search

opera-adt / COMPASS

COregistered Multi-temPorAl Sar Slc
Apache License 2.0
38 stars 17 forks source link

Add separate generation of CSLC-S1 static layers #177

Closed vbrancat closed 1 year ago

vbrancat commented 1 year ago

This PR allows the SAS CSLC-S1 to independently generate the static layers product. Therefore, the new CSLC-S1 SAS behavior is the following:

  1. when product_type: CSLC_S1, the SAS generates only the CSLC-S1 product
  2. when product_type: CSLC_S1_STATIC, the SAS generates only the CSLC-S1 static layers product

Flags to run rdr2geo and geocode the static layers have been removed from the schema. The default SAS product_type has been set to always generate the CSLC-S1 product.

Note: would be good if this PR can be merged ASAP so that the integration of the LOS vectors into the CSLC-S1 static layers workflow will be much easier. Thank you :)

seongsujeong commented 1 year ago

A comment not much related to the point of this PR: I see the checks in the recent PRs in this repo are failing on "Read the Docs build". Will merging the PR #170 solve the issue?

vbrancat commented 1 year ago

This is a question for @scottstanie!

vbrancat commented 1 year ago

@LiangJYu and @seongsujeong all your comments should have been addressed. Please, have another look to this PR. One thing I have notice is that there is something off with measuring the runtime of the static layers. This is what I get by running the static layers workflow:

journal: elapsed time (GEO-IN) [s]: 6.704
journal: s1_geocode_metadata burst successfully ran in -19067 days, 8:07:32 (hr:min:sec)
journal: s1_static_layers burst successfully ran in 0:01:30 (hr:min:sec)

Not sure why s1_geocode_metadata gives that awful result :s.

LiangJYu commented 1 year ago

@LiangJYu and @seongsujeong all your comments should have been addressed. Please, have another look to this PR. One thing I have notice is that there is something off with measuring the runtime of the static layers. This is what I get by running the static layers workflow:

journal: elapsed time (GEO-IN) [s]: 6.704
journal: s1_geocode_metadata burst successfully ran in -19067 days, 8:07:32 (hr:min:sec)
journal: s1_static_layers burst successfully ran in 0:01:30 (hr:min:sec)

Not sure why s1_geocode_metadata gives that awful result :s.

I'll look into it

vbrancat commented 1 year ago

@seongsujeong do you mind having another look to this PR? Thanks :)