Scanning all new published packages on PyPI I know that the quality is often quite bad. I try to filter out the worst ones and list here the ones which might be worth a look, being followed or inspire you in some way.

• fuzzyroutines
FuzzyRoutines library contains some routines for work with fuzzy logic operators, fuzzy datasets and fuzzy scales.

• gqlmod
GraphQL Query Modules. ‘gqlmod’ allows you to keep your GraphQL queries in ‘.gql’ files and import them
as modules.
• Validation of queries at import time
• Validation of queries against the schema

• nevolve
Neuro-Evolution Library for Reinforcement Learning.

• sonosco
Framework for training deep automatic speech recognition models. Roboy Sonosco – a library for Speech Recognition based on Deep Learning models

• caispp
High level ML library used in CAIS++ Curriculum. This is package allows for high level ML model creation. It uses Keras with a Tensorflow backend, and was originally created to be used for the curriculum of USC’s CAIS++ (Center for AI in Society, Student Branch).

• coda-to-end
A collection of ML algorithms for Compositional Data Analysis. coda-to-end is a Python package which includes several end to end algorithms for Compositional Data Analysis (CoDA). The user inputs the compositional data and the feature vector (either real numbers for regression, or class indices for classification). For regression problems, the CodaRegress class is used, while classification problems will call the CodaClass class.

• discovery-transition-ds
Advanced data cleaning, data wrangling and feature extraction tools for ML engineers. Accelerated Machine Learning is a unique approach around machine learning that innovates the data science discovery vertical and productization of the data science delivery model. More specifically, it is an incubator project that shadowed a team of Ph.D. data scientists in connection with the development and delivery of machine learning initiatives to define measurable benefit propositions for customer success. To accomplish this, the project developed specific and unique knowledge regarding transition and preparation of data sets for algorithmic execution and augmented knowledge, which is at the core of the projects services offerings. From this the project developed a new approach to data science discovery and productization dubbed ‘Accelerated Machine Learning’.

• multiprefixspan
Implementation of Prefixspan for multiple items

• autodisambiguator
A framework for automatic disambiguation based on secondary categories. Autocategorization of lemma into lemma with index for lemma whose index is defined by POS or similar

• classmail
A simple framework for automatise mail classification task. Mail classification Python library optimized for french mails in the field of insurance. Classmail was created to automate mail classification workflow in quick experiments. Developped during my internship at [Covéa](https://www.covea.eu ).