Skip to content

Commit

Permalink
Merge 6059d5c into e80ac65
Browse files Browse the repository at this point in the history
  • Loading branch information
justaddcoffee committed Aug 10, 2020
2 parents e80ac65 + 6059d5c commit 0498ad6
Show file tree
Hide file tree
Showing 4 changed files with 46 additions and 0 deletions.
8 changes: 8 additions & 0 deletions Jenkinsfile
Original file line number Diff line number Diff line change
Expand Up @@ -100,6 +100,14 @@ pipeline {
}
}

stage('QC'){
steps {
dir('./gitrepo') {
sh '. venv/bin/activate && python3.7 queries/qc/ensmallen_report.py'
}
}
}

stage('Make blazegraph journal'){
steps {
dir('./gitrepo/blazegraph') {
Expand Down
26 changes: 26 additions & 0 deletions queries/qc/ensmallen_report.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,26 @@
from ensmallen_graph import EnsmallenGraph
import tarfile
import compress_json

tar = tarfile.open("kg-covid-19.tar.gz")

tar.extractall()
graph = EnsmallenGraph.from_csv(
edge_path="merged-kg_edges.tsv",
sources_column="subject",
destinations_column="object",
directed=False,
edge_types_column="edge_label",
default_edge_type="biolink:association",
node_path="merged-kg_nodes.tsv",
nodes_column="id",
node_types_column="category",
default_node_type="biolink:NamedThing",
ignore_duplicated_edges=True,
ignore_duplicated_nodes=True,
force_conversion_to_undirected=True
)

json_report = graph.report()
compress_json.dump(json_report, "kg-covid-19-ensmallen-report.json")

12 changes: 12 additions & 0 deletions queries/qc/ensmallen_report_expected_values.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
degrees_mode
bidirectional_rate
strongly_connected_components_number
nodes_number
edges_number
density
singleton_nodes
unique_edge_types_number
connected_components_number
unique_node_types_number
degrees_median

Binary file added tests/resources/qc/kg-covid-19-TEST.tar.gz
Binary file not shown.

0 comments on commit 0498ad6

Please sign in to comment.