This should produce a 48MYU_20180813_20210424_v001.gpq file under the data/embeddings/ folder. Sample file (need to unzip): 48MYU_20180813_20210424_v001.gpq.zip
Extra configuration options can be found using python trainer.py predict --help
Still many things that could be improved, such as sharing duplicated code between model_vit.py and model_clay.py, but will merge in to main first for the first release.
What I am changing
{MGRS:5}_{MINDATE:8}_{MAXDATE:8}_v{VERSION:3}.gpq
(e.g.12ABC_20210101_20231231_v001.gpq
), following discussions at https://github.com/Clay-foundation/model/pull/86#discussion_r1430759687How I did it
predict_step
, implement the logic to do the forward pass and save-to-gpq stepSample output table would look like this (same as #86):
TODO in this PR:
predict_step
to generategpd.GeoDataFrame
tableon_predict_epoch_end
to mergegpd.GeoDataFrame
tables and output to GeoParquet file(s)TODO in the future:
model_vit.py
andmodel_clay.py
?How you can test it
s3://clay-tiles-02/02/
s3://clay-model-ckpt/v0/mae_epoch-02_val-loss-0.52.ckpt
to thecheckpoints/
folder.48MYU_20180813_20210424_v001.gpq
file under thedata/embeddings/
folder. Sample file (need to unzip): 48MYU_20180813_20210424_v001.gpq.zippython trainer.py predict --help
To load the embeddings from the GeoParquet file:
Related Issues
Towards #3