Skip to content

Latest commit

 

History

History
490 lines (316 loc) · 35.9 KB

CHANGELOG.md

File metadata and controls

490 lines (316 loc) · 35.9 KB
Raw

1.24.0 (2022-05-22)

Features

  • sgd classifier can not train on categorical variables, as well as one-hot encoded variables (10141cd)

1.23.0 (2022-05-19)

Features

  • added test case for custom callbacks. works great and somehow serializes. (7fa5c42)
  • custom modelfitargs for linear models (2ddcad9)

1.22.0 (2022-05-18)

Features

  • added back in logistic regression tests (dc2ec4a)
  • first pass at removing tensorflow from bundle (7562da2)
  • more tests moved over (76509d8)
  • removed hard dependency on tensorflow (0f2736e)
  • removed unneeded build steps (d3814ca)
  • updated serialize / deserialize to avoid tfjs error (1bf508d)

1.21.0 (2022-05-08)

Features

1.20.0 (2022-04-26)

Features

1.19.0 (2022-04-26)

Features

  • changed lodash imports to support building on esm.sh (3eabad9)

1.18.0 (2022-04-26)

Features

  • removed seedrandom in favor of inlining to help build on esm.sh (245d49c)

1.17.0 (2022-04-21)

Features

  • added automated tests to test our code in the browser (87e06a2)
  • renamed files, added to repo (9a63da4)

1.16.0 (2022-04-19)

Features

  • fixed loadBoston calls. Need to do the others (05c9d9a)
  • fixed tests (3f6654d)
  • remove data loading logic in favor of using dfd.readCSV(url) (3251738)

1.15.0 (2022-04-18)

Features

  • remove rollup from the build process, replace with esbuild (1f16ef8)
  • updated readme (7e70aba)

1.14.0 (2022-04-17)

Features

  • commented out tests (77b6ab6)
  • commenting out svc, svr code until it can be built in browser (dd95256)
  • disable libsvm until we can ship to the browser (fdc3214)
  • updated tests (6938b32)

1.13.0 (2022-04-17)

Features

1.12.0 (2022-04-17)

Features

  • only import from tensorflow and not subpackages (f971942)

1.11.0 (2022-04-17)

Features

1.10.0 (2022-04-17)

Features

  • removed danfo as a dependency (b8b5578)

1.9.0 (2022-02-27)

Bug Fixes

Features

  • add custom serializer to sgdclassifier (wip) (9c3f3dc)
  • add loss and optimizer type to enable easy parsing (08809ec)
  • add loss types and initializer types (33e1d2c)
  • add more estimators and makes serializer flexible for ensembles and pipeline (e2d319b)
  • add optimizer, loss and intializer caller (8778f73)
  • add serializer to criterion (8cbb737)
  • add serializer to criterion (92f765e)
  • add serializer to decision tree and update test (83ef949)
  • add serializer to kNeighborBase (5314044)
  • add serializer to labelencoder (6a1c362)
  • add serializer to linear model base class (9701a3e)
  • add serializer to NaiveBayes (84e747e)
  • add serializer to pipeline (6c425e1)
  • add serializer to splitter (763d7e1)
  • add serializer to SVC AND SVR (672328f)
  • add serializer to votingclassifier (2a21f31)
  • add serializer to votingRegressor (6d59faf)
  • allow ClassifierMixin to extends Serialize class (48d15f1)
  • allow Kmeans to inherit from serializer (6b36f4c)
  • implement generic class to Serialize models and transformers e.t.c (7f617bc)
  • implement serialize ensembles for ensemble class (bbd9fac)
  • make TransformerMixin and RegressorMixin extends serialize (5a001fd)
  • update linear model with new args to enable easy serialization (ec559e1)
  • update serialize to easily parse tensors (232cb62)
  • update Serialize to handle serialization of tensors (924b050)
  • update serialize to return inherited class (0733023)
  • update serializer for sgdclassifier (4971b3c)

1.8.0 (2022-01-28)

Bug Fixes

  • k-neighbors-regressor now supports no params (9656d6b)
  • kd-tree index issue fix + docs (5eba76c)
  • kd-tree protection copy + tfjs-core import (5c4348d)

Features

  • k-neighbors now lists available algorithms (fcfcb87)

1.7.0 (2022-01-23)

Bug Fixes

  • cross-val-score and k-fold fixes+improvements (21a566b)
  • cross-val-score api improvement etc (efe63f9)
  • k-fold memory leak (2f5529d)

Features

  • cross-val-score and k-fold implemented (6bc3ee3)
  • rand-utils create-rng (553232f)

1.6.0 (2022-01-21)

Features

1.5.0 (2022-01-20)

Features

  • add makeRegression function (5337ecf)

1.4.0 (2022-01-18)

Features

  • added ability for decision tree to handle negative input (a6cf53f)
  • first pass at decision tree classifier (550551e)
  • first pass at regression tree (849469a)

1.3.0 (2022-01-14)

Features

  • gaussian naive bayes classifier (8174ae1)
  • gaussian naive bayes classifier (d520b1a)

1.2.0 (2022-01-02)

Features

  • seeing if this package.json exports does the trick (4a73f7c)

1.1.0 (2021-12-31)

Bug Fixes

  • added fast-check dev dependency (fe9e693)
  • change max length on commit message (fe4ce57)
  • change max length on commit message (f4a8672)
  • commented out tests failing in test:browser (0fe0fe1)
  • k-neighbors-classifier await super.fit() (01632f4)

Features

  • k-neighbors kd-tree algorithm (59d40de)
  • kd-tree first draft (354979a)

Performance Improvements

  • k-neighbors kd-tree performance improvements (158506c)

1.1.0 (2021-12-31)

Bug Fixes

  • added fast-check dev dependency (fe9e693)
  • change max length on commit message (f4a8672)
  • commented out tests failing in test:browser (0fe0fe1)
  • k-neighbors-classifier await super.fit() (01632f4)

Features

  • k-neighbors kd-tree algorithm (59d40de)
  • kd-tree first draft (354979a)

Performance Improvements

  • k-neighbors kd-tree performance improvements (158506c)

1.0.3 (2021-12-31)

Bug Fixes

  • fixing any type to correct usage (0084771)

1.0.2 (2021-12-31)

Bug Fixes

  • fixing any type to correct usage (3f5c288)

1.0.1 (2021-12-31)

Bug Fixes

  • fixing any type to correct usage (4496805)

1.0.0 (2021-12-31)

Bug Fixes

  • broken UMD browser script (10f2e34)
  • fix lint (28c876d)
  • k-neighbors inverse distance weighting (a162baa)
  • k-neighbors predict now checks n_features (35efd93)

Features

  • add SVC (a5fe596)
  • added cut 1 of voting classifier (4045b81)
  • added tests and basic implementation of votingregressor (d7011e7)
  • added voting classifier (d7ab9c6)
  • broke out sgdlinear into sgdregressor and sgdclassifier (81fbee8)
  • changed imports (390375c)
  • finish (825ebb7)
  • First pass at VotingRegressor (ffb3393)
  • implemented kNeighborsRegression (a1a7174)
  • import libsvm (f0f0cc8)
  • k-neighbors regressor (94f6a69)
  • k-neighbors regressor (225e167)
  • k-neighbors-classifier implemented (d120257)
  • k-neighbors-regressor (050cec6), closes #111
  • k-neighbors-regressor (cb0a8b0)
  • linear svr (7a0534d)
  • simple first pass addition of linear-svc (483117d)
  • train test split implementation (97b89a5)
  • updated index to export linear-svc and updated docs (b5c116b)

1.0.0 (2021-12-31)

Bug Fixes

  • broken UMD browser script (10f2e34)
  • fix lint (28c876d)
  • k-neighbors inverse distance weighting (a162baa)
  • k-neighbors predict now checks n_features (35efd93)

Features

  • add SVC (a5fe596)
  • added cut 1 of voting classifier (4045b81)
  • added tests and basic implementation of votingregressor (d7011e7)
  • added voting classifier (d7ab9c6)
  • broke out sgdlinear into sgdregressor and sgdclassifier (81fbee8)
  • changed imports (390375c)
  • finish (825ebb7)
  • First pass at VotingRegressor (ffb3393)
  • implemented kNeighborsRegression (a1a7174)
  • import libsvm (f0f0cc8)
  • k-neighbors regressor (94f6a69)
  • k-neighbors regressor (225e167)
  • k-neighbors-classifier implemented (d120257)
  • k-neighbors-regressor (050cec6), closes #111
  • k-neighbors-regressor (cb0a8b0)
  • linear svr (7a0534d)
  • simple first pass addition of linear-svc (483117d)
  • train test split implementation (97b89a5)
  • updated index to export linear-svc and updated docs (b5c116b)

1.0.0 (2021-12-31)

Bug Fixes

  • broken UMD browser script (10f2e34)
  • fix lint (28c876d)
  • k-neighbors inverse distance weighting (a162baa)
  • k-neighbors predict now checks n_features (35efd93)

Features

  • add SVC (a5fe596)
  • added cut 1 of voting classifier (4045b81)
  • added tests and basic implementation of votingregressor (d7011e7)
  • added voting classifier (d7ab9c6)
  • broke out sgdlinear into sgdregressor and sgdclassifier (81fbee8)
  • changed imports (390375c)
  • finish (825ebb7)
  • First pass at VotingRegressor (ffb3393)
  • implemented kNeighborsRegression (a1a7174)
  • import libsvm (f0f0cc8)
  • k-neighbors regressor (94f6a69)
  • k-neighbors regressor (225e167)
  • k-neighbors-classifier implemented (d120257)
  • k-neighbors-regressor (050cec6), closes #111
  • k-neighbors-regressor (cb0a8b0)
  • linear svr (7a0534d)
  • simple first pass addition of linear-svc (483117d)
  • train test split implementation (97b89a5)
  • updated index to export linear-svc and updated docs (b5c116b)

1.0.0 (2021-12-31)

Bug Fixes

  • broken UMD browser script (10f2e34)
  • fix lint (28c876d)
  • k-neighbors inverse distance weighting (a162baa)
  • k-neighbors predict now checks n_features (35efd93)

Features

  • add SVC (a5fe596)
  • added cut 1 of voting classifier (4045b81)
  • added tests and basic implementation of votingregressor (d7011e7)
  • added voting classifier (d7ab9c6)
  • broke out sgdlinear into sgdregressor and sgdclassifier (81fbee8)
  • changed imports (390375c)
  • finish (825ebb7)
  • First pass at VotingRegressor (ffb3393)
  • implemented kNeighborsRegression (a1a7174)
  • import libsvm (f0f0cc8)
  • k-neighbors regressor (94f6a69)
  • k-neighbors regressor (225e167)
  • k-neighbors-classifier implemented (d120257)
  • k-neighbors-regressor (050cec6), closes #111
  • k-neighbors-regressor (cb0a8b0)
  • linear svr (7a0534d)
  • simple first pass addition of linear-svc (483117d)
  • train test split implementation (97b89a5)
  • updated index to export linear-svc and updated docs (b5c116b)

1.0.0 (2021-12-31)

Bug Fixes

  • broken UMD browser script (10f2e34)
  • fix lint (28c876d)
  • k-neighbors inverse distance weighting (a162baa)
  • k-neighbors predict now checks n_features (35efd93)

Features

  • add SVC (a5fe596)
  • added cut 1 of voting classifier (4045b81)
  • added tests and basic implementation of votingregressor (d7011e7)
  • added voting classifier (d7ab9c6)
  • broke out sgdlinear into sgdregressor and sgdclassifier (81fbee8)
  • changed imports (390375c)
  • finish (825ebb7)
  • First pass at VotingRegressor (ffb3393)
  • implemented kNeighborsRegression (a1a7174)
  • import libsvm (f0f0cc8)
  • k-neighbors regressor (94f6a69)
  • k-neighbors regressor (225e167)
  • k-neighbors-classifier implemented (d120257)
  • k-neighbors-regressor (050cec6), closes #111
  • k-neighbors-regressor (cb0a8b0)
  • linear svr (7a0534d)
  • simple first pass addition of linear-svc (483117d)
  • train test split implementation (97b89a5)
  • updated index to export linear-svc and updated docs (b5c116b)

1.0.0 (2021-12-31)

Bug Fixes

  • broken UMD browser script (10f2e34)
  • fix lint (28c876d)
  • k-neighbors inverse distance weighting (a162baa)
  • k-neighbors predict now checks n_features (35efd93)

Features

  • add SVC (a5fe596)
  • added cut 1 of voting classifier (4045b81)
  • added tests and basic implementation of votingregressor (d7011e7)
  • added voting classifier (d7ab9c6)
  • broke out sgdlinear into sgdregressor and sgdclassifier (81fbee8)
  • changed imports (390375c)
  • finish (825ebb7)
  • First pass at VotingRegressor (ffb3393)
  • implemented kNeighborsRegression (a1a7174)
  • import libsvm (f0f0cc8)
  • k-neighbors regressor (94f6a69)
  • k-neighbors regressor (225e167)
  • k-neighbors-classifier implemented (d120257)
  • k-neighbors-regressor (050cec6), closes #111
  • k-neighbors-regressor (cb0a8b0)
  • linear svr (7a0534d)
  • simple first pass addition of linear-svc (483117d)
  • train test split implementation (97b89a5)
  • updated index to export linear-svc and updated docs (b5c116b)