pykg2vec 0.0.37

A python library for Knowledge Graph Embedding

Python Knowledge Graph Embedding Library

This library is an outcome of a bold and optimistic attempt to bring all the state-of-the-art knowledge graph embedding algorithms into one single python library.

Implemented Methods

We aim to implement all the latest state-of-the-art knowledge graph embedding library. So far these are the implemented algorithms:

Latent Feature Models

These modles utilize a latent feature of either entities or relations to explain the triples of the Knowledge graph. The features are called latent as they are not directly observed. The interaction of the entities and the relations are captured through their latent space representation.

Latent Distance Models

These models utilized the distance-based scoring functions to embed the knowledge graph triples.

• TransE: TransE is an energy based model which represents the relationships as translations in the embedding space. Which means that if (h,l,t) holds then the embedding of the tail 't' should be close to the embedding of head entity 'h' plus some vector that depends on the relationship 'l'. Both entities and relations are vectors in the same space[1].

• TransH: TransH follows the general principle of the TransE. However, compared to it, it introduces relation-specific hyperplanes. The entities are represented as vecotrs just like in TransE, however, the relation is modeled as a vector on its own hyperplane with a normal vector. The entities are then projected to the relation hyperplane to calculate the loss.

• TransR: TransR is pretty similar to TransH, the only difference being that rather than having one relation hyperplane, it introduces relation-specific hyperplanes. The entities are vecotr in entity space and each relation is a vector in relation specific space. For calculating the loss, the entities are projected to relation specific space using the projection matrix.

Semantic Matching Models

Semantic matching models are latent feature models which represents triples by using a pairwise interactions of latent features.

• RESCAL: Rescal is a latent feature model where each relation is represented as a matrix modeling the iteraction between latent factors. It utilizes a weight matrix which specify how much the latent features of head and tail entities interact in the relation.

• Semantic Matching Energy (SME): SME utilizes a neural network architecture to perform the semantic matching. The energy of the triples are computed by a parameterized function which relies on matching criterion computed between both sides of the triples. The semantic energy function learns to distinguish plausible combinations of entities from implausible ones. It consists of two variation SMElinear and SMEbilinear.

Datasets

We intend to provide the libraries to test the knowledge graph algorithms against all the well-known datasets available online. So far the library is able to work with the following datasets:

• Freebase: Freebase is a large collaborative knowledge base consisting of data composed mainly by its community members. It is an online collection of structured data harvested from many sources, including individual, user-submitted wiki contributions [2].

Repository Structure

• pyKG2Vec/config: This folder consists of the configuration module. It provides the necessary configuration to parse the datasets, and also consists of the baseline hyperparameters for the knowledge graph embedding algorithms.
• pyKG2Vec/core: This folder consists of the core codes of the knowledge graph embedding algorithms. Inside this folder, each algorithm is implemented as a separate python module.
• pyKG2Vec/utils: This folders consists of modules providing various utilities, such as data preparation, data visualization, and evaluation of the algorithms.

Dependencies

The goal of this library is to minimize the dependency on other libraries as far as possible to rapidly test the algorithms against different dataset. We emphasize that in the beginning, we will not be focus in run-time performance. However, in the future, may provide faster implementation of each of the algorithms. We encourage installing the tensorflow-gpu version for optimal usage.

• networkx==2.2
• setuptools==40.8.0
• tensorflow==1.13.1
• matplotlib==3.0.3
• numpy==1.16.2
• seaborn==0.9.0
• scikit_learn==0.20.3

Install

For best performance, we encourage the users to create a virtual environment and setup the necessary dependencies for running the algorithms.

Prepare your environment:

```bash
   sudo apt update
   sudo apt install python3-dev python3-pip
   sudo pip3 install -U virtualenv     
```

Create a virtual environment:

```bash
   virtualenv --system-site-packages -p python3 ./venv
```

Activate the virtual environment using a shell-specific command:

```bash
   source ./venv/bin/activate
``` 

Upgrade pip:

```bash
   pip install --upgrade pip
```

Install pyKG2Vec:

```bash
   (venv) $ pip install pykg2vec
``` 

Usage Example

```python
    import tensorflow as tf
    from pykg2vec.config.config import TransEConfig
    from pykg2vec.utils.dataprep import DataPrep
    from argparse import ArgumentParser
    from pykg2vec.core.TransE import TransE
    import os


    def main(_):
        parser = ArgumentParser(description='Knowledge Graph Embedding with TransE')
        parser.add_argument('-b', '--batch', default=128, type=int, help='batch size')
        parser.add_argument('-t', '--tmp', default='./intermediate', type=str, help='Temporary folder')
        parser.add_argument('-ds', '--dataset', default='Freebase', type=str, help='Dataset')
        parser.add_argument('-l', '--epochs', default=10, type=int, help='Number of Epochs')
        parser.add_argument('-tn', '--test_num', default=5, type=int, help='Number of test triples')
        parser.add_argument('-ts', '--test_step', default=5, type=int, help='Test every _ epochs')
        parser.add_argument('-lr', '--learn_rate', default=0.01, type=float, help='learning rate')
        parser.add_argument('-gp', '--gpu_frac', default=0.4, type=float, help='GPU fraction to use')

        args = parser.parse_args()

        if not os.path.exists(args.tmp):
            os.mkdir(args.tmp)

        data_handler = DataPrep(args.dataset)

        config = TransEConfig(learning_rate=args.learn_rate,
                              batch_size=args.batch,
                              epochs=args.epochs,
                              test_step=args.test_step,
                              test_num=args.test_num,
                              gpu_fraction=args.gpu_frac)

        model = TransE(config=config,data_handler=data_handler)
        model.summary()
        model.train()


    if __name__ == "__main__":
        tf.app.run()
```  

The output of code will be as follows:

```angular2
    Number of batches: 461
    iter[0] ---Train Loss: 53589.23212 ---time: 3.06
    ---------------Test Results: iter0------------------
    iter:0 --mean rank: 11484.90 --hit@10: 0.10
    iter:0 --filter mean rank: 11484.90 --filter hit@10: 0.10
    iter:0 --norm mean rank: 11413.10 --norm hit@10: 0.10
    iter:0 --norm filter mean rank: 11413.10 --norm filter hit@10: 0.10
    -----------------------------------------------------
    iter[1] ---Train Loss: 45554.93086 ---time: 2.91
    iter[2] ---Train Loss: 42788.54883 ---time: 2.80
    iter[3] ---Train Loss: 40905.94272 ---time: 2.76
    iter[4] ---Train Loss: 39643.85038 ---time: 2.68
    iter[5] ---Train Loss: 39153.04682 ---time: 2.87
    iter:5 --mean rank: 19915.50 --hit@10: 0.10
    iter:5 --filter mean rank: 19915.50 --filter hit@10: 0.10
    iter:5 --norm mean rank: 19856.20 --norm hit@10: 0.20
    iter:5 --norm filter mean rank: 19856.20 --norm filter hit@10: 0.20
    iter[6] ---Train Loss: 38518.77916 ---time: 2.66
    iter[7] ---Train Loss: 38320.69923 ---time: 2.60
    iter[8] ---Train Loss: 37865.60836 ---time: 2.58
    iter[9] ---Train Loss: 37619.98050 ---time: 2.51
    iter:9 --mean rank: 27790.10 --hit@10: 0.10
    iter:9 --filter mean rank: 27790.10 --filter hit@10: 0.10
    iter:9 --norm mean rank: 28324.30 --norm hit@10: 0.20
    iter:9 --norm filter mean rank: 28324.30 --norm filter hit@10: 0.20

    ----------SUMMARY----------
           margin : 1.0
           epochs : 10
     loadFromData : False
    disp_triple_num : 5
         test_num : 5
         testFlag : False
        test_step : 5
        optimizer : gradient
          L1_flag : True
       batch_size : 128
    learning_rate : 0.01
             data : Freebase
      hidden_size : 100
    ---------------------------
         reducing dimension to 2 using TSNE!
    dimension self.h_emb (5, 100)
    dimension self.r_emb (5, 100)
    dimension self.t_emb (5, 100)
    dimension self.h_emb (5, 2)
    dimension self.r_emb (5, 2)
    dimension self.t_emb (5, 2)
         drawing figure!
```

The red nodes represent head entity, green nodes represent the relations and the blue node represents the tail entities.

Learn More

Here are some links to get you started with knowledge graph embedding methodologies.

• A Review of Relational Machine Learning for Knowledge Graphs
• Knowledge Graph Embedding: A Survey of Approaches and Applications

Cite

Please kindly cite us if you found the library helpful.

@online{pykg2vec,
author = {Rokka Chhetri, Sujit and  Yu, Shih-Yuan and  Salih Aksakal, Ahmet and  Goyal, Palash and  Canedo Arquimedes, Martinez},
title = {{pykg2vec: Python Knowledge Graph Embedding Library},
year = 2019,
url = {https://pypi.org/project/pykg2vec/}
}