Skip to content

Latest commit

 

History

History

knowledgebases

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
pmlb_regression_results-small.tsv
pmlb_regression_results-small.tsv
 
 
processing
processing
 
 
test
test
 
 
user
user
 
 
README.md
README.md
 
 
__init__.py
__init__.py
 
 
add_ids_to_kb.py
add_ids_to_kb.py
 
 
pmlb_classification_metafeatures.csv.gz
pmlb_classification_metafeatures.csv.gz
 
 
pmlb_regression_metafeatures.csv.gz
pmlb_regression_metafeatures.csv.gz
 
 
pmlb_regression_results-small.tsv.gz
pmlb_regression_results-small.tsv.gz
 
 
pmlb_regression_results.json.gz
pmlb_regression_results.json.gz
 
 
pmlb_regression_results.pkl.gz
pmlb_regression_results.pkl.gz
 
 
pmlb_regression_results.tsv.gz
pmlb_regression_results.tsv.gz
 
 
sklearn-benchmark-data-knowledgebase-r6-small.tsv.gz
sklearn-benchmark-data-knowledgebase-r6-small.tsv.gz
 
 
sklearn-benchmark-data-knowledgebase-r6.tsv.gz
sklearn-benchmark-data-knowledgebase-r6.tsv.gz
 
 
sklearn-benchmark5-data-edited-formatted-filtered.tsv.gz
sklearn-benchmark5-data-edited-formatted-filtered.tsv.gz
 
 
sklearn-benchmark5-data-knowledgebase-small.json.gz
sklearn-benchmark5-data-knowledgebase-small.json.gz
 
 
sklearn-benchmark5-data-knowledgebase-small.pkl.gz
sklearn-benchmark5-data-knowledgebase-small.pkl.gz
 
 
sklearn-benchmark5-data-knowledgebase-small.tsv.gz
sklearn-benchmark5-data-knowledgebase-small.tsv.gz
 
 
sklearn-benchmark5-data-knowledgebase.json.gz
sklearn-benchmark5-data-knowledgebase.json.gz
 
 
sklearn-benchmark5-data-knowledgebase.pkl.gz
sklearn-benchmark5-data-knowledgebase.pkl.gz
 
 
sklearn-benchmark5-data-knowledgebase.tsv.gz
sklearn-benchmark5-data-knowledgebase.tsv.gz
 
 
sklearn-benchmark5-data-short-formatted.tsv.gz
sklearn-benchmark5-data-short-formatted.tsv.gz
 
 

README.md

Knowledgebases

Knowledgebases are collections of previous results from machine learning analyses that are used to bootstrap Aliro.

The results are stored in a .tsv.gz file. By default Aliro loads results from the benchmark of scikit-learn described in these papers:

  • Olson, Randal S., William La Cava, Patryk Orzechowski, Ryan J. Urbanowicz, and Jason H. Moore. “PMLB: A Large Benchmark Suite for Machine Learning Evaluation and Comparison.” BioData Mining, 2017. arxiv

  • Olson*, Randal S., William La Cava*, Zairah Mustahsan, Akshay Varik, and Jason H. Moore. “Data-Driven Advice for Applying Machine Learning to Bioinformatics Problems.” In Pacific Symposium on Biocomputing (PSB), 2017. arxiv

A knowledgebase also needs the metafeatures of the datasets it contains. The processing folder contains scripts to facilitate generating them, given a set of datasets and the aformented set of results. See generate_metafeatures.py for an example.