Giters

AnacletoLAB / grape

🍇 GRAPE is a Rust/Python Graph Representation Learning library for Predictions and Evaluations

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Edge prediction models not working in edge prediction evaluation pipeline

DylanSteinecke opened this issue · comments

commented

When using the edge prediction evaluation pipeline, I tried running a few different edge prediction models (e.g., KipfGCNEdgePrediction, not for the MLP, Decision Tree, or Random Forest), I got this error:

Traceback (most recent call last):
File [script I ran]
<module> KipfGCNEdgePrediction(),
File "/home/dylansteinecke/anaconda3/lib/python3.9/site-packages/embiggen/utils/abstract_models/model_stub.py", line 101, in init
super().init(**parent_class.smoke_test_parameters())
File "/home/dylansteinecke/anaconda3/lib/python3.9/site-packages/embiggen/utils/abstract_models/abstract_model.py", line 124, in smoke_test_parameters
raise NotImplementedError((
NotImplementedError: The smoke_test_parameters method must be implemented in the child classes of abstract model.

Is there a way I can address it or are these models not currently permitted in the edge prediction evaluation pipeline? Thank you.

commented

I think that is an error caused by a combination of errors, do you have TensorFlow and sklearn installed? I am on discord if you want to do a quick call.