Data server #4
Data server #4
Conversation
| raise ValueError("Currently only supporting nrs/cellmap and /groups/cellmap/cellmap") | ||
| # TODO should be done better | ||
| ds = get_ds(dataset_path, filetype) | ||
| add_layer(s, ds, "raw") |
There was a problem hiding this comment.
@davidackerman i want to support multiscale. do you know a better way to open LocalVolume ?
There was a problem hiding this comment.
what type of multiscale? also it seems like in the newest neuroglancer you don't need to specify n5:// or zarr://, itll just know it from the https://path
There was a problem hiding this comment.
e.g. /nrs/cellmap/data/jrc_mus-pancreas-5/jrc_mus-pancreas-5.zarr/recon-1/em/fibsem-uint8
currently i am only supporting one res
There was a problem hiding this comment.
i haven't used funlib.show. but you can just add a multiscale layer to neuroglancer just like this:
with viewer.txn() as s:
s.layers["raw"] = neuroglancer.ImageLayer(
source=f"zarr://s3://janelia-cosem-datasets/{current_collection}/{current_collection}.zarr/recon-1/em/fibsem-uint8",
shaderControls={
"normalized": {"range": [contrast_min, contrast_max]}
},
)is that what you mean?
There was a problem hiding this comment.
but i can't do that for local data. i can't feed a path
There was a problem hiding this comment.
there seems to be an unsanswered question about this here: google/neuroglancer#515
but maybe something like this would work if zarr allows reading in multiscale. then you would just open the zarr dataset and pass that to local volume:
https://github.com/google/neuroglancer/blob/5a54a489ed3fd679a5beae9906338c602371a9fa/python/examples/interactive_inference.py#L47-L52
No description provided.