Skip to content

Changelog

v1.0.0

  • Implemented modules:
    • Reimplementation of vaccinations as a module;
    • Conversion of diagnostics as a module;
    • Vaccination trial module;
    • Reinfection module;
    • Transport module;
    • Schools monitoring module;
    • Visualizer module;
  • New Airborne network/infection mechanic for use in closed spaces;
  • Better organization of the lab tools for multiple simulations and analysis;
  • New population methods for creation of attributes with output management;
  • Implementation of new auxiliary classes:
    • RandomDistibution: used with parameters that can be passed as a distribution;
    • coefficient_container: used to apply coefficients to child classes (used in the disease class to modulate infection attributes);
  • New dynamic execution list on the community with add_to_execution() method, priority control and time of the day control;
  • Elimination of legacy distance mechanics;
  • Implementation of internal random network generator to improve performance;
  • Reorganization of network generation and states transition inside a single method to improve performance;
  • Reorganization of network generators as classes and creation of add_net_generator to dynamically set generators for different placements.
  • Implemented stop_filter to dynamically stop simulations;

v0.3.0

  • Implemented new flexible services, workers and rooms interface;
  • Implemented new configurable quarantines;
  • Implemented new configurable vaccination campaigns;
  • Improved parsing of parameters;
  • Added a mechanism to configure random distributions to determine infection probability and times between states transitions;
  • Added new plots: Symptoms or Infections in homes, Histogram of time between transitions, Infection tree;
  • Updated visualizer to also work with networks.
  • Implemented new clock with a start date, days of the week and functions to convert time;
  • Added the ability to select with days of the week a service opens;
  • New homes and environmental networks generators.
  • New networks available for services:
    • NET_HOSPITALS: used in hospitals (internal implementation);
    • NET_STREET_MARKETS: used in street markets (internal implementation);
  • Separated network generators on it's own file;
  • Separated vaccination, quarantine and decisions mechanics on it's own files;
  • Implemented new hdf5 interface (using new h5dict library);
  • Added the ability to decrease workers or visitors mid-simulation on services;
  • Update infections defaults according to new calibration.

v0.2.0

  • Implemented new mechanics to generate encounters in each location using local networks instead of distances;
  • Networks available in this release (using NetworkX):
    • NET_DEFAULT: used in homes, hospitals, non essential services and street markets (Barabasi Albert generator from NetworkX);
    • NET_MARKETS: used in markets (internal implementation);
    • NET_SCHOOLS: used in schools (internal implementation);
    • NET_RESTAURANTS_1 and NET_RESTAURANTS_2: used in restaurants (internal implemetation);
  • Profiler branched into it's own repository, named nanoProfiler;
  • Added methods to visualize encounters networks and infection tree (using matplotlib and pygraphviz);
  • Creation of defaults.py with default values for all parameters and their descriptions;
  • New decision criteria using number of symptomatic cases, or by following an inputted time series with all the decisions set at specific intervals;
  • Organized internal simulation data to create a mode with low memory usage;
  • Added viral shedding and anti-body curves generation to disease object;
  • Implemented vaccination mechanics;
  • Implemented neighborhood distinction for particles and services;
  • Implemented an infection tree to store information respective to the propagation of the disease;
  • Implemented susceptibility and mortality defined by age group;
  • Added deceased state;
  • Implemented a progress bar and text output of main results;
  • Fixed errors in quarantine by tracing;
  • Added setup.py script for installation of COMORBUSS as a module.

v0.1.2

  • Classes delegated to their own files;
  • Creation of disease class for biological models;
  • Creation of standard plot class for loading and visualizing results;
  • Creating of profiler class;
  • Better integration of laboratory and experiment classes, now based on plots class;
  • Integration with database for Brazilian demography;
  • Implementation of social isolation as a time series;
  • Positions inside services on a subgrid;
  • Services and homes organized along third dimension (apartments, offices etc);
  • Compute_distances_isolate_homes redundant and eliminated;
  • Non-essential services implemented - clarification between contact and encounter;
  • Better handling of the changes for placement, activity and movement (all in mark_for_new_position), which feeds a single instance for updating positions (update_positions).

v0.1.1

  • Complete CLI - new interface, where the internal mechanisms are easily turned on or off, and user are exposed only to basic configuration parameter;
  • Introduced (experimental) lab module for automation of simulations and analysis;
  • Introduced decoupled decision and action mechanics;
  • Decoupled diagnostic mechanics (only PCR for now) from actions;
  • Decoupled hospitalization of serious cases from quarantine actions;
  • Abstracted quarantines to easily configure different types in separate comorbuss;
  • Corrected and optimized boundary conditions;
  • Decoupled configuration of movement from action in services;
  • Improved visualization;
  • Introduced simple quarantines by tracing;
  • Unified loader/saver;
  • Triage and demographic normalization inside community class;
  • Introduced generation of network contact matrix from agent tracing (experimental).