Closed Matthew-J-Payne closed 1 year ago
In the photo, I wrote "scihub" and "dataframe", when I mean to compare "scihub" with "DATASPACE" ðŸ˜
Gonna have to solve the merge conflicts on this branch too
Looks more or less fine to me. Has this been run already, i.e. is it throwing any errors?
When the do_download_from_dataspace
if branch was uncommented (starts line 164 in acd_by_tile_raster
), a dataframe is returned fine from query_by_polygon
.
So the todo is to create a Pandas DataFrame that matches the expected columns and formatting (df_all
) of the scihub derived version of df_all
(in the photo).
Clone this branch and give it a go, work your magic! 🪄
This is the progress so far:
We run the whole pipeline (raster and vector) from
run_acd_national.py
from CLI:python pyeo_1/apps/acd_national/run_acd_national.py pyeo_matt.ini
If within the ini the raster processing flags are turned on, like
do_raster = True
anddo_download_from_dataspace = True
, then the raster processing functions will be called.The trouble we currently have is coercing the dataframe returned from
query_by_polygon()
(line 184 ofacd_by_tile_raster.py
) to mirror the dataframe returned from the scihub equivalent, as this is what the rest of the pyeo_1 raster pipeline expects.Our last efforts were to compare the dataframes from the dataspace and scihub queries: