-
Notifications
You must be signed in to change notification settings - Fork 2
RFC: add channel_map and version to codebook #11
base: master
Are you sure you want to change the base?
Conversation
2f1703d
to
79f3615
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
how about "codebook" instead of "codes"? I like the list of codewords being the codebook, but either way is fine with me
@berl I like that suggestion. Can you think of a more general term for the |
} | ||
], | ||
"channel_map": [ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
why not do a map of channel name to index val? at least you gain channel name uniqueness as a freebie.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I could also see having it as an array. cF. https://github.com/IDR/idr-metadata/blob/403b0817c1c57dd6b509cfb0d0d41bd35ec1b938/idr0006-fong-nuclearbodies/screenA/idr0006-screenA-renderdef.yml
1:
label: DAPI
min: 0
max: 1000
color: "0000FF"
2:
label: TRITC
min: 50,
max: 800
since an ordering is guaranteed.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. @berl @ambrosejcarr is channel_map.json required for the spaceTx pipeline? does it make more sense for this to live in experiment.json?
channel_map.json is not strictly required for the pipeline, but it would make some things a little easier to read. E.g. asking to filter images differently based on channel, like |
Thanks all for comments. I'm going to wait to land this until we construct the dependency tree between sptx-format, slicedimage, and starfish. |
landing post-merge might be a serious PITA. |
This PR is a first attempt at adding a version and code_map to the codebook. Note that the diff is much smaller than git makes it appear -- to add properties to the codebook it needed to be converted from an
array
to anobject
, which caused it to get shifted up a level.See examples for the proposed codebook structure.
@berl does this address the concern about being able to detect the channels?
Addresses #3, #10