Vivid Store is a Python based algorithm repository with integrated evaluation and ranking. This allows the resolution of abstract code requests by currently best fitting algorithms, as required by the cloud-assisted meta programming paradigm.
The resolution mechanism of a Vivid Store is based on the comprehensive evaluation and ranking of all hosted algorithms. This evaluation, which uses an integrated Vivid Node, is performed with respect to all evaluation metrices, that are known to the store and cached in corresponding evaluation lookup tables. Important examples for evaluation metrices are the prediction errors for fixed sets of gold standard samples of different data domain, like latin handwriting samples, spoken word samples, TCGA gene expression data, etc. Accordingly Vivid Stores use an universal data interface, implemented by Vivid DB.
The communication usually is initiated with an
EVALUATION REQUEST to a Vivid Store. This request comprises (E1) an Algorithm Category, (E2) the used Data Domain and (E3) the applied Evaluation Metric. Thereupon the Vivid Store uses it’s evaluation lookup tables to respond to this request with an
CODE OFFER, that includes the above given information, as well as (O1) an Evaluation Score and (O2) an Algorithm ID. If the offer is accepted the Vivid Store receives a
CODE REQUEST, including (C1) the Algorithm ID and (C2) a Cryptographic Token, that identifies the user. Finally, the transaction is finished, when the store responds to the code request with a
CODE ANSWER. This answer depends on the authorization of the user: If the user is unknown or not allowed to receive the algorithm, the answer is constituted by (A1) the Algorithm ID and (A2) a respective Error Notification Flag. If the user, however is authorized the error flag is empty and the answer also comprises (A3) the encoded algorithm.
This approach allows Vivid Stores to dynamically resolve requests for currently best fitting algorithms. Thereupon the separation between evaluation requests and code requests using user authorizations allow collaborations at an entirely new level: Cluster of self contained Vivid Stores.
Installation using PIP
$ pip install vivid-store
Since almost 10 years the rate of data science publications has been growing enormously! For scientists and developers, it is therefore becoming more and more difficult to keep track of suitable current approaches. ... read more
For the current development- and exploration process in data science three obstacles in particular appear as outstanding hurdles, when it comes up to realize projects - and even more, when it comes up to venture collaborations. ... read more