Emgraph

Emgraph is a Python toolkit for graph embedding.

$ pip install emgraph

Simple example

Embedding wordnet11 graph using TransE model:

from sklearn.metrics import brier_score_loss, log_loss
from scipy.special import expit
from emgraph.datasets import BaseDataset, DatasetType
from emgraph.models import TransE


def train_transe():
    X = BaseDataset.load_dataset(DatasetType.WN11)
    model = TransE(batches_count=64, seed=0, epochs=20, k=100, eta=20,
                   optimizer='adam', optimizer_params={'lr': 0.0001},
                   loss='pairwise', verbose=True, large_graphs=False)

    model.fit(X['train'])

    scores = model.predict(X['test'])

    print("Scores: ", scores)
    print("Brier score loss:", brier_score_loss(X['test_labels'], expit(scores)))


# Executing the function

if __name__ == '__main__':
    train_transe()

Evaluating ComplEx model after training:

import numpy as np
from emgraph.datasets import BaseDataset, DatasetType
from emgraph.models import ComplEx
from emgraph.evaluation import evaluate_performance


def complex_performance():
    X = BaseDataset.load_dataset(DatasetType.WN18)
    model = ComplEx(batches_count=10, seed=0, epochs=20, k=150, eta=1,
                    loss='nll', optimizer='adam')
    model.fit(np.concatenate((X['train'], X['valid'])))
    filter_triples = np.concatenate((X['train'], X['valid'], X['test']))
    ranks = evaluate_performance(X['test'][:5], model=model,
                                 filter_triples=filter_triples,
                                 corrupt_side='s+o',
                                 use_default_protocol=False)
    return ranks


# Executing the function

if __name__ == '__main__':
    ranks = complex_performance()
    print("ranks {}".format(ranks))

More examples

Embedding wordnet11 graph using DistMult model:

from sklearn.metrics import brier_score_loss, log_loss
from scipy.special import expit
from emgraph.datasets import BaseDataset, DatasetType
from emgraph.models import DistMult


def train_dist_mult():
    X = BaseDataset.load_dataset(DatasetType.WN11)
    model = DistMult(batches_count=1, seed=555, epochs=20, k=10, loss='pairwise',
                     loss_params={'margin': 5})

    model.fit(X['train'])

    scores = model.predict(X['test'])

    print("Scores: ", scores)
    print("Brier score loss:", brier_score_loss(X['test_labels'], expit(scores)))


# Executing the function

if __name__ == '__main__':
    train_dist_mult()

Future work

• Modulate the functions
• Add more algorithms
• Run on CUDA cores
• Make it faster using vectorization etc.
• Add more preprocessors
• Add dataset, graph, and dataframe manipulations
• Unify and reconstruct the architecture and eliminate redundancy

If you found it helpful, please give us a :star:

License

Released under the BSD license

Copyright © 2019-2022 Emgraph Developers
Soran Ghaderi (soran.gdr.cs@gmail.com)
Taleb Zarhesh (taleb.zarhesh@gmail.com)