v0.2.0

EnsembleKalmanProcesses v0.2.0

Diff since v0.1.2

Closed issues:

• Module and module file names are inconsistent (#6)
• Using EKP with deterministic and non-deterministic models (#23)
• Ensemble Kalman Sampler in docs doesn't match the implementation (#50)
• Random seeds (#55)
• Should we be building the Docs with Julia v1.6.4? (#79)
• Standardize UKI constructor to be consistent with EKS, document kwargs (#84)
• nightly causing test fails (#89)
• Simplify interface (#92)
• Covariance matrix obs_noise_cov is restricted to Array{FT, 2} (#99)

Merged pull requests:

• EKS docs update. (#69) (@agarbuno)
• [WIP] Add implementation of sparse EKI (#70) (@jinlong83)
• modified EKI/EKS runtest (#73) (@odunbar)
• Update GitHub Actions workflows to use Julia 1.6.4 (#85) (@navidcy)
• Standardize UKI constructor to be consistent with EKS, document kwargs (#86) (@Zhengyu-Huang)
• CompatHelper: add new compat entry for Convex at version 0.14, (keep existing compat) (#87) (@github-actions[bot])
• CompatHelper: add new compat entry for SCS at version 0.8, (keep existing compat) (#88) (@github-actions[bot])
• remove nightly from test pipeline (#90) (@odunbar)
• Allow quadrature of scalar metrics in UKI, vectorize mean quadrature. (#91) (@ilopezgp)
• Udpdate module names to new convention. (#93) (@ilopezgp)
• Update julia tests to 1.7 (#94) (@ilopezgp)
• Pass explicit RNG object to methods involving stochasticity (#95) (@tsj5)
• Extending functions from packges (#98) (@odunbar)
• Substitute types by abstractions. (#100) (@ilopezgp)