Implement query flow. top_k proteins saved to results/results.json - Githubissues

georgeamccarthy / protein_search

The neural search engine for proteins.

GNU Affero General Public License v3.0

15 stars 6 forks source link

Implement query flow. top_k proteins saved to results/results.json #19

Closed georgeamccarthy closed 3 years ago

georgeamccarthy commented 3 years ago

Pull Request Type

Select the type of changes and add some context. After selecting, remove the rest of the types.

🏆 Enhancements - Random protein from samelength.csv is printed in command line output, then used as an input to the search. The top_k (currently = 5) most similar proteins are printed.
🏠 Internal - path variables moved to backend_config, top_k moved to backend config. New embeddings directory separated proteins.json from csv input files. Results saved to new dir results in results/results.json.

Purpose

Implements searching.

Why?

Essential feature

Feedback required over

A quick pair of :eyes: on the code The current implementation is a base implementation from which we will build on. Next step will be to link the input and output to the frontend.

Mentions

@fissoreg

georgeamccarthy commented 3 years ago

Need to change tests to open flows instead of classing MyIndexer methods directly.