gene_order

[bjstewart1]

Using 0.1.0.post1.dev12+ga195722
Using query model downloaded here https://zenodo.org/records/8286452

Getting stuck trying to find gene_order.tsv which is not in the unpackaged query model directory

cq = CellQuery(model_path)

FileNotFoundError: [Errno 2] No such file or directory: '..../query_model_v1/gene_order.tsv'

[jsalignon]

Same issue here.
I moved the "query_model_v1/train_meta.csv" to "annotation_model_v1/cellsearch/full_kNN_meta.csv". It worked for a while but then I got stuck as well with:
KeyError: "Only a column name can be used for the key in a dtype mappings argument. 'prediction_nn_dist' not found in columns." Then I renamed the first line in the csv file from nn_dist to prediction_nn_dist. But then I had another issue... Seems like the code does not fit with the provided files.
I could not make the docker work as well; neither the tagged release.

[gheimberg]

Thanks for the issue, we are looking at this. We should have an answer early next week.

[javh]

Sorry for the delay on this. I think our code and the structure of the published model got out of sync. We just pushed a new model (v1.1) to Zenodo here:
https://zenodo.org/records/10685499

We are planning to tag a new release of the API shortly, but this model should work with the current tip of master and the docker container that was built today.

[jsalignon]

Sounds great! Thank you!