Skip to content

Latest commit

 

History

History
78 lines (58 loc) · 1.48 KB

experiments.md

File metadata and controls

78 lines (58 loc) · 1.48 KB
Raw

Reproduce Experiments

We use the open source library Guild AI to simplify experiments reproducibility.

You can run every models defined in the yaml file as follows:

$ guild run model:train

or

$ python model/train.py

To reproduce our experimental results

Planetoid

$ guild run planetoid:evaluate100

ChebNet

$ guild run chebnet:evaluate100

GCN

$ guild run gcn:evaluate100

GAT

$ guild run gat:evaluate100

Calculate means and standard deviations

Planetoid

$ guild compare -o planetoid --csv -> results/planetoid.csv
$ python3 results.py -p results/planetoid.csv

ChebNet

$ guild compare -o chebnet --csv -> results/chebnet.csv
$ python3 results.py -p results/chebnet.csv

GCN

$ guild compare -o gcn --csv -> results/gcn.csv
$ python3 results.py -p results/gcn.csv

GAT

$ guild compare -o gat --csv -> results/gat.csv
$ python3 results.py -p results/gat.csv

Test the models

In the test scripts you can test the models and create the t-SNE plot of the learned hidden space.

Example

  • train and save a model
    $ python3 GAT/train.py --dataset cora --checkpoint-path GAT_ckpt
  • test the model and make t-SNE plot
    $ python3 GAT/test.py --dataset cora --checkpoint-path GAT_ckpt --tsne

T.B.N.:

  • --checkpoint-path specifies the path wherethe model will be saved/loaded;
  • --tsne specifies to create the t-SNE plot.