Light

vaibhavad / python-wrapper-openie5 Goto Github PK

View Code? Open in Web Editor NEW

41.0 1.0 5.0 10 KB

Python wrapper for OpenIE5

Home Page: https://pypi.org/project/pyopenie/

License: MIT License

Python 100.00%

openie5-server openie information-extraction python-wrapper

python-wrapper-openie5's Introduction

pyopenie

Python wrapper for OpenIE5. This simply wraps the API from the server included with OpenIE 5.1.0.

Install

pip install pyopenie

Usage

First make sure you have the OpenIE5 server running. See the instructions here for how to do that.

Then the setup just requires you to pass in the url of the server:

>>> from pyopenie import OpenIE5
>>> extractor = OpenIE5('http://localhost:9000')

Any English sentence can be passed to OpenIE5 server.

>>> extractions = extractor.extract("The U.S. president Barack Obama gave his speech to thousands of people.")

The result is a JSON list of extractions with confidence, offset and other properties.

>>> extractions
[
    {
        "confidence": 0.38089450366724514,
        "sentence": "The U.S. president Barack Obama gave his speech on Tuesday to thousands of people.",
        "extraction": {
            "arg1": {
                "text": "Barack Obama",
                "offsets": [...]
            },
            "rel": {
                "text": "[is] president [of]",
                "offsets": [...]
            },
            "arg2s": [
                {
                    "text": "United States",
                    "offsets": [...]
                }
            ],
            "context": null,
            "negated": false,
            "passive": false
        }
    },
    {
        "confidence": 0.9168198459177435,
        "sentence": "The U.S. president Barack Obama gave his speech on Tuesday to thousands of people.",
        "extraction": {
            "arg1": {
                "text": "The U.S. president Barack Obama",
                "offsets": [...]
            },
            "rel": {
                "text": "gave",
                "offsets": [...]
            },
            "arg2s": [
                {
                    "text": "his speech",
                    "offsets": [...]
                },
                {
                    "text": "on Tuesday",
                    "offsets": [...]
                },
                {
                    "text": "to thousands of people",
                    "offsets": [...]
                }
            ],
            "context": null,
            "negated": false,
            "passive": false
        }
    }
]

Individual properties can also be accessed.

>>> extractions[0]['confidence']
0.38089450366724514

>>> extractions[0]['extraction']['arg1']['text']
'Barack Obama'

python-wrapper-openie5's People

Contributors

Stargazers

Watchers

Forkers

bwang12 zypchenfeng mlpotter wangqiqing yuanweij

python-wrapper-openie5's Issues

How to compute confidence score？

Hi,

I was wondering how you get the confidence score for every extracted triple?

requests.exceptions.ConnectionError: ('Connection aborted.', RemoteDisconnected('Remote end closed connection without response',))

Hello, i'm having this issue when trying to segmentation then send every sentence to extract, i don't use --split because i have to reuse those sentences later.
Help!

Server does not seem to be able to take incoming requests from multiple IPs

I am hosting the server on an EC2 instance. I then make a request from another computer and it seems to work fine. However, when a second computer tries to make a request, the code hangs and the request doesn't seem to be able to make it to the server. Weirdly enough, the original computer will start having the same problem too.

Offset

Hi @vaibhavad ,

Can you please elaborate more on the offset with an example?

Thanks.

<class 'json.decoder.JSONDecodeError'>

For some reason, I am getting a lot of these issues.

I am loading the text from .txt for context
bob_dylan.txt

Error:
<class 'json.decoder.JSONDecodeError'>

Would it be possible that I need to parse the text through a json first before sending it off? Or is this a problem in the response? Are all characters accepted?

How do I pass properties such as split, binary?

I have tried passing a dictionary such as properties = {'binary':True, 'split': True} when invoking extractor.extract but it is not making a difference.

Can you provide a sample of the colab version?

Hello, I can't get started quickly in the process of using it. Could you please provide a colab（notebook） version of the tutorial?
In this way, it helps me to apply quickly.

How to get SRL result in retrieved triples ?

HttpResponse receive through API server currently does not contain SRL arguments (SpatialArgument and TemporalArgument).

Can you guide me a way to get those arguments?

Failing sentences

The sentences containing “ fail, maybe an encoding issue with my documents.

JSONDecodeError: Expecting value: line 1 column 1 (char 0)

A very long text leads to the following error:
JSONDecodeError: Expecting value: line 1 column 1 (char 0)
Server started with
java -Xmx10-XX:+UseConcMarkSweepGC -jar openie-assembly-5.0-SNAPSHOT.jar --split --ignore-errors --httpPort 8000
The input text was text extracted from a research paper.

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.