Skip to content

Latest commit

 

History

History
135 lines (91 loc) · 7.03 KB

RELEASES.md

File metadata and controls

135 lines (91 loc) · 7.03 KB
Raw

Releases

The simulation does not follow a fixed release schedule and new versions are typically finalized when a publication is submitted for peer review. This also drives the feature that are added to the simulation since the current research needs dictate the features currently under development.

Change Log

Version 4.1.8

Version 4.1.8 is currently in progress and will update the simuation to run in the compute environment at Temple University, and includes the following updates:

  1. Build script updated, most dependent libraries will be managed by vcpkg. As of 2024-01-30, the versions of dependent libraries are:
    • fmt 10.2.1
    • gsl 2.7.1#3
    • libpq 16.0
    • libpqxx 7.8.1
    • sqlite3 3.43.2#1
    • yaml-cpp 0.8.0#1
    • date 3.0.1#5
    • cli11 2.3.2
    • easyloggingpp 9.97.1
    • catch2 3.5.2
  2. Added SQLiteDistrictReporter and SQLitePixelReporter to support reporting to SQLite database file.
  3. Removal of last-change error handler in main.cpp for Linux platforms (will find a better solution later)
  4. Add clang-format to the project to enforce code style.
  5. Add clang-tidy to the project to enforce code quality.
  6. Add TravelTracking feature to keep track and report the percentage of population travels in the past 30, 60, or 90 days.
  • Two new attributes to the Person class to track the initiation days of the last trip and the last trip outside the district: day_that_last_trip_was_initiated and day_that_last_trip_outside_district_was_initiated.
  • Compile-time flag ENABLE_TRAVEL_TRACKING to toggle the new travel tracking feature, allowing it to be enabled or disabled at compile time for performance considerations.
  • Added TravelTrackingReporter to report (by district) the percentages of the population that have traveled in the past 30, 60, or 90 days to csv file named travel_tracking_{job_id}.csv
  • Noted that, if a person traveled in the past 30 days, he will also be counted as traveled in the past 60 days and 90 days.

  1. SQLiteDistrictReporter and SQLitePixelReporter report clinical episodes by age group (the structure of age group defined in the input file)

  2. Introduce relative_probability_that_child_travels_compared_to_adult, refer to Issue #7 for more details.

  3. Introduces the MFTAgeBasedStrategy, an efficient mechanism for assigning therapies based on individual ages within the simulation. We've optimized therapy assignment by precomputing a mapping from integer ages to therapy indices, thus eliminating the need for repetitive boundary checks. Refer to Issue #8 for more details.

Usage

To turn on the TravelTracking feature, you need to specify the -DENABLE_TRAVEL_TRACKING=ON option to cmake when building the simulation:

# Generate the build folder with cmake and turn on the `TravelTracking` feature
make generate ENABLE_TRAVEL_TRACKING=ON
make build

To use this SQLiteDistrictReporter reporter, you need to specify the -r SQLiteDistrictReporter option when running the simulation:

MaSim -i input.yaml -r SQLiteDistrictReporter -j 1 # this will generate a SQLite database named monthly_data_1.db

Version 4.1.7.1

The version 4.1.7.1 release is the last version of the simulation while the project was based at Penn State, and includes new features needed for research, with significant improvements to the documentation.

  1. Added change_circulation_percent_event
  2. Added rotate_treatment_strategy_event

Version 4.1.7

Version 4.1.7 was an internal version focused on adding new features needed for research:

  1. Added the DistrictMftStrategy treatment strategy to the simulation.

Version 4.1.5

Version 4.1.5 was an internal version focused on adding new features needed for research:

  1. Added the WesolowskiSurface movement model to the simulation.
  2. Added the AgeBandReporter specialist reporter to the simulation.
  3. Updated the default seeding of random numbers to use std::random_device.
  4. Corrected district level reporting of infected individuals when genotype data is not also captured.

Version 4.1.4

Version 4.1.4 was an internal version focused on adding new features needed for research:

  1. Added update_beta_raster_event

Version 4.1.3

Version 4.1.3 release was an internal version focused on adding new features needed for research:

  1. Add the relevant classes and events needed for the RAPT protocol.
  2. Additional input validation added for input data files.
  3. Updated console logging to report physical and virtual memory usage at end of simulation.

Version 4.1.2.1

The version 4.1.2.1 release is focused on bug fixes and adding new features needed for research:

  1. Removed the configuration field p_compliance.
  2. Updated simulation and configuration to support variable treatment compliance rates based upon the day of treatment.

Version 4.1.1.1

Version 4.1.1.1 was an internal version specifically intended for work with manuscript revisions.

  1. Updated the Marshall movement model to precompute the kernel.

Version 4.1.1

The version 4.1.1 release is focused on adding new features needed for research:

  1. Added new seasonality method based upon rainfall data.
  2. Added new event (introduce_mutant_event) which allows mutations to be introduced in the population explicitly.
  3. Added the genotype carriers reporter to the simulation.
  4. Added the seasonal immunity reporter to the simulation.
  5. Updated DbReporter to use ASCII character one (SOH or ☺) to encode \n\r instead of removing them. Stored configurations can now have formatting restored.
  6. Updated -r switch to support multiple comma delimited reporter types.
  7. Refactored movement model for better flexibility, better Burkina Faso model performance.

Version 4.1.0

The version 4.1.0 release focused on quality of life updates as well as correcting bugs found in the 4.0.0 release:

  1. Added the DbReporterDistrict allowing for reporting data to be aggregated to the district level when reporting, database schema also updated to support this.
  2. Weighted frequency has been removed from the genotype table (sim.monthlygenomedata.weightedfrequency), this will break code written for the 4.0.0 schema, but will reduce the storage requirements studies with many replicates and genotypes.
  3. Deprecated values have been removed from the configuration (min_clinical_probablity).
  4. Fixes for various bugs that were found in version 4.0.0.

Version 4.0.0

The 4.0.0 release of MaSim marks a significant upgrade from previous versions, although backwards comparability has been maintained where possible. The major changes with this version of the model are:

  1. Increased spatial support - models can now use a ESRI ASCII Raster for geographic data such as population distribution or a country's political organization.
  2. Reporting of fine-grained information to a PostgreSQL database.
  3. Reporting of agent movement during model execution.