1. 08 Mar, 2021 1 commit
    • Chris Jewell's avatar
      Adaptive Hamiltonian Monte Carlo within Gibbs implementation · 1676ca5d
      Chris Jewell authored
      CHANGES:
      
      1. Kernel builder functions moved to `mcmc_kernel_factory.py`
      2. Windowed adaptive MCMC a la STAN implemented in `inference.py`
      3. Prior on beta1 tightened to improve stability.
      4. Depends on `gemlib`@develop branch for tf-nightly and tfp-nightly
      1676ca5d
  2. 15 Feb, 2021 2 commits
  3. 02 Feb, 2021 1 commit
  4. 25 Jan, 2021 1 commit
  5. 11 Jan, 2021 2 commits
  6. 10 Jan, 2021 1 commit
    • Chris Jewell's avatar
      Pipeline bugfixes · 9fe819d7
      Chris Jewell authored
      CHANGES:
      
      1. Replaced `read_phe_cases` with `CasesData`
      2. Column-name fix in `case_exceedance`
      3. Fix column heading in `summarize.infec_incidence`
      4. Add Rt_exceed to `summarize.rt`
      5. Add layer option to `summary_geopackage`
      6. Update `ruffus_pipeline.py` to reflect above changes
      7. Update `template_config.yaml` to reflect above changes
      9fe819d7
  7. 08 Jan, 2021 1 commit
  8. 07 Jan, 2021 1 commit
  9. 06 Jan, 2021 1 commit
  10. 04 Jan, 2021 1 commit
  11. 09 Dec, 2020 1 commit
  12. 03 Dec, 2020 1 commit
    • Chris Jewell's avatar
      As of analysis 2020-11-30 · 595ea2eb
      Chris Jewell authored
      CHANGES:
      
      1. Re-introduced Tier lockdown covariates.  National lockdown covariate is set to 0 because
         we cannot identify this from the time-varying baseline force of infection.
      
      2. Added inference on variance parameter for the temporal Gaussian process on beta.
      595ea2eb
  13. 27 Nov, 2020 1 commit
  14. 19 Nov, 2020 1 commit
    • Chris Jewell's avatar
      Implemented weekday effect for I->R · 107186d1
      Chris Jewell authored
      CHANGES:
      
      1. Implemented log-linear model for weekday effect in I->R hazard;
      2. Updated summary.py and within_between.py to match;
      3. Implemented linear random walk for new gamma0 parameter.
      4. Implemented per-parameter data structure in output HDF5 file.
      107186d1
  15. 09 Nov, 2020 2 commits
  16. 08 Nov, 2020 1 commit
  17. 03 Nov, 2020 1 commit
  18. 02 Nov, 2020 1 commit
  19. 01 Nov, 2020 1 commit
  20. 26 Oct, 2020 1 commit
  21. 22 Oct, 2020 5 commits
  22. 05 Oct, 2020 1 commit
    • Chris Jewell's avatar
      Major new pipelining structure · bfcaf904
      Chris Jewell authored
      ==============================
      
      enqueue_pipeline.sh --> covid_pipeline.sge --> prepare_config.py --> inference.py --> summary.py
      
      enqueue_pipeline.sh identifies the dataset and analysis period, launching 4 SGE jobs [P1, P1+2]\times[specimen_date, report_date]
      
      Each SGE job runs the inference and summary scripts, using information contained in a config.yaml file.
      
      The config.yaml file, together with results outputs, are stored in a results directory specified the config file
      and over-ridden in the top-level enqueue_pipeline.sh script.
      bfcaf904
  23. 03 Oct, 2020 4 commits
  24. 01 Oct, 2020 2 commits
  25. 27 Sep, 2020 1 commit
  26. 25 Sep, 2020 2 commits
    • Chris Jewell's avatar
      Code tidy. · d247c771
      Chris Jewell authored
      d247c771
    • Chris Jewell's avatar
      Refactored model specification · e52283b6
      Chris Jewell authored
      Changes:
      
      1. Created a TFP JointDistribution to represent full probability model;
      2. Renamed CovidUKStochastic --> DiscreteTimeStateTransitionModel;
      3. DiscreteTimeStateTransitionModel now inherits from tfp.Distribution.
      e52283b6
  27. 23 Sep, 2020 1 commit
  28. 11 Sep, 2020 1 commit