v0.2.3

Polynomials4ML v0.2.3

Diff since v0.2.2

Closed issues:

• mul! in LinearLayer is calling a "slow" function (#62)

Merged pull requests:

• quick fix on rtri and monobasis (#61) (@CheukHinHoJerry)
• Fixing mul! problem in LinearLayer (#63) (@CheukHinHoJerry)

v0.2.2

Polynomials4ML v0.2.2

Diff since v0.2.1

Closed issues:

• Jacobi: replace quadgk with explicit normalisation (#1)
• rrules and frules (#2)
• Product Basis (#16)
• Vararg deprecation (#30)
• Full ACE Integration (#33)
• Second backprop (#46)
• CachedArrays for Linear layer (#56)
• ObjectPools 0.3 (#59)

Merged pull requests:

• WIP: CachedArray in Linear layer and migrating to unwrap from parent (#58) (@CheukHinHoJerry)

v0.2.1

Polynomials4ML v0.2.1

Diff since v0.2.0

v0.2.0

Polynomials4ML v0.2.0

Diff since v0.1.5

Merged pull requests:

• 2nd-order rrules (#49) (@cortner)
• Simpler Static Product (#50) (@cortner)
• WIP: Alternative AA evaluator (#51) (@cortner)
• Dense/Linear Layer (#53) (@CheukHinHoJerry)
• docs update (#55) (@cortner)

v0.1.5

Polynomials4ML v0.1.5

Diff since v0.1.4

Closed issues:

• Precompilation (#47)
• Very Strange Bug (#48)

Merged pull requests:

• trainable AORlayer + some bug fixed (#44) (@CheukHinHoJerry)

v0.1.4

Polynomials4ML v0.1.4

Diff since v0.1.3

Closed issues:

• Chebyshev Basis (#21)
• Allocation tests (#29)
• Laplacian via HyperDuals (#38)

Merged pull requests:

• Bringing ACE in line with P4ML (#37) (@cortner)
• Many frules and rrules (#39) (@cortner)
• Faster Chebyshev Basis (#40) (@cortner)

v0.1.3

Polynomials4ML v0.1.3

Closed issues:

• Ylms (#3)
• Real Spherical Harmonics (#4)
• Batching is sometimes slower (#7)
• Solidify and document Interface (#14)
• Name of package (#15)
• Array Sizes assert bug in-place Ylm evaluation (#17)
• Normalization Conventions and Condon-Shortley Sign Conventions (#25)

Merged pull requests:

• Ylm (#5) (@cortner)
• WIP: Real trig polys (#10) (@cortner)
• Real Spherical Harmonics (#11) (@cortner)
• rnl+rlm basis (#13) (@DexuanZhou)
• Cleanup for Version 0.1.0 (#20) (@cortner)
• Product basis (#22) (@CheukHinHoJerry)
• Shared Functionality (#23) (@cortner)
• Ready for v0.1.0? (#24) (@cortner)
• Much More Cleanup and Simplification (#26) (@cortner)
• Move ACEcore into P4ML (#32) (@cortner)