Skip to content

README.md

Latest commit

 

History

History
95 lines (66 loc) · 2.35 KB

README.md

File metadata and controls

95 lines (66 loc) · 2.35 KB

RAKI Demo

The RAKI Demo consists of the Biopax data by default and provides the RAKI Endpoint such as the DRILL Endpoint and the OWL Verbalizer Endpoint.

Deploy

Read deploy with your data to deploy the RAKI Demo with other data.

Clone

Clones this repository to your local folder: raki-deploy.

git clone --recurse-submodules https://github.com/raki-project/raki-deploy.git

Build

Builds your application in your local folder.

cd drill && unzip LPs.zip
sudo docker compose build

Start

Starts your application:

sudo docker compose up

Deploy with your Data

For Drill: update the docker-compose.yml file and add your data to the Drill service build environment. For the Verbalizer: copy your ontology to the server/ontology folder.

RAKI Endpoint

Send a HTTP POST request that requires two parameters to http://localhost:9081/raki

The two parameters are:

  • input A JSON file that contains an input for Drill as described in DRILL,

  • ontology A RDF/OWL file, an ontology.

By default the rule-based verbalizer is set. The parameter type with the value model will switch to the trained network in the beta version (e.g., http://localhost:9081/raki?type=model).

GUI

In your browser open http://localhost:9081/ to request the GUI.

Example

An input can be created with, for instance:

jq '
   .problems
     ."((pathwayStep ⊓ (∀INTERACTION-TYPE.Thing)) ⊔ (sequenceInterval ⊓ (∀ID-VERSION.Thing)))"
   | {
      "positives": .positive_examples,
      "negatives": .negative_examples
     }' LPs/Biopax/lp.json > input.json

The files in the LPs folder are given in DRILL.

Request example with Curl:

ontology="@biopax.owl"
input="@input.json"
response=$0.json

curl \
	-F ontology=$ontology \
	-F input=$input \
	-H "charset=utf-8" \
	-o $response \
	http://localhost:9081/raki

DRILL Endpoint

Send a HTTP request to http://localhost:9080/concept_learning

Details to parameters: DRILL

OWL Verbalizer Endpoint

Send a HTTP POST request that requires two parameters to http://localhost:9081/verbalize

The two parameters are:

  • axioms A RDF/OWL file to verbalize.

  • ontology A RDF/OWL file, an ontology.