Giter Site home page Giter Site logo

daens's Introduction

DAENS

Requirements:

torch 1.11.0+cu113

torch-geometric 2.1.0.post1

Python 3.9+

Results from Paper

Algo Cora CiteSeer Dblp PubMed WikiCS Am.Comp Am.Photo Co.CS
DGI 82.6±0.4 68.8±0.7 83.2±0.1 86.0±0.1 75.35±0.14 83.95±0.47 91.61±0.22 92.15±0.63
GRACE 83.3±0.4 72.1±0.5 84.2±0.1 86.7±0.1 80.14±0.48 89.53±0.35 92.78±0.45 91.12±0.20
GCA N/A N/A N/A N/A 78.35±0.05 88.94±0.15 92.53±0.16 93.10±0.01
BGRL 83.83±1.61 72.32±0.89 84.07±0.23 86.03±0.33 79.98±0.10 *90.34±0.19 93.17±0.30 93.31±0.13
MVGRL *86.80±0.5 73.30±0.5 N/A 80.10±0.70 77.52±0.08 87.52±0.11 91.74±0.07 92.11±0.12
LG2AR 82.70±0.70 N/A N/A 81.50±0.70 77.80±0.50 89.60±0.30 94.10±0.40 93.60±0.30
AFGRL N/A N/A N/A N/A 77.62±0.49 89.88±0.33 93.22±0.28 93.27±0.17
COSTA 84.30±0.20 72.90±0.30 84.50±0.10 86.20±0.10 79.12±0.02 88.32±0.03 92.56±0.45 92.95±0.12
AFGCL 83.16±0.13 71.96±0.42 N/A 79.05±0.75 79.01±0.51 89.68±0.19 92.49±0.31 91.92±0.10
GRACE* 84.57±0.35 74.29±0.33 84.36±0.31 85.96±0.23 80.51±0.41 89.88±0.14 93.04±0.14 OOM
LG2AR* 83.56±0.64 72.11±0.35 83.34±0.53 81.32±0.61 78.23±0.29 89.57±0.35 94.07±0.41 93.45±0.48
BGRL* 83.87±0.76 73.01±0.46 83.96±0.18 85.94±0.38 79.86±0.11 *90.49±0.17 93.65±0.21 93.25±0.15
DAENS_C 87.13±0.00 79.64±0.00 86.07±0.76 87.48±0.04 83.29±0.21 90.44±0.19 93.44±0.25 93.19±0.13
DAENS_T 89.71±0.00 75.81±0.29 84.99±0.14 86.72±0.06 83.62±0.00 91.60±0.24 94.68±0.42 93.96±0.04

Hyper-parameter Settings

The table below presents the Hyper-parameter settings while the file HyperParameters.py``HyperParameters.py contains these settings with seeds contains these settings with seed values to regenerate the above results.

Dataset $p_{e1}$ $p_{f1}$ $p_{e2}$ $p_{f2} = F_d \times F_r$ Encoder Hetro Training epochs Learning rate Activation function Hidden layers
Cora 0.4 0.4 0.2 0.7 × 0.43 GAT 1D,2D 750 0.0005 ReLU 128
CiteSeer 0.4 0.4 0.2 0.8 × 0.375 GCN 2D,2D 1200 0.00001 PReLU 256
DBLP 0.1 0.1 0.2 0.375 × 0.8 GCN 2D,2D 1450 0.001 ReLU 256
PubMed 0.1 0.1 0.2 0.43 × 0.7 GCN 1D,2D 1800 0.001 ReLU 256
WikiCS 0.4 0.1 0.3 0.4 × 0.75 GAT 1D,2D 1800 0.01 ReLU 256
Am.Comp 0.5 0.1 0.5 0.7 × 0.43 GAT 2D,2D 1900 0.001 ReLU 128
Am.Photo 0.5 0.1 0.3 0.8 × 0.375 GAT 1D,2D 1200 0.01 ReLU 256
Co.CS 0.4 0.3 0.2 0.75 × 0.4 GAT 1D,2D 2000 0.0005 ReLU 256

daens's People

Contributors

mhadnanali avatar

Watchers

 avatar

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

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

  • Typescript photo Typescript

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

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo 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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

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

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.