks but have some questions regarding the evaluation of said notebooks.
We plan on using the RWTH JupyterHub as a substitute for Exercises and would like to automatically grade the jupyter notebooks with unittest on the output. Are there already recommended methods to do this or can you use something like nbgrader (https://nbgrader.readthedocs.io/en/stable/) on the RWTH JupyterHub?
Apart from that, are there ways to integrate the submission and/or grading of jupyter notebooks into RWTHmoodle?
The workflow should look like this:
1. The students get their assignments in form of a jupyter notebook with gaps they have to fill.
2. After they are done working on their notebooks, they can submit them for grading.
3. The notebook is evaluated using unittests.
Ideally the notebooks can only be submitted before a certain deadline and after a notebook gets graded, the grade is automatically transmitted to RWTHmoodle.
Best regards and thank you in advance,
Lars Pieschel
Dear users of the RWTH JupyterHub cluster,
I would like to use this mail to remind you of the ongoing online evaluation of the cluster.
Unfortunately, the feedback has been relatively low until today.
This is partially due to some technical problems in the first week which have been resolved now.
Since the JupyterHub cluster is now transitioning into an official ITC service, your feedback and comments are valuable to further improve the service and offered features within the cluster.
To participate in the survey, please login to https://jupyter.rwth-aachen.de or use the following link:
https://www.global-assess.rwth-aachen.de/jupyter_eval/frontend/www/index.ph…
If possible, we would like also to ask you to ask your students to participate in the evaluation.
I also have to make a small correction to my first mail:
The Center for Teaching and Learning Services (CLS) which is overseeing the evaluation is reachable via the following mail address: evaluation(a)cls.rwth-aachen.de
Best regards,
Steffen Vogel
On 03.08.20, 11:56, "Vogel, Steffen" <svogel2(a)eonerc.rwth-aachen.de> wrote:
Dear user of the RWTH JupyterHub cluster,
Since the beginning of the current summer semester, students and faculty of the RWTH have had access to the interactive computer platform JupyterHub.
During your course(s) you could get to know the platform in the past months and gain versatile experience in using it.
The admins of the cluster together with Center for Teaching and Learning Services (CLS) designed a questionnaire to find out how you get along with the platform.
We would like you to share your experiences with us and answer the following questions. Completing the questionnaire takes between five and twelve minutes.
The questionnaire targets both lecturers and students.
So we would like to ask to participate yourself as well as to re-distribute this invitation to your students.
You can start the questionnaire by login into the cluster: https://jupyter.rwth-aachen.de
The Center for Teaching and Learning Services (CLS) is responsible for implementing this survey.
If you have any questions regarding the survey, please contact Mr. Michael Michels (evaluation(a)rwth-aachen.de).
With the end of the current semester, we aim to conclude the pilot phase and transition the platform during the next months into an official service provided by our IT Center.
We thank you very much and wish you all the best
Michael Michels (CLS)
Steffen Vogel (EONERC, ACS)
Institute for Automation of Complex Power Systems
Center for Teaching and Learning Services
IT Center
Dear user of the RWTH JupyterHub cluster,
Since the beginning of the current summer semester, students and faculty of the RWTH have had access to the interactive computer platform JupyterHub.
During your course(s) you could get to know the platform in the past months and gain versatile experience in using it.
The admins of the cluster together with Center for Teaching and Learning Services (CLS) designed a questionnaire to find out how you get along with the platform.
We would like you to share your experiences with us and answer the following questions. Completing the questionnaire takes between five and twelve minutes.
The questionnaire targets both lecturers and students.
So we would like to ask to participate yourself as well as to re-distribute this invitation to your students.
You can start the questionnaire by login into the cluster: https://jupyter.rwth-aachen.de
The Center for Teaching and Learning Services (CLS) is responsible for implementing this survey.
If you have any questions regarding the survey, please contact Mr. Michael Michels (evaluation(a)rwth-aachen.de).
With the end of the current semester, we aim to conclude the pilot phase and transition the platform during the next months into an official service provided by our IT Center.
We thank you very much and wish you all the best
Michael Michels (CLS)
Steffen Vogel (EONERC, ACS)
Institute for Automation of Complex Power Systems
Center for Teaching and Learning Services
IT Center
Hallo Herr Özmaden,
Ja, Julia wird bereits vom „Datascience“ Profil als Kernel unterstützt (Julia 1.3.1 + R + Python).
Dieses können Sie beim Starten direkt in der Profilauswahl wählen.
Viele Grüße
Steffen Vogel
From: "Özmaden, Can" <can.oezmaden(a)rwth-aachen.de>
Date: Friday, 31. July 2020 at 13:33
To: "jupyter(a)lists.rwth-aachen.de" <jupyter(a)lists.rwth-aachen.de>
Subject: [Jupyter] Jupyter Hub Julia
Sehr geehrtes Jupyter RWTH Team,
erstens, vielen Dank für diese tolle Initiative!
Ich wollte gerne wissen ob eine Unterstützung für Julia Notebooks geplant ist.
Danke!
Viele Grüße
Can Özmaden
Sehr geehrtes Jupyter RWTH Team,
erstens, vielen Dank für diese tolle Initiative!
Ich wollte gerne wissen ob eine Unterstützung für Julia Notebooks geplant ist.
Danke!
Viele Grüße
Can Özmaden
Liebe Juypter-Interessierte,
wir bereits angekündigt planen wir, dass die neue Jupyter-Instanz für das nächste Sommersemester einsetzbar ist. Wir sind nun bereit weitere Vorlesungen und Übungen mit in die Plattform aufzunehmen. Dazu benötigen wir von jedem Teilnehmer Informationen über den angebotenen Kurs. Bitte füllen Sie den Fragebogen (siehe unten) aus, wenn Sie die Jupyter-Instanz nutzen wollen und dies noch nicht getan haben. Senden Sie die Antworten bitte an jupyter-admin(a)lists.rwth-aachen.de und wir werden uns bei Ihnen melden. Eine Anleitung zur Einrichtung neuer Kurse finden Sie unter https://jupyter.pages.git-ce.rwth-aachen.de/documentation/instructors/Prepe…
Viele Grüße
Stefan Lankes
Profile name: Modelling and Simulation of Complex Power Systems
Short acronym for the course: MSP
Git repository: https://git.rwth-aachen.de/acs/teaching/msp-material
Docker image or Dockerfile: registry.git.rwth-aachen.de/acs/teaching/msp-material
Index Notebook (will be opened initially): /Index.ipynb
# In case the Git repo or registry are not public, we also need a GitLab deploy token
# Note: The token needs read-access for both the registry and repository.
# Note: The token is also required for internal repositories.
Username: gitlab+deploy-token-25
Token: XXXXXXXXXXXXXXX
# Under which license are your materials published?
License: MIT, GPL-2, ...?
# Should we always update/pull the latest state of the Git repo containing the Notebooks?
Update Git on login: yes/no
# Should we (re-)fetch new Docker images whenever a new profile is started?
Update Docker image on session start: yes/no
# Do you have estimates for ressource usage per User:
RAM: 250 MiB
Storage per User: 5 MiB
GPUs required: yes/no
Number of users per Semester: 100
Contact Person: Steffen Vogel <svogel2(a)eonerc.rwth-aachen.de>
Planned date of first use: SS2020 (1.4.2020)
Planned date of last use: indefinite
Frequency of usage: daily, weekly, once
Type of usage: In Class / At Home & Voluntary / Mandatory
Cycle of usage: every semester, summer/winter semester, only once?
--
Dr. rer. nat. Stefan Lankes
Institute for Automation of Complex Power Systems
E.ON Energy Research Center, RWTH Aachen University
Mathieustraße 10
52074 Aachen, Germany
Phone: +49-241-80-49740
