# DeepLearning540 Instruction Briefing

present: David Brunner, Jonas Rübenach, Sascha Diefenbacher, Lucas Wiens, Tobias Lösche, Peter Steinbach

## Say Hello

- Guten Tag
- Helene says "Hi"
- Moin!
- hi
- Hello there! General Kenobi!
- :smirk:

## Agenda

- mini intro to [hackmd.io](https://hackmd.io) (use your github account for getting access)
- introduction to learning material
- prepare hackmd documents for each group

## Introduction fo learning material

- share [website](https://deeplearning540.github.io/) with learners, ideally after lesson 1
- Lesson Design: information about agenda of each lesson
- first learners watch video
- tip for learners: open notebook and make notes while watching (starting in the 2nd lesson)
- then they discuss the "check-your-learning" questions as a team within zoom
- these questions can be found on the website
- please, put the answers into hackmd document
- ask learner to put +1 to correct answer (only one is correct) and then to comment on their choice
- important: if there are wrong answers DISCUSS why they chose this answer and help them to find correct answer (don't use word 'wrong')
- verbalise why you think the other answer is correct
- this leads to better learning
- make at least one of the check-your-learning exercises to build the team to support each other and to speak openly about misunderstandings
- keep the time in mind and (if necessary) interrupt discussion
- goal of these questions: learners get a chance to see if they understood the lessons content
- there are always 2 versions of a notebook. here is an example for lesson 01:
- there is the `lesson.ipynb` which is typically located in the lesson repo, i.e. https://github.com/deeplearning540/lesson01/blob/main/lesson.ipynb (this only contains half of the content)
- the full notebook (which also is the script for the video), is located elsewhere: https://github.com/deeplearning540/deeplearning540.github.io/blob/main/source/lesson01/script.ipynb
- both scripts are linked from the lesson page under the `Content` heading: https://deeplearning540.github.io/lesson01/content.html#content

---

**Demo Question**:

NaN stands for not-a-number and is commonly met in data science workflows if …

1. Input files contain string values in a column
+1
2. Computational Problems occurred, like computing the square root of a negative number
+1 +1 +1
3. Data could not be parsed correctly when reading input files into memory
+1

4. Predictions for unseen data are off the charts

- there was some discussion if answer 2 is also correct?
- in the context of the video, answer 3 is the most likely correct one
- in other contexts than loading data with pandas answer 2 is also correct

---

After the questions:
- then the practical exercises start
- every motivated learner should do these three steps:
- watch video
- answer check-yourself-question
- do exercise
- learners might be faster or slower
- is there enough time left for the exercise? if not: skip it
- video skript is also available on the website, so leaners have excess to the code
- give a fixed time frame for exercise and then see how far everyone got
- if you are not able to answer a question: no problem!!! be honest and post it into the Mattermost channel, so it stays not ananswered

## Questions
How does the course start on monday?
- short introduction for everyone, then group work
- no breakout rooms, but extra zoom sessions for each group
- central zoom session for everyone will always be open to be entered for questions
- additionally the Mattermost channel is a place to post questions
- Peter + Helene will be available to answer them

What should go into hackmd?
- check-your-learning questions + discussion notes
- mini agenda for group work
- links to the video and information about further exercises
- place to document questions/problems

Do we directly start with lesson 2 when we are finished with lesson 1?
- yes, there is no coming back (planned) to the main zoom session inbetween

Are there two different notebooks for the script and for the learners?
- yes, the learner get a notebook without the code, that they can use to take notes and make the exercise
- the other notebook is the one from the video script with everything in it which can be shared if you could not finish with everything in time

Should lerners use Google collab?
- they can choose themselves
- advantage of google collab: everinments are all available
- you are not responsible to make their local environments running

---

(Template for Team HackMD Documents - copy and paste the text below to the hackmd document of your team)

# Introduction to Data Science, ML and Deep Learning - Deeplearning540

Mar 1 - 3, 2021 @ 2nd Terascale School for Machine Learning, https://indico.desy.de/event/28296/

Team XX

**Important:** You can use this hackmd pad to take notes together, identify key questions, and document progress. Please be constructive, inclusive and positive in your communication with your peers.

## House Keeping

Some Important links to not get lost:

### Video Conferencing

- the main zoom room is here:
- our team's zoom room is here:

### Staying in Touch

- the main mattermost channel:
- our team's mattermost channel

## Learning

Each lesson always follows the same structure and is expected to last about 1h.

1. learners watch the video :cinema:
2. learners answer at least one check-your-learning questions as a team (at best in a hackmd document) :heavy_check_mark:
3. learners dive into the exercise on their own if time permits :clock1:

:question: Instructors help with show stoppers like syntax errors where they can.
:computer: if you like to conduct the exercises, or code along during the videos, we suggest to use [google colab](colab.research.google.com/). Note, you may need a google account for this.

Each lesson has a jupyter notebook, that is half filled. The video lectures start from this notebook and provide content to fill in.

## Lessons

- Lesson 00: Preface
- Lesson 01: Diving into Regression [video](), [learner notebook](https://github.com/deeplearning540/lesson01/blob/main/lesson.ipynb)
- Lesson 02: Enter Clustering [video](), [learner notebook](https://github.com/deeplearning540/lesson02/blob/main/lesson.ipynb)
- Lesson 03: From Clustering To Classification [video](), [learner notebook](https://github.com/deeplearning540/lesson03/blob/main/lesson.ipynb)
- Lesson 04: Classification Performance ROCs [video](), [learner notebook](https://github.com/deeplearning540/lesson04/blob/main/lesson.ipynb)
- Lesson 05: Neural Networks as Code [video](), [learner notebook](https://github.com/deeplearning540/lesson05/blob/main/lesson.ipynb)
- Lesson 06: How did we train [video](), no jupyter notebook for this lesson
- Lesson 07: CNNs [video](), [learner notebook](https://github.com/deeplearning540/lesson06/blob/main/lesson.ipynb)
- Lesson 08: Deep Learning [video](), [learner notebook](https://github.com/deeplearning540/lesson07/blob/main/lesson.ipynb)

## Team Notes

### Lesson 01

- Lesson 01: Diving into Regression [video](), [learner notebook](https://github.com/deeplearning540/lesson01/blob/main/lesson.ipynb)

- Check your Learning