Release Notes

0.1.0 (2023-06-22)

  • First release on PyPI.

Features

Credentials Management

  • create_profile() initializes a Credentials object from a username and password.

  • create_token_profile() initializes a Token object from a REST API token.

REST API data pulls

  • get_api_response() retrieves a response from an API service using a provided profile of type Credentials or Token. Used as a wrapper to the .get_api_response() method of a provided profile of type Credentials or Token.

  • get_api_data() retrieves data (in JSON or CSV) from an API service using a provided profile of type Credentials or Token. Used as a wrapper to the .get_api_data() method of a provided profile of type Credentials or Token.

  • get_api_graph() retrieves a graph from an API service using a provided profile of type Credentials or Token. Used as a wrapper to the .get_api_graph() method of a provided profile of type Credentials or Token.

  • get_essence_data() retrieves data from the NSSP-ESSENCE API service using a provided profile of type Credentials or Token.

Anomaly Detection and Trend Classification

  • alert_ewma() implements the EWMA time series anomaly detection algorithm.

  • alert_regression() implements the Multiple Adaptive Regression time series anomaly detection algorithm.

  • alert_switch() implements the Regression/EWMA Switch time series anomaly detection algorithm

  • alert_nbinom() implements the Negative Binomial Regression time series anomaly detection algorithm.

  • alert_serfling() implements the original and modified Serfling method for time series anomaly detection.

  • classify_trend() fits rolling binomial models to a daily time series of percentages or proportions to classify the overall trend.

Added Data

  • load_simulated_ts() loads a dataframe of simulated time series.

  • load_nssp_stopwords() loads a dataframe of NSSP-curated stopwords.

  • get_scenario1() loads a a subset of the simulated time series data (“scenario #1”).

  • get_scenario2() loads a a subset of the simulated time series data (“scenario #2”).

Utility Functions

  • change_dates() modifies the start and end dates in a given URL to new dates, if provided.

  • webscrape_icd() scrapes ICD Codes from the Web.

Classes

  • Credentials is an abstract representation of a profile object given a username and password strings.

  • Token is an abstract representation of a profile object given a token string.

  • APIGraph is an abstract representation of a graph object returns from an API service.

  • NSSPContainer encapsulates a value or an object to store.