Skip to content

Latest commit

 

History

History
24 lines (21 loc) · 888 Bytes

README.md

File metadata and controls

24 lines (21 loc) · 888 Bytes

This is the code for the PFN internship selection test I did last golden week (2019). I implemented the Graph Neural Network for a graph classification task, using numerical differentiation method. Three gradient based optimizer were implemented : Stochastic Gradient Descenti(SGD), Stochastic Gradient Descent with Momentum(SGDM) and Adaptive Moment Estimation (ADAM).

requirements :

Python 3.6.3
Numpy 1.13.3

how to run :

python train.py
    --T step
    --D nfeature
    --op optimizer #SGD/SGDM/ADAM
    --lr learningrate
    --e upsilon
    --iter iteration
    --momen momentum
    --batchsize size
    --ep epoch
    --datadir dir
    --testsize testratio