1. 01 Dec, 2020 1 commit
    • Bentriou Mahmoud's avatar
      Test of distance G automaton passes! · f4c41acb
      Bentriou Mahmoud authored
      The two Cosmos tests checks the statistical correctness of estimated
      distance value for R1,R2,R3,R5 experiments. Also, they test that all
      trajectories are accepted.
      A bunch of errors were fixed in automata files (Cosmos .lha and julia
      ./automata files) and simulation function that introduced statistical
      bias, i.e. approximated values that were closed but not in the
      confidence interval. The general structure of simulation and was
      reworked in order to suits well te behavior of Cosmos.
  2. 30 Nov, 2020 1 commit
    • Bentriou Mahmoud's avatar
      A major fix in simulation was done, side effect issue, an efficient · c095bee4
      Bentriou Mahmoud authored
      copyto! is implemented for LHA states.
      Automaton G works almost well, a bias in the distance is identified
      compared to Cosmos when one of the model parameters is zero.
      Implementation of volatile_simulate, which simulates SynchronizedModel
      without saving values: it only returns the last LHA state of the
      Implementation of distributed computations of variable automata with
      Tests passed.
  3. 29 Nov, 2020 2 commits
  4. 27 Nov, 2020 2 commits
  5. 26 Nov, 2020 2 commits
  6. 25 Nov, 2020 1 commit
  7. 24 Nov, 2020 3 commits
  8. 22 Nov, 2020 1 commit
  9. 21 Nov, 2020 2 commits
  10. 20 Nov, 2020 2 commits
  11. 19 Nov, 2020 1 commit
  12. 18 Nov, 2020 2 commits
  13. 17 Nov, 2020 3 commits
  14. 16 Nov, 2020 3 commits
  15. 15 Nov, 2020 4 commits
    • Bentriou Mahmoud's avatar
      Improvement of memory allocation. · d690d6b9
      Bentriou Mahmoud authored
      In fine I give the Abstract and Static arrays types up.
    • Bentriou Mahmoud's avatar
      Fix of type instability in SIR model: · 384b4ad7
      Bentriou Mahmoud authored
      - Real => Float64 for better performance (Real is an abstract type)
      - Should not create a type inside the model because the compiler fails
      at estimating the type of the output of the f function (detected with
      - Intentionaly ER is not as well improved as SIR in order to check the
      differences of performance.
    • Bentriou Mahmoud's avatar
      - Add of ER model + tests · 06dca928
      Bentriou Mahmoud authored
      - Add of benchmark scripts that compares perf wrt pygmalion.
      => As expected it is bad in terms of reading cost (because of row by row
      matrix read) but it is also not good in terms of simulation cost.
      hcat seems to perform badly, should investigate.
    • Bentriou Mahmoud's avatar
      Commit that implement the essentials for the lauch of a simulation of a · 2f7fea8f
      Bentriou Mahmoud authored
      SIR model.
      - Evolution of the ContinuousTimeModel type
      - simulate, easy access to Trajectory fields
      - fix of several errors
  16. 14 Nov, 2020 2 commits