# Meeting 21 January

* conceptual model - continue working
* look at more NIPS papers
* continue exercises

# Meeting 21 January

* conceptual model abstractions: define examples
* look at more NIPS papers
* continue with exercises

# Meeting 14 January

* look at NIPS/JMLR publications, pick 20 papers check how learning tasks are described and introduced formally.
* Keep working on coceptual model
* Excercise 5.28
* look again at 5.29 and 5.27
* continue with exercises from chapter 6
* next meeting 12:00

# Meeting 17 December
1. define the task properly
2. define the methods properly: fit/predict
3. continue with sorting exercise
4. exercises on Chapter 5: definitions

5. Draft a base class for a directional forecaster
6. Two examples of a directional forecasters
7. Redo problems 5.27-5.29
8. Write base class for directional forecaster

# Meeting 3 December

1. Do more math exercises from chapter 5
2. revise examples chapter 4, also not where comments explicitly
3. do problems 5.25-5.29
4. try to define abstract scientific type for a directional forecaster. Start with three examples.

# Meeting 19 November

action points
1. redo 4.2, 4.3, 4.4 (one wrong), 4.5(d redo) ,f (implication)
2. sorting algorithm and other ADT write properties of ADT - properties of sorting alg, etc.
3. start reading section 5

# Meeting 12 November

action points:

1. work on conceptula model sorting, workflows for coneptual model on sorting. give examples,
2. continue with chapter 4 exercises
3. 3.18, more exercises from chapter 3 with English

# Meeting 5 November

action points
1. continue chapter 3 exercises
2. conceptual model for sorting algorithms

Agenda points:

1. Review of actions
2. show solution of equidistant points
3.

# Meeting 29 October 2021

action points

1. section 1 problems: work on 1 and 2
2. problem 2.6 h find correct answer
3. problem 2.10, 2.14, 2.15, 2.20
4. start working on conceptual model for directional forecasting study

# Meeting 22 October

action points

1. sorting algorithm: continue
2. isort precommit for sktime
3. exercises from chapter 2 and 1 corrct wrong, equidistant point, bathroom door

# Meeting 15 October 2021

Franz comments

1. Regarding content: a little bit off. The examples are not good.
2. How to proceed: exercise sheet. First I should have done the exercises.

Action points

1. do the exercises from the exercise sheet on the LTCC lecture 1 on conceptual model.
2. Solve problem 1.7 from reading, writing, proving.
3. Chapter 1, do more exercises

# Meeting Notes 17 September 2021
Present: Franz, Ramin, Viktor

Franz, Ramin and Viktor agree that progress is nost as quick as it should be. Crucial focus has to be on writing a paper on directional forecasting.

Discussed math exercises. Franz pointed out mistakes and encouraged Viktor to continue with exercises.

Franz and Ramin pointed out that Viktor continues to write math with imprecisions and should focus on improving this.

Franz suggested that Viktor should avoid writing math sections for the paper until Viktor improves his math writing skills by doing more exercises from Reading, Writing, Proving.

**Short-term objectives**
* Start building components required for directional forecasting benchmarking experiement.
* Start with anomaly destction algorithms. Check what is already implemented in Python and how it can be used in a sktime network pipeline.
* Continue doing literature review on directional forecasting.
* Do math exercises from Reading, Writing, Proving

**Mid-term objective**
* write paper on directional forecasting

**Action Points**
* Keep high level TODO list
* Write down a plan for conducting benchmarking experiment
* Keep notes in HackMD. Going forward, HackMD notebook should be shared before each meeting.