simon987/music-graph-scripts
1
0
Fork 0
mirror of https://github.com/simon987/music-graph-scripts.git synced 2025-04-19 10:26:43 +00:00
Code Issues Projects Releases Wiki Activity

Normalize tag weight

Browse Source
This commit is contained in:
simon 2019-05-25 08:51:53 -04:00
parent 14327c62cc
commit 6460e98ce0
5 changed files with 30 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
make_neoj4_db.sh
View File

@ -1,6 +1,6 @@
#!/bin/bash #!/usr/bin/env bash
export NEO4J_HOME="/home/drone/Downloads/neo4j-community-3.5.3" export NEO4J_HOME="/home/drone/Documents/neo4j"
export REPOSITORY="http://localhost:9999" export REPOSITORY="http://localhost:9999"
export DATABASE="graph.db" export DATABASE="graph.db"

0
make_release_to_rg_map.py → map_release_to_rg_map.py
View File

33
process_mb_dump.py
View File

@ -313,40 +313,59 @@ with open("in/tag") as f:
out.write(cols[0] + ",\"" + cols[1].replace("\"", "\"\"") + "\"\n") out.write(cols[0] + ",\"" + cols[1].replace("\"", "\"\"") + "\"\n")
with open("repo/release_tag.csv", "w") as out: with open("repo/release_tag.csv", "w") as out:
out.write(":START_ID(Release),:END_ID(Tag),weight:int\n") out.write(":START_ID(Release),:END_ID(Tag),weight:float\n")
# get max count
max_count = 0
with open("in/release_group_tag") as f: with open("in/release_group_tag") as f:
for line in f: for line in f:
cols = line.split("\t") cols = line.split("\t")
max_count = max(max_count, int(cols[2]))
max_count = max_count / 4
if int(cols[2]) <= 0: # weight is linear
with open("in/release_group_tag") as f:
for line in f:
cols = line.split("\t")
count = int(cols[2])
if count <= 0:
continue continue
out.write(",".join(( out.write(",".join((
release_groups[cols[0]][1], release_groups[cols[0]][1],
cols[1], cols[1],
cols[2], str(max(min(count / max_count, 1), 0.2)),
)) + "\n") )) + "\n")
with open("repo/artist_tag.csv", "w") as out: with open("repo/artist_tag.csv", "w") as out:
out.write(":START_ID(Artist),:END_ID(Tag),weight:int\n") out.write(":START_ID(Artist),:END_ID(Tag),weight:float\n")
# get max count
max_count = 0
with open("in/artist_tag") as f:
for line in f:
cols = line.split("\t")
max_count = max(max_count, int(cols[2]))
max_count = max_count / 4
# Weight is linear
with open("in/artist_tag") as f: with open("in/artist_tag") as f:
for line in f: for line in f:
cols = line.split("\t") cols = line.split("\t")
if int(cols[2]) <= 0: count = int(cols[2])
if count <= 0:
continue continue
out.write(",".join(( out.write(",".join((
artists[cols[0]][1], artists[cols[0]][1],
cols[1], cols[1],
cols[2], str(max(min(count / max_count, 1), 0.2)),
)) + "\n") )) + "\n")
with open("repo/tag_tag.csv", "w") as out: with open("repo/tag_tag.csv", "w") as out:
out.write(":START_ID(Tag),:END_ID(Tag),weight:int\n") out.write(":START_ID(Tag),:END_ID(Tag),weight:int\n")
# TODO: normalize weight so it's between [0,1]
with open("in/tag_relation") as f: with open("in/tag_relation") as f:
for line in f: for line in f:
cols = line.split("\t") cols = line.split("\t")

1
seed.cypher
View File

@ -1,2 +1,3 @@
CREATE INDEX ON :Artist(id); CREATE INDEX ON :Artist(id);
CREATE INDEX ON :Artist(name);
CREATE INDEX ON :Release(id); CREATE INDEX ON :Release(id);

2
seed_neo4j_db.sh
View File

@ -1,5 +1,5 @@
#!/usr/bin/env bash #!/usr/bin/env bash
export NEO4J_HOME="/home/drone/Downloads/neo4j-community-3.5.3" export NEO4J_HOME="/home/drone/Documents/neo4j"
cat seed.cypher | ${NEO4J_HOME}/bin/cypher-shell cat seed.cypher | ${NEO4J_HOME}/bin/cypher-shell