This is a limited proof of concept to search for research data, not a production system.

Search the MIT Libraries

Title: MLP link prediction models in H5 format

Type Dataset Reese, Justin (2020): MLP link prediction models in H5 format. Zenodo. Dataset. https://zenodo.org/record/4031401

Author: Reese, Justin ;

Links

Summary

The KG-COVID-19 graph from this Zenodo URL was used:

https://zenodo.org/record/4011267/files/kg-covid-19-skipgram-aug-2020.tar.gz

To produce these embeddings (Skipgram, 80/20 training/test split, seed=42, 500 epochs max, delta 0.0001)

https://zenodo.org/record/4019808/files/SkipGram_80_20_training_test_epoch_500_delta_0.0001_embedding.npy

These embeddings were used to train link prediction classifiers using this Jupyter notebook:

https://github.com/justaddcoffee/kg_covid_19_drug_analyses/blob/master/Link%20prediction.ipynb

SkipGram_weightedL2_finalized_model.h5

SkipGram_weightedL2.csv

SkipGram_weightedL1_finalized_model.h5

SkipGram_weightedL1.csv

SkipGram_hadamard_finalized_model.h5

SkipGram_hadamard.csv

SkipGram_average_finalized_model.h5

SkipGram_average.csv all_reports.csv

More information

  • DOI: 10.5281/zenodo.4031401

Dates

  • Publication date: 2020
  • Issued: September 15, 2020

Rights


Much of the data past this point we don't have good examples of yet. Please share in #rdi slack if you have good examples for anything that appears below. Thanks!

Format

electronic resource

Relateditems

DescriptionItem typeRelationshipUri
IsVersionOfhttps://doi.org/10.5281/zenodo.4031400
IsPartOfhttps://zenodo.org/communities/zenodo