Home

Awesome

jRDF2Vec

Java CI Python Tests Publish Docker image Coverage Status License

jRDF2Vec is a Java implementation of <a href="http://rdf2vec.org/">RDF2Vec</a>. It supports multi-threaded, in-memory (or disk-access-based) walk generation and training. You can generate embeddings for any NT, NQ, OWL/XML, RDF HDT, TDB 1, or TTL file.

Found a bug? Don't hesitate to <a href="https://github.com/dwslab/jRDF2Vec/issues">open an issue</a>.

How to cite?

Portisch, Jan; Hladik, Michael; Paulheim, Heiko. RDF2Vec Light - A Lightweight Approach for Knowledge Graph Embeddings. Proceedings of the ISWC 2020 Posters & Demonstrations. 2020. [to appear]

An open-access version of the paper is available here.

How to use the jRDF2Vec Command-Line Interface?

Download this project, execute mvn clean install. Alternatively, you can download the packaged JAR of the latest successful: commit <a href="https://github.com/dwslab/jRDF2Vec/tree/jars/jars">here</a>.

System Requirements

You can check if you set up the environment (Python 3 + dependencies) correctly by running:

java -jar jrdf2vec-1.1-SNAPSHOT.jar -checkInstallation

The command line output will list missing requirements or print Installation is ok ✔.

Command-Line Interface (jRDF2Vec CLI) for Training and Walk Generation

Use the resulting jar from the target directory.

Minimal Example

java -jar jrdf2vec-1.1-SNAPSHOT.jar -graph ./kg_file.hdt

Required Parameters

Optional Parameters

jRDF2Vec follows the <a href="https://en.wikipedia.org/wiki/Convention_over_configuration">convention over configuration</a> design paradigm to increase usability. You can overwrite the default values by setting one or more optional parameters.

Parameters for the Walk Configuration

Parameters for the Training Configuration

Advanced Parameters

Command-Line Interface (jRDF2Vec CLI) - Additional Services

Besides generating walks and training embeddings, the CLI offers additional services which are described below.

Generating a Vector Text File

(1) Full Vocabulary<br/> jRDF2vec is compatible with the <a href="https://github.com/mariaangelapellegrino/Evaluation-Framework">evaluation framework for KG embeddings (GEval)</a>. The latter framework requires the vectors to be present in a text file. If you have a gensim model or vector file, you can use the following command to generate this file:

java -jar jrdf2vec-1.1-SNAPSHOT.jar -generateTextVectorFile ./path-to-your-model-or-vector-file

You can find the file (named vectors.txt) in the directory where the model/vector file is located. If you want to specify the file name/path yourself, you can use option -newFile <file_path>.

(2) Subset of the Vocabulary<br/> If you want to write a vectors.txt file that contains only a subset of the vocabulary, you can additionally specify the entities of interest using the -light <entity_file> option (The <entity_file> should contain one entity (full URI) per line.):

java -jar jrdf2vec-1.1-SNAPSHOT.jar -generateTextVectorFile ./path-to-your-model-or-vector-file -light ./path-to-entity-file

You can find the file (named vectors.txt) in the directory where the model/vector file is located. If you want to specify the file name/path yourself, you can use option -newFile <file_path>. If the vector concepts contain surrounding tags that you want to remove in the process, use option -noTags. This command also works if ./path-to-your-model-or-vector-file is an existing vector text file that shall be reduced.

Generating a Vocabulary Text File

jRDF2vec provides functionality to print all concepts for which a vector has been trained. One word of the vocabulary will be printed per line to a file named vocabulary.txt. The model or vector file needs to be specified. If you have a gensim model or vector file, you can use the following command to generate this file:

java -jar jrdf2vec-1.1-SNAPSHOT.jar -generateVocabularyFile ./path-to-your-model-or-vector-file

Converting a Text Vector File

jRDF2vec generates a vectors.txt file where one line represents a vector. This is the format also used by GloVe, for instance. In some cases, however, other file formats are required. You can use jRDF2vec to convert text vector files to other common formats. The vector file does not have to be generated by jRDF2vec.

(1) Converting to w2v Format<br/> To create a word2vec formatted file from the text file, you can use the following command:

java -jar jrdf2vec-1.1-SNAPSHOT.jar -convertToW2V <txt_file_path> <new_file.w2v>

(2) Converting to kv Format<br/> The provided txt file (first parameter) can be either in txt format or in w2v format. Make sure you use the correct file ending (.txt/.w2v).

You can run the command as follows:

java -jar jrdf2vec-1.1-SNAPSHOT.jar -convertToKv <txt_file_path> <new_file.kv>

(3) Converting to Tensorflow Projector Format<br/> If you want to visualize your embedding space by using the Tensorflow Projector, you can do so by converting your vectors.txt file to the two files required by the tool. Use the following command:

java -jar jrdf2vec-1.1-SNAPSHOT.jar -convertToTfProjector <txt_file_path> [<vectors.tsv> <metadata.tsv>]

Two additional .tsv files will be generated. You can find them in the same directory where <txt_file_path> is located.

Optionally, you can specify the paths of the files to be written as indicated in the command above.

Analyzing the Embedding Vocabulary

For RDF2Vec, it is not always guaranteed that all concepts in the graph appear in the embedding space. For example, some concepts may only appear in the object position of statements and may never be reached by random walks. In addition, the word2vec configuration parameters may filter out infrequent words depending on the configuration (see -minCount above, for example). To analyze such rather seldom cases, you can use the -analyzeVocab function specified as follows:

java -jar jrdf2vec-1.1-SNAPSHOT.jar -analyzeVocab <model> <training_file|entity_file>

A report will be printed. For large models, you may want to redirect that into a file ([...] &> somefile.txt).

Merge of All Walk Files Into One

By default, jRDF2vec serializes walks in different gzipped files. If you require a single, uncompressed file, you can use the -mergeWalks keyword. You need to provide a -walkDirectory <dir> and you can optionally specify the output file using -o <file_path>. (Files not ending with .gz in <dir> will be skipped.)

java -jar jrdf2vec-1.1-SNAPSHOT.jar -mergeWalks -walkDirectory <dir> -o <file_to_write>

Converting the Graph File

(1) Converting to PajekNet<br/> To create a graph file in the <a href="https://gephi.org/users/supported-graph-formats/pajek-net-format/">PajekNet format</a> (e.g. for graph analysis), you can use the following command:

java -jar jrdf2vec-1.1-SNAPSHOT.jar -convertToPajek <graph> <file_to_write>

How to use the jRDF2Vec as library in Java projects?

Stable releases are available through the maven central repository:

<dependency>
    <groupId>de.uni-mannheim.informatik.dws</groupId>
    <artifactId>jrdf2vec</artifactId>
    <version>1.0</version>
</dependency>

Run jRDF2Vec using Docker

Publish Docker image

Optionally, Docker can be used to run jRDF2Vec. This functionality has been added by <a href="https://github.com/vemonet">Vincent Emonet</a>.

Run

The Docker image can be used with the same arguments as the Jar file, refer to the documentation above for more details on the different jRDF2Vec arguments.

Test run to get the help message:

docker run -it --rm ghcr.io/dwslab/jrdf2vec -help

The best way to mount your local files in the docker container is to mount a folder on /data in the container:

Here is an example generating embeddings using sample config files for DBpedia found in src/test/resources in this repository. Use this command from the root folder of this repository on Linux or MacOS, change the $(pwd) to ${PWD} for Windows:

docker run -it --rm \
  -v $(pwd):/data \
  ghcr.io/dwslab/jrdf2vec \
  -light /data/src/test/resources/sample_dbpedia_entity_file.txt \
  -graph /data/src/test/resources/sample_dbpedia_nt_file.nt

Embeddings will be generated in the folders walks and python_server from where you ran the command.

Build

A new docker image is automatically built and published to the GitHub Container Registry by a GitHub Actions workflow:

Build from source code:

docker build -t ghcr.io/dwslab/jrdf2vec .

Developer Documentation

The most recent JavaDoc sites generated from the latest commit can be found <a href="https://dwslab.github.io/jRDF2Vec/">here</a>.<br/>

Special Applications

Ordered RDF2Vec ("Putting RDF2vec in Order")

The following steps are necessary to obtain ordered RDF2vec embeddings (see publication Putting RDF2vec in Order for conceptional details).

Step 1: Generate Walks<br/> Run jRDF2Vec to generate only walks (option -onlyWalks) on your desired dataset.

Step 2: Merge the Walks in a single, uncompressed file<br/> By default, jRDF2Vec serializes the walks in multiple gzipped files. For this application, however, we need a single, uncompressed walk file.

You can use the corresponding jRDF2Vec command line service to do so.

Step 3: Compile wang2vec<br/> Download the C implementation of wang2vec from GitHub. Compile the files with make.

Step 4: Run and have fun<br/> Run the compiled wang2vec implementation on the merged walk file from step 2. In case you receive a segfault error, set the capping parameter to 1 (-cap 1).

Call Syntax<br/>

./word2vec -train <your walk file> -output <desired file to be written> -type <2 (cwindow) or 3 (structured 
skipgram>) -size <vector size> -threads <number of threads> -min-count 0 -cap 1

Exemplary Call<br/>

./word2vec -train walks.txt -output v100.txt -type 3 -size 100 -threads 4 -min-count 0 -cap 1

Not working? Contact us or open an issue.

Please do not forget to cite the corresponding papers:

(1)  Portisch, Jan; Paulheim, Heiko. Putting RDF2vec in Order. In: Proceedings of the International Semantic Web 
Conference - Posters and Demos, ISWC 2021. 2021. 

(2) Ling, Wang; Dyer, Chris; Black, Alan; Trancoso, Isabel. Two/too simple adaptations of word2vec for syntax 
problems. In: NAACL HLT 2015. pp. 1299–1304. ACL (2015)

Frequently Asked Questions (FAQs)

I have Python installed, but it is not accessible via command python. How to resolve this?<br/> Create a file python_command.txt in directory ./python_server (created when first running the jar). Write the command to call Python 3 in the first line of the file.

The program starts and immediately shuts down. Nothing seems to happen.<br/> Make sure your system is set-up correctly, in particular whether you have installed Python 3 and the required dependencies.

Can I run the command multiple times in parallel on the same machine?<br/> Yes, you can. You need to make sure that for each command, you use (1) a different -port and (2) a different -walkDirectory.