https://github.com/dahong67/gcpdecompositions.jl - v0.3.1

GCPDecompositions v0.3.1

Diff since v0.3.0

Merged pull requests: - CPD Normalize Function (#58) (@bakergg) - Add Array function (#61) (@PedroSophiaaa) - Make normalizecomps more flexible (#65) (@dahong67) - Add permutecomps/permutecomps! (#66) (@dahong67) - Update the doc setup (#90) (@dahong67) - Add a logo (#91) (@dahong67)

Closed issues: - Conversion to Array (#57) - Normalize components (#64)

- Julia
Published by github-actions[bot] 12 months ago

https://github.com/dahong67/gcpdecompositions.jl - v0.3.0

GCPDecompositions v0.3.0

Diff since v0.2.0

Merged pull requests: - New Demo - Monkey BMI Data (#55) (@bakergg) - Rename ncomponents to ncomps (#63) (@dahong67)

- Julia
Published by github-actions[bot] almost 2 years ago

https://github.com/dahong67/gcpdecompositions.jl - v0.2.0

GCPDecompositions v0.2.0

Diff since v0.1.2

Merged pull requests: - Change API to use AbstractLoss types (#13) (@dahong67) - Add more losses (#20) (@alexmul1114) - Change design for dispatch (#22) (@dahong67) - Setup system for constraints (#25) (@dahong67) - Support multiple algorithms (#26) (@dahong67) - Add Benchmarks (#30) (@alexmul1114) - Eagerly instantiate the matricized tensor in mttkrp (#31) (@dahong67) - Extra work on benchmarks (#32) (@dahong67) - Faster MTTKRP (#35) (@alexmul1114) - FastALS method (#39) (@alexmul1114) - Faster Khatri-Rao (#41) (@dahong67) - In-place versions of mttkrp and khatrirao (#43) (@dahong67) - Add norm method for CPD (#44) (@dahong67) - Remove old _gcp signature (#46) (@dahong67) - Improve module/file organization (#47) (@dahong67) - Make init one of the arguments (#49) (@dahong67) - Make loss a kwarg in gcp (#50) (@dahong67) - Initial setup of docs (#51) (@dahong67) - Improved CPD Indexing (#52) (@alexmul1114)

Closed issues: - Change API to use AbstractLoss types and handle custom case with new type (#12) - Add the remaining losses (#15) - Setup support for multiple algorithms and add ALS (#21) - Remove old method signature (#23) - Setup more systematic handling of constraints (#24) - Improve module/file organization (#27) - Improve ALS implementation (#28) - Add benchmarks (#29) - Faster MTTKRP (#33) - Faster Khatri-Rao product (#34) - Add in-place mttkrp! and khatrirao! (#42) - Make init one of the arguments (#48)

- Julia
Published by github-actions[bot] about 2 years ago

https://github.com/dahong67/gcpdecompositions.jl - v0.1.2

GCPDecompositions v0.1.2

Diff since v0.1.1

Merged pull requests: - Restrict LossFunctions.jl support to DistanceLoss and MarginLoss (#14) (@dahong67)

- Julia
Published by github-actions[bot] over 2 years ago

https://github.com/dahong67/gcpdecompositions.jl - v0.1.1

GCPDecompositions v0.1.1

Diff since v0.1.0

Merged pull requests: - Use autodiff to compute derivative of loss by default (#9) (@alexmul1114) - Add compatibility with LossFunctions.jl, test with least squares (#11) (@alexmul1114)

Closed issues: - Register v0.1.0 (#1) - Subtype AbstractArray? (#3) - Use autodiff to compute derivative of loss by default (#5) - Allow user to specify loss function using LossFunctions.jl (#10)

- Julia
Published by github-actions[bot] over 2 years ago

https://github.com/dahong67/gcpdecompositions.jl - v0.1.0

GCPDecompositions v0.1.0

Initial release!

- Julia
Published by github-actions[bot] over 3 years ago