Skip to content

Commit aeceb46

Browse files
Analyse day 3
1 parent 29d9d08 commit aeceb46

15 files changed

Lines changed: 60 additions & 52 deletions

evaluations/20260423_day_3/README.md

Lines changed: 21 additions & 42 deletions
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,8 @@
55

66
## Survey at end
77

8-
- [Evaluation results (csv)](survey_end.csv)
8+
- [Evaluation results (csv)](evaluation_results.csv)
9+
- [Evaluation results (xlsx)](evaluation_results.xlsx)
910
- [Analysis script](analyse.R)
1011
- [Average confidence per question (.csv)](average_confidences.csv)
1112
- [Success score](success_score.txt):
@@ -18,61 +19,39 @@
1819

1920
### [Pace](pace.txt)
2021

21-
- Richèl's sessions had a good pace,
22-
the rest were way too fast with too little time for exercises
23-
and too little time for breaks.
22+
- good
23+
- good
24+
- Richèl's sessions had a good pace, the rest were way too fast with too little time for exercises and too little time for breaks.
2425
- more practice
2526
- good
2627
- Good, did well balancing levels across a large group
27-
- I like the way that Richèl is teaching.
28-
He makes the concepts clear and you can follow his instructions and learn.
29-
However, I cannot follow the class when Bjorn is teaching, I feel very lost.
28+
- I like the way that Richèl is teaching. He makes the concepts clear and you can follow his instructions and learn. However, I cannot follow the class when Bjorn is teaching, I feel very lost.
3029
- The overall pace was OK
3130
- Pace seems to be OK, it's probably me who is slower than the pace :)
31+
- A bit uneven. I think more time is needed
32+
- OK
33+
3234

3335
### [Future topics](future_topics.txt)
3436

3537
- We did not even get to practice what is already in the course properly
3638
- GNNs
37-
- More information about how analyze big data
38-
such as genomic or transcriptomic data.
39+
- More information about how analyze big data such as genomic or transcriptomic data.
40+
- More examples on how to run Python codes as batch jobs. More about finding modules
41+
- USPEX
42+
3943

4044
### [Other comments](comments.txt)
4145

42-
- more cases, like, an `oom` killed job,
43-
how to `seff` it and solve the issue to run it properly
44-
- Use Richèl's teaching style for all topics.
45-
Keep the amount of content in the course as it is,
46-
but spread it over 5 days instead of 4 days.
47-
There's simply way too much to go through,
48-
way to many potential technical difficulties that take time,
49-
and way too little time to do exercises to actually absorb
50-
all the content and enjoy the learning process.
51-
And I would suggest that the course on how to connect and work
52-
in an HPC environment be moved to before this course during the same semester,
53-
so that people don't have to deal with having to learn that at the same time.
54-
Right now the "Online training events for new users:
55-
NAISS Introduction training days" course is after this course
56-
during the semester, which makes little sense considering that
57-
its contents should be a prerequisite for this course.
46+
- more cases, like, an oom killed job , how to seff it and solve the issue to run it properly
47+
- Use Richèl's teaching style for all topics. Keep the amount of content in the course as it is, but spread it over 5 days instead of 4 days. There's simply way too much to go through, way to many potential technical difficulties that take time, and way too little time to do exercises to actually absorb all the content and enjoy the learning process. And I would suggest that the course on how to connect and work in an HPC environment be moved to before this course during the same semester, so that people don't have to deal with having to learn that at the same time. Right now the "Online training events for new users: NAISS Introduction training days" course is after this course during the semester, which makes little sense considering that its contents should be a prerequisite for this course.
5848
- Where should we improve? materials, exercises, and structure
59-
- For me personally, the time spent interacting with the cluster
60-
and getting familiarised was the most useful.
61-
This is personally though an I'm not sure whether this
62-
would be the case across the board.
63-
- Today (day3), the topics were more of what I expected from
64-
an Introduction to Python and Using Python in an HPC environment course.
65-
The morning session about pandas, matplotlib and seaborn was interesting,
66-
as well as the big data with python.
67-
I will probably use Dask in the future so it was a good explanation
68-
of how Dask works.
69-
I still don't understand the time spent talking about bash and sending jobs.
70-
We could use that time learning
71-
and using common python packages in an HPC cluster.
72-
- It think that we need more documentation about Dardel.
73-
Also, we need more hands-on and demonstrations.
74-
- I feel some exercises are not in "step-by-step" format
75-
that makes difficult to complete it.
49+
- For me personally, the time spent interacting with the cluster and getting familiarised was the most useful. This is personaly though an I'm not sure whether this would be the case across the board.
50+
- Today (day3), the topics were more of what I expected from an Introduction to Python and Using Python in an HPC environment course. The morning session about pandas, matplotlib and seaborn was interesting, as well as the big data with python. I will probably use Dask in the future so it was a good explanation of how Dask works. I still don't understand the time spent talking about bash and sending jobs. We could use that time learning and using common python packages in an HPC cluster.
51+
- It think that we need more documentation about Dardel. Also, we need more hands-ons and demonstrations.
52+
- I feel some exercises are not in "step-by-step" format that makes difficult to complete it.
53+
- I think there is very little material on the pandas/matplotlib/seaborn part and would like there to be more so I can redo it later myself. This just tells you were to go look for the info and that could be put in a box at the bottom. I am not fond of the way Richel teaches (sorry - he is very nice). I prefer to have more theory and material to read and then also show some demos. Maybe if there was more time for this course. I really think you should have it a full week.
54+
- At first I didn't understand why we need to learn about batch since most people use IDEs now but I realized that batch is also part of HPC and that there may be some people who still submit jobs to batch and doesn't just work in jupyter
7655

7756
## [Any other feedback](any_other_feedback.csv)
7857

1.47 KB
Loading

evaluations/20260423_day_3/analyse.R

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
#!/bin/env Rscript
22

33
read_data <- function() {
4-
readr::read_csv("survey_end.csv", show_col_types = FALSE)
4+
readr::read_csv("evaluation_results.csv", show_col_types = FALSE)
55
}
66

77
t <- read_data()
@@ -187,7 +187,6 @@ t_sessions_taught <- unique(t_tidy$question)
187187
# Cut out sessions if needed
188188
t_sessions_taught <- stringr::str_subset(t_sessions_taught, "Pandas|Seaborn|matplotlib")
189189

190-
191190
testthat::expect_true(all(t_sessions_taught %in% t_tidy$question))
192191

193192
confidences_on_taught_sessions <- t_tidy |> dplyr::filter(question %in% t_sessions_taught)
Lines changed: 7 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
11
question,mean
2-
I can decide on useful file formats for big data,2.36
3-
I can make a plot with Seaborn,3.6
4-
I can make a plot with matplotlib,4
5-
I can read and write tabular data with Pandas,3.45
6-
I can submit a script to the job scheduler,3.45
7-
"I can work (i.e. inspect, clean, sort, etc) on my data with Pandas",3.18
8-
I can write a batch script,3.3
2+
I can decide on useful file formats for big data,2.15
3+
I can make a plot with Seaborn,3.17
4+
I can make a plot with matplotlib,3.75
5+
I can read and write tabular data with Pandas,3.31
6+
I can submit a script to the job scheduler,3.69
7+
"I can work (i.e. inspect, clean, sort, etc) on my data with Pandas",3
8+
I can write a batch script,3.5
587 Bytes
Loading

evaluations/20260423_day_3/comments.txt

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -5,3 +5,5 @@ For me personally, the time spent interacting with the cluster and getting famil
55
Today (day3), the topics were more of what I expected from an Introduction to Python and Using Python in an HPC environment course. The morning session about pandas, matplotlib and seaborn was interesting, as well as the big data with python. I will probably use Dask in the future so it was a good explanation of how Dask works. I still don't understand the time spent talking about bash and sending jobs. We could use that time learning and using common python packages in an HPC cluster.
66
It think that we need more documentation about Dardel. Also, we need more hands-ons and demonstrations.
77
I feel some exercises are not in "step-by-step" format that makes difficult to complete it.
8+
I think there is very little material on the pandas/matplotlib/seaborn part and would like there to be more so I can redo it later myself. This just tells you were to go look for the info and that could be put in a box at the bottom. I am not fond of the way Richel teaches (sorry - he is very nice). I prefer to have more theory and material to read and then also show some demos. Maybe if there was more time for this course. I really think you should have it a full week.
9+
At first I didn't understand why we need to learn about batch since most people use IDEs now but I realized that batch is also part of HPC and that there may be some people who still submit jobs to batch and doesn't just work in jupyter
2.01 KB
Loading
1.56 KB
Loading
Lines changed: 24 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,24 @@
1+
ID,Start time,Completion time,Email,Name,I can read and write tabular data with Pandas,"I can work (i.e. inspect, clean, sort, etc) on my data with Pandas",I can make a plot with Seaborn,I can write a batch script,I can submit a script to the job scheduler,I can decide on useful file formats for big data,I can make a plot with matplotlib,"Overall, how would you rate this training event?","What do you think about the pace of teaching overall today?
2+
3+
","Would you recommend this course to your colleagues?
4+
","Which future training topics would you like to be provided by the training host(s)?
5+
","Do you have any additional comments?
6+
7+
Suggestions/ideas:
8+
9+
- What did you like best? (materials, exercises, structure)
10+
- Where should we improve? (materials, exercises, structure)
11+
- Training organi...",Last modified time
12+
1,4/23/26 13:07:06,4/23/26 14:17:41,anonymous,,I have some confidence I can do this,I have some confidence I can do this,I have low confidence I can do this,I have some confidence I can do this,I have good confidence I can do this,I have some confidence I can do this,I have some confidence I can do this,8,good,Yes,,"more cases, like, an oom killed job , how to seff it and solve the issue to run it properly",
13+
2,4/23/26 15:30:41,4/23/26 15:31:21,anonymous,,I have low confidence I can do this,I have some confidence I can do this,I have good confidence I can do this,I have low confidence I can do this,I have good confidence I can do this,I have low confidence I can do this,I have good confidence I can do this,6,good,Yes,,,
14+
3,4/23/26 15:30:17,4/23/26 15:39:05,anonymous,,I have some confidence I can do this,I have low confidence I can do this,I have some confidence I can do this,I have good confidence I can do this,I have good confidence I can do this,I have low confidence I can do this,I have good confidence I can do this,6,"Richèl's sessions had a good pace, the rest were way too fast with too little time for exercises and too little time for breaks.",No,,"Use Richèl's teaching style for all topics. Keep the amount of content in the course as it is, but spread it over 5 days instead of 4 days. There's simply way too much to go through, way to many potential technical difficulties that take time, and way too little time to do exercises to actually absorb all the content and enjoy the learning process. And I would suggest that the course on how to connect and work in an HPC environment be moved to before this course during the same semester, so that people don't have to deal with having to learn that at the same time. Right now the ""Online training events for new users: NAISS Introduction training days"" course is after this course during the semester, which makes little sense considering that its contents should be a prerequisite for this course.",
15+
4,4/23/26 15:52:17,4/23/26 15:54:41,anonymous,,I have low confidence I can do this,I have no confidence I can do this,I have no confidence I can do this,I have some confidence I can do this,I have some confidence I can do this,I have some confidence I can do this,I have some confidence I can do this,7,more practice ,Not sure,We did not even get to practice what is already in the course properly ," Where should we improve? materials, exercises, and structure",
16+
5,4/23/26 15:57:42,4/23/26 15:58:46,anonymous,,I have some confidence I can do this,I have some confidence I can do this,I have some confidence I can do this,I have no confidence I can do this,I have no confidence I can do this,I have no confidence I can do this,I have some confidence I can do this,6,,Yes,,,
17+
6,4/23/26 15:57:55,4/23/26 15:59:07,anonymous,,I can absolutely do this!,I can absolutely do this!,I can absolutely do this!,I can absolutely do this!,I can absolutely do this!,I have some confidence I can do this,I can absolutely do this!,8,good ,Yes,GNNs,,
18+
7,4/23/26 15:56:58,4/23/26 15:59:23,anonymous,,I have good confidence I can do this,I have good confidence I can do this,I have good confidence I can do this,I have good confidence I can do this,I have good confidence I can do this,I have some confidence I can do this,I have good confidence I can do this,9,"Good, did well balancing levels across a large group",Yes,,"For me personally, the time spent interacting with the cluster and getting familiarised was the most useful. This is personaly though an I'm not sure whether this would be the case across the board. ",
19+
8,4/23/26 15:34:26,4/23/26 16:00:19,anonymous,,I can absolutely do this!,I can absolutely do this!,I can absolutely do this!,I can absolutely do this!,I can absolutely do this!,I have some confidence I can do this,I can absolutely do this!,7,,Not sure,,"Today (day3), the topics were more of what I expected from an Introduction to Python and Using Python in an HPC environment course. The morning session about pandas, matplotlib and seaborn was interesting, as well as the big data with python. I will probably use Dask in the future so it was a good explanation of how Dask works. I still don't understand the time spent talking about bash and sending jobs. We could use that time learning and using common python packages in an HPC cluster.",
20+
9,4/23/26 16:00:02,4/23/26 16:04:52,anonymous,,I have good confidence I can do this,I have good confidence I can do this,I can absolutely do this!,I have some confidence I can do this,I have some confidence I can do this,I have no confidence I can do this,I can absolutely do this!,7,"I like the way that Richèl is teaching. He makes the concepts clear and you can follow his instructions and learn. However, I cannot follow the class when Bjorn is teaching, I feel very lost. ",Yes,More information about how analyze big data such as genomic or transcriptomic data.,"It think that we need more documentation about Dardel. Also, we need more hands-ons and demonstrations.",
21+
10,4/23/26 15:59:00,4/23/26 16:06:39,anonymous,,I have good confidence I can do this,I have some confidence I can do this,I have good confidence I can do this,I have some confidence I can do this,I have some confidence I can do this,I have low confidence I can do this,I have good confidence I can do this,8,The overall pace was OK ,Yes,-,,
22+
11,4/23/26 16:00:33,4/23/26 16:08:36,anonymous,,I have some confidence I can do this,I have low confidence I can do this,I did not attend that session,,I have low confidence I can do this,I have some confidence I can do this,I did not attend that session,7,"Pace seems to be OK, it's probably me who is slower than the pace :)",Yes,,"I feel some exercises are not in ""step-by-step"" format that makes difficult to complete it. ",
23+
12,4/24/26 14:31:27,4/24/26 14:37:20,anonymous,,I have some confidence I can do this,I have no confidence I can do this,I have no confidence I can do this,I can absolutely do this!,I can absolutely do this!,I have no confidence I can do this,I have some confidence I can do this,8,A bit uneven. I think more time is needed,Yes,More examples on how to run Python codes as batch jobs. More about finding modules ,I think there is very little material on the pandas/matplotlib/seaborn part and would like there to be more so I can redo it later myself. This just tells you were to go look for the info and that could be put in a box at the bottom. I am not fond of the way Richel teaches (sorry - he is very nice). I prefer to have more theory and material to read and then also show some demos. Maybe if there was more time for this course. I really think you should have it a full week.,
24+
13,4/24/26 17:26:07,4/24/26 17:28:56,anonymous,,I have low confidence I can do this,I have some confidence I can do this,I have no confidence I can do this,I have good confidence I can do this,I can absolutely do this!,I have no confidence I can do this,I have low confidence I can do this,8,OK,Yes,USPEX,At first I didn't understand why we need to learn about batch since most people use IDEs now but I realized that batch is also part of HPC and that there may be some people who still submit jobs to batch and doesn't just work in jupyter,

evaluations/20260424_day_4/Intro to Python and Using Python in an HPC environment, 20 + 22-24 April, 2026 - Day 3(1-13).xlsx renamed to evaluations/20260423_day_3/evaluation_results.xlsx

File renamed without changes.

0 commit comments

Comments
 (0)