The world's first universal vectorizer for Machine Learning applications.
One of the most challenging tasks when performing natural language processing is to numerically encode the text while ensuring optimality and reproducibility. Usually, this is done by a local dictionary with tokens, which must be shared across your programs. However, with Agano Shibata, you can rapidly encode/decode your tokens based on their occurrence, length, or name with just 2 API calls.
As of now, we have managed to cover most of the English, French, German and Spanish vocabulary and are actively working on including more languages while improving the present ones.