Instructor notes: Introducing analysis modules #157
Conversation
There was a problem hiding this comment.
This looks like a good start. I had some thoughts on a bit of ordering and some details about the download script. There may also be a few decisions we need to make about how we are really going to present the data.
Finally, a general concern is what happens if connectivity means we can't really use LSfR? I don't know how heavy the bandwidth requirements are, but if it gets really laggy, the experience could kind of suck.
| - Download test data for an opportunity to explain the benefits of developing with test data: | ||
| - `./download-data.py --test-data` |
There was a problem hiding this comment.
This is still a healthy download: They should probably include the same flags as below:
./download-data.py --format AnnData --projects SCPCP0000001 --test-data
Note that --include processed should be the default, so they will not need to enter that, but you might introduce it to show what other data there might be. But I would probably not at this point.
| ```bash | ||
| # The openscpca-hello-python environment should have an asterisk next to it | ||
| conda env list |
There was a problem hiding this comment.
I would show people this sooner: before activating?
| Now, run the module: | ||
|
|
||
| ```bash | ||
| cd path/to/analysis/module |
There was a problem hiding this comment.
They should already be there when creating the conda environment
| ## Activity | ||
|
|
||
| Instructors will explain and demonstrate this activity while participants follow along. | ||
| We expect that participants will take all steps along with the instructor. |
There was a problem hiding this comment.
Just as a question: which environment are we planning to use here? VSCodium?
There was a problem hiding this comment.
Yeah, that was my thought (see #160), since we'll be in python land
|
|
||
| Now instructors and partipants will all run the module together, following the instructions in its README. | ||
|
|
||
| First, make sure we are working in the correct conda environment, if this is not reflected in the terminal directly: |
There was a problem hiding this comment.
Do we know what this will look like on LS? My memory is we get bash by default, and the conda default will be to show the environment in the prefix. I think showing conda env list is a good idea still, but just to clarify the expectation for this section of the notes.
There was a problem hiding this comment.
My memory is we get bash by default, and the conda default will be to show the environment in the prefix
This is also my memory! I'll point it out here.
| - `./download-data.py --test-data` | ||
| - Do a _dry run_ of a download: | ||
| - `./download-data.py --format AnnData --include processed --projects SCPCP000001 --dryrun` | ||
| - Download processed AnnData files from a single project for use with `hello-python`: |
There was a problem hiding this comment.
It might be worth showing that if they run it a second time it won't re-download?
They should also be shown the data folder structure after download: show how the data is versioned by date but linked to current. And that the test data directory can be swapped in by rerunning the test download, or switch back by rerunning without test data... This may get a bit in the weeds, but I think it is handy.
| - `./download-data.py --format AnnData --include processed --projects SCPCP000001 --dryrun` | ||
| - Download processed AnnData files from a single project for use with `hello-python`: | ||
| - `./download-data.py --format AnnData --include processed --projects SCPCP000001` | ||
| - Or, if connectivity is an issue, `./download-data.py --format AnnData --include processed --samples SCPCS000001` |
There was a problem hiding this comment.
Connectivity shouldn't be an issue if we are using LSfR, as everything will be internal to AWS. The only question is whether LSfR is too laggy to use, in which case we will have to switch everything to local work, I guess?
Noting that it looks like our first iteration of this workshop will be primarily Macs, so this will make our lives here a little smoother! |
Closes #152
This PR adds instructor notes for the activity where we introduce analysis modules, download data, and run
hello-python. I added this to a new nested directoryinstructor_notes/openscpcaalong with a briefREADMEfile.Let me know where I can expand, reduce, or rephrase!