v0.4.2

MCIntegration v0.4.2

Diff since v0.4.1

Merged pull requests:

• Fix inconsistency with 1 dof (#55) (@lxvm)

Closed issues:

• Fail to deal with integrands with singularities: a simple case (#52)

v0.4.1

MCIntegration v0.4.1

Diff since v0.4.0

Closed issues:

• refactor user API (#38)
• Bug with vegasmc and vegas with user-defined datatype (#44)

Merged pull requests:

• Fix typos (#47) (@dcerkoney)
• Bugfix: vegas fails whe dof is different between the integrands (#49) (@kunyuan)

v0.4.0

MCIntegration v0.4.0

Diff since v0.3.6

Merged pull requests:

• improve docs (#43) (@kunyuan)
• Bugfix (#45) (@kunyuan)
• Bugfix (#46) (@kunyuan)

v0.3.6

MCIntegration v0.3.6

Diff since v0.3.5

Merged pull requests:

• fix a critical bug with observables with different types for differen… (#42) (@kunyuan)

v0.3.5

MCIntegration v0.3.5

Diff since v0.3.4

Merged pull requests:

• fix bug of double-call MPI.Init() (#40) (@houpc)
• fix MPI reweight bug and add its unit test (#41) (@houpc)

v0.3.4

MCIntegration v0.3.4

Diff since v0.3.3

Closed issues:

• Incompatibility with built-in functions when var is provided as a tuple. (#35)

Merged pull requests:

• Bugfix (#37) (@kunyuan)
• Bugfix (#39) (@kunyuan)

v0.3.3

MCIntegration v0.3.3

Diff since v0.3.2

Closed issues:

• Sample from learned variables directly (#34)

Merged pull requests:

• clean up (#31) (@kunyuan)
• Bugfix for report (#32) (@dcerkoney)
• fix a bug with error report (#33) (@kunyuan)
• Refactor variable type system and improve readme (#36) (@kunyuan)

v0.3.2

MCIntegration v0.3.2

Diff since v0.3.1

Closed issues:

• improve MC sampling algorithm (#6)
• cached variable (#16)
• Thread support (#22)

Merged pull requests:

• Add thread support (#28) (@kunyuan)

v0.3.1

MCIntegration v0.3.1

Diff since v0.3.0

Merged pull requests:

• Bugfix (#24) (@kunyuan)
• fix tests (#25) (@kunyuan)
• Bugfix (#26) (@dcerkoney)
• add support for inplace integrand (#27) (@kunyuan)

v0.3.0

1. Support Vegas algorithm.
2. Implement a new type of Vegas algorithm based on the Markov-chain Monte Carlo. It is as fast as the Vegas algorithm at low dimensions and becomes faster and more robust at high dimensions.
3. Improve documentation.