reasonable 0.2.1-alpha13

An OWL 2 RL reasoner with reasonable performance
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125

import rdflib
import glob
import brickschema
import time

def load_file(g, f):
    if f.endswith('.ttl'):
        g.parse(f, format='ttl')
    elif f.endswith('.n3'):
        g.parse(f, format='ntriples')

ontology_files = [
    "../../example_models/ontologies/Brick.n3",
    "../../example_models/ontologies/owl.n3",
    "../../example_models/ontologies/rdfs.ttl",
]


# data_files = [
#     "../../example_models/small1.n3",
#     "../../example_models/soda_hall.n3",
# ]
data_files = glob.glob("buildings/*.ttl")
data_sizes = {}
for f in data_files:
    g = rdflib.Graph()
    g.parse(f, format='ttl')
    data_sizes[f] = len(g)

N = 1

all_samples = {
    'owlrl': {},
    'allegro': {},
    'owlready2': {},
    'reasonable': {},
}


# owlready2
import owlready2
for data_file in data_files:
    print(f"Benching owlready2 on {data_file}")
    all_samples['owlready2'][data_file] = []
    for i in range(N):
        g = rdflib.Graph()
        load_file(g, data_file)
        for ont_file in ontology_files:
            load_file(g, ont_file)
        g.serialize("_owlready2_input.ttl", format="ttl")
        world = owlready2.World()
        onto = world.get_ontology(f"file://./_owlready2_input.ttl").load()
        t0 = time.time()
        owlready2.sync_reasoner(world, infer_property_values =True)
        G = world.as_rdflib_graph()
        t1 = time.time()
        print(f"    owlready2: {data_file} took {t1-t0}")
        all_samples['owlready2'][data_file].append({'duration': t1-t0, 'triples': len(G)})


# OWLRL
import owlrl
for data_file in data_files:
    print(f"Benching owlrl on {data_file}")
    all_samples['owlrl'][data_file] = []
    for i in range(N):
        g = rdflib.Graph()
        load_file(g, data_file)
        for ont_file in ontology_files:
            load_file(g, ont_file)

        t0 = time.time()
        owlrl.DeductiveClosure(owlrl.OWLRL_Semantics).expand(g)
        t1 = time.time()
        print(f"    owlrl: {data_file} took {t1-t0}")
        all_samples['owlrl'][data_file].append({'duration': t1-t0, 'triples': len(g)})

# Allegro
from brickschema.inference import OWLRLAllegroInferenceSession
for data_file in data_files:
    print(f"Benching owlrl on {data_file}")
    all_samples['allegro'][data_file] = []
    for i in range(N):
        g = rdflib.Graph()
        load_file(g, data_file)
        for ont_file in ontology_files:
            load_file(g, ont_file)
        sess = OWLRLAllegroInferenceSession()

        t0 = time.time()
        g = sess.expand(g)
        t1 = time.time()
        print(f"    allegro: {data_file} took {t1-t0}")
        all_samples['allegro'][data_file].append({'duration': t1-t0, 'triples': len(g)})

# reasonable
import reasonable
for data_file in data_files:
    print(f"Benching reasonable on {data_file}")
    all_samples['reasonable'][data_file] = []
    for i in range(N):
        g = rdflib.Graph()
        load_file(g, data_file)
        for ont_file in ontology_files:
            load_file(g, ont_file)
        r = reasonable.PyReasoner()
        r.from_graph(g)

        t0 = time.time()
        triples = r.reason()
        t1 = time.time()
        print(f"    reasonable: {data_file} took {t1-t0}")
        all_samples['reasonable'][data_file].append({'duration': t1-t0, 'triples': len(triples)})


import pandas as pd
records = []
for reasoner, defn in all_samples.items():
    for data_file, samples in defn.items():
        for sample in samples:
            records.append((reasoner, data_file, sample['duration'], sample['triples']))
df = pd.DataFrame.from_records(records)
df.columns=['reasoner','data_file','duration','triples']
df.loc[:, 'src_size'] = df['data_file'].map(lambda x: data_sizes[x])
df.to_csv('benchmark.csv', header=True, index=False)