reVX.rpm.rpm_output.RPMOutput

class RPMOutput(rpm_clusters, cf_fpath, excl_fpath, excl_dict, techmap_dset, excl_area=None, include_threshold=0.001, n_profiles=1, rerank=True, cluster_kwargs=None, max_workers=None, trg_bins=None, trg_dset='lcoe_fcr', pre_extract_inclusions=False, target_crs=None)

Bases: object

Framework to format and process RPM clustering results.

Parameters:

• rpm_clusters (pd.DataFrame | str) – Single DataFrame with (gid, gen_gid, cluster_id, rank), or str to file.

• cf_fpath (str) – Path to reV .h5 file containing desired capacity factor profiles

• excl_fpath (str | None) – Filepath to exclusions data (must match the techmap grid). None will not apply exclusions.

• excl_dict (dict | None) – Dictionary of exclusion LayerMask arugments {layer: {kwarg: value}}

• techmap_dset (str) – Dataset name in the exclusions file containing the exclusions-to-resource mapping data.

• excl_area (float | None) – Area in km2 of one exclusion pixel. None will calculate the exclusion pixel area from the exclusion projection profile.

• include_threshold (float) – Inclusion threshold. Resource pixels included more than this threshold will be considered in the representative profiles. Set to zero to find representative profile on all resource, not just included.

• n_profiles (int) – Number of representative profiles to output.

• rerank (bool) – Flag to rerank representative generation profiles after removing excluded generation pixels.

• cluster_kwargs (dict) – RPMClusters kwargs

• max_workers (int, optional) – Number of parallel workers. 1 will run serial, None will use all available., by default None

• trg_bins (str | list | None) – TRG bins as an ordered list of bin edge values or string to a csv containing a single column with bin edge values. None will ignore trgs.

• trg_dset (str) – Dataset associated with TRG bins that can be found in the cf_fpath file.

• pre_extract_inclusions (bool) – Flag to pre-extract the inclusion mask using excl_fpath and excl_dict. This is advantageous if the excl_dict is highly complex and if you’re processing a lot of points. Default is False.

• target_crs (str) – Target coordinate reference system, e.g. “EPSG:4326”. Note that everything is in EPSG:4326/WGS84 by default (raw lat/lon with lon=0 at Greenwich).

Methods

apply_exclusions()	Calculate exclusions for clusters, adding data to self._clusters.
apply_trgs()	Apply TRG's if requested.
export_all(out_dir[, job_tag])	Run RPM output algorithms and write to CSV's.
extract_profiles(rpm_clusters, cf_fpath, out_dir)	Use pre-formatted RPM cluster outputs to generate profile outputs.
make_shape_file(fpath_shp)	Make shape file containing all clusters.
process_outputs(rpm_clusters, cf_fpath, ...)	Perform output processing on clusters and write results to disk.
run_rerank([groupby, rank_col])	Re-rank rep profiles for included resource in generic groups.

Attributes

cluster_summary	Summary dataframe with cluster_id primary key.
excl_lat	Get the full 2D array of latitudes of the exclusion grid.
excl_lon	Get the full 2D array of longitudes of the exclusion grid.

property excl_lat

Get the full 2D array of latitudes of the exclusion grid.

Returns:

_excl_lat (np.ndarray) – 2D array representing the latitudes at each exclusion grid cell

property excl_lon

Get the full 2D array of longitudes of the exclusion grid.

Returns:

_excl_lon (np.ndarray) – 2D array representing the latitudes at each exclusion grid cell

apply_exclusions()

Calculate exclusions for clusters, adding data to self._clusters.

Returns:

self._clusters (pd.DataFrame) – self._clusters with new columns for exclusions data.

apply_trgs()

Apply TRG’s if requested.

run_rerank(groupby='cluster_id', rank_col='rank_included')

Re-rank rep profiles for included resource in generic groups.

Parameters:

• groupby (str | list) – One or more columns in self._clusters to groupby and rank profiles within each group.

• rank_col (str) – Column to add to self._clusters with new rankings.

property cluster_summary

Summary dataframe with cluster_id primary key.

Returns:

s (pd.DataFrame) – Summary dataframe with a row for each cluster id.

make_shape_file(fpath_shp)

Make shape file containing all clusters.

Parameters:

fpath_shp (str) – Filepath to write shape_file to.

export_all(out_dir, job_tag=None)

Run RPM output algorithms and write to CSV’s.

Parameters:

• out_dir (str) – Directory to dump output files.

• job_tag (str | None) – Optional name tag to add to the csvs being saved. Format is “rpm_cluster_output_{tag}.csv”.

classmethod extract_profiles(rpm_clusters, cf_fpath, out_dir, n_profiles=1, job_tag=None, max_workers=None, key=None, forecast_fpath=None)

Use pre-formatted RPM cluster outputs to generate profile outputs.

Parameters:

• rpm_clusters (pd.DataFrame | str) – Single DataFrame with (gid, gen_gid, cluster_id, rank), or str to file.

• cf_fpath (str) – reV generation output file.

• out_dir (str) – Directory to dump output files.

• n_profiles (int) – Number of representative profiles to output.

• job_tag (str | None) – Optional name tag to add to the output files. Format is “rpm_cluster_output_{tag}.csv”.

• max_workers (int, optional) – Number of parallel workers. 1 will run serial, None will use all available., by default None

• key (str | None) – Column in clusters to sort ranks by. None will allow for default logic.

• forecast_fpath (str) – reV generation output file for forecast data. If this is input, profiles will be taken from forecast fpath instead of fpath gen based on a NN mapping.

classmethod process_outputs(rpm_clusters, cf_fpath, excl_fpath, excl_dict, techmap_dset, out_dir, job_tag=None, max_workers=None, cluster_kwargs=None, excl_area=None, include_threshold=0.001, n_profiles=1, rerank=True, trg_bins=None, trg_dset='lcoe_fcr', pre_extract_inclusions=False, target_crs=None)

Perform output processing on clusters and write results to disk.

Parameters:

• rpm_clusters (pd.DataFrame | str) – Single DataFrame with (gid, gen_gid, cluster_id, rank), or str to file.

• cf_fpath (str) – Path to reV .h5 file containing desired capacity factor profiles

• excl_fpath (str | None) – Filepath to exclusions data (must match the techmap grid). None will not apply exclusions.

• excl_dict (dict | None) – Dictionary of exclusion LayerMask arugments {layer: {kwarg: value}}

• techmap_dset (str) – Dataset name in the techmap file containing the exclusions-to-resource mapping data.

• out_dir (str) – Directory to dump output files.

• job_tag (str | None) – Optional name tag to add to the output files. Format is “rpm_cluster_output_{tag}.csv”.

• max_workers (int, optional) – Number of parallel workers. 1 will run serial, None will use all available., by default None

• excl_area (float | None) – Area in km2 of one exclusion pixel. None will calculate the exclusion pixel area from the exclusion projection profile.

• include_threshold (float) – Inclusion threshold. Resource pixels included more than this threshold will be considered in the representative profiles. Set to zero to find representative profile on all resource, not just included.

• n_profiles (int) – Number of representative profiles to output.

• rerank (bool) – Flag to rerank representative generation profiles after removing excluded generation pixels.

• trg_bins (str | list | None) – TRG bins as an ordered list of bin edge values or string to a csv containing a single column with bin edge values. None will ignore trgs.

• trg_dset (str) – Dataset associated with TRG bins that can be found in the cf_fpath file.

• pre_extract_inclusions (bool) – Flag to pre-extract the inclusion mask using excl_fpath and excl_dict. This is advantageous if the excl_dict is highly complex and if you’re processing a lot of points. Default is False.

• target_crs (str) – Target coordinate reference system, e.g. “EPSG:4326”. Note that everything is in EPSG:4326/WGS84 by default (raw lat/lon with lon=0 at Greenwich).