EpistasisLab/tpot2

Issues

• Public API / documentation for # generations executed

  #161 opened a month ago by chimaerase
  0

• Remove jupyter as a dependency (+maybe others)

  #146 opened 4 months ago by chimaerase
  4

• Replace remaining calls to legacy `np.random.seed()`

  #112 opened a year ago by chimaerase
  1

• allow individual modules to be specified by string for config dicts. For example ["LogisticRegression", "KNN"] etc.

  #60 opened 3 months ago by perib
  1

• #TODO, mutate all steps or just one?

  #154 opened 3 months ago by jgh9094
  0

• Forgot to instantiate the rng.

  #151 opened 3 months ago by jgh9094
  2

• Should 'classification=True' be 'classification=self.classification'?

  #150 opened 3 months ago by jgh9094
  2

• 'np.random.normal(0, 1)' should be 'rng.normal(0, 1)'

  #152 opened 3 months ago by jgh9094
  0

• Add a destroyer function to properly close dask clients when the evolver or estimator instance is deleted.

  #70 opened 10 months ago by perib
  0

• Learning Machine

  #117 opened a year ago by RADj375
  1

• Can we improve the configuration files to be easier to read and write?

  #78 opened a year ago by perib
  0

• Fix miscellaneous invalid hyperparameter options in the configuration dictionary. And add n_jobs=1 when possible.

  #59 opened a year ago by perib
  2

• API : improve selector module organization

  #87 opened a year ago by perib
  0

• TPOT1 feature Parity: cuML configuration dictionaries

  #101 opened a year ago by perib
  0

• Rename survival_percentage to something clearer.

  #89 opened a year ago by perib
  0

• Automatically generate names for lambda/partial functions used as custom objective functions

  #88 opened a year ago by perib
  0

• Implement Meta Learner

  #91 opened a year ago by perib
  0

• Idea for a two step preprocessing pipeline

  #100 opened a year ago by perib
  0

• Add support for ordinal categorical columns

  #98 opened a year ago by perib
  0

• Better integrate Optuna Optimization

  #90 opened a year ago by perib
  0

• implement a callback function/class

  #99 opened a year ago by perib
  0

• Genetic Feature Selection Improvements

  #96 opened a year ago by perib
  0

• optimization of preprocessing pipeline

  #95 opened a year ago by perib
  0

• add support for regressors in the complexity objective function

  #94 opened a year ago by perib
  0

• This issue for TPOT1 brought up a good idea about outlier detection https://github.com/EpistasisLab/tpot/issues/1166

  #93 opened a year ago by perib
  0

• Add option for successive halving based on time or convergence rather than number of generations.

  #92 opened a year ago by perib
  0

• add in unit tests

  #80 opened a year ago by perib
  0

• TPOT1 Feature Parity: Write an export function that takes a graphpipeline and generates a .py file similar to TPOT1

  #58 opened a year ago by perib
  0

• TPOT1 Feature Parity: add "sample_weight" and "groups" parameters to the fit function in TPOTEstimator and the graphpipeline

  #57 opened a year ago by perib
  0

• Add successive halving algorithm to steady state

  #66 opened a year ago by perib
  0

• add a debugging tutorial

  #64 opened a year ago by perib
  0

• Avoid duplicate work for steady state and base evolvers. Have one evolver that can do both generational EA, but also allow for extra individuals to be generated on the fly to make use of idle cores.

  #65 opened a year ago by perib
  0

• Support custom initializations / manually defined initial pipelines -

  #61 opened a year ago by perib
  0

• rename baseevolver.optimize to .evolve? To avoid confusion with individual.optimize

  #79 opened a year ago by perib
  0

• Licensing text in source files

  #77 opened a year ago by perib
  0

• Make the cv early stop functions into standalone functions? Or remove?

  #69 opened a year ago by perib
  1

• Test that TPOTEstimator, TPOTClassifier, TPOTRegressor have a) same params, b) meet the sklearn baseestimator conventions

  #81 opened a year ago by perib
  0

• shuffle cv folds every generation?

  #82 opened a year ago by perib
  0

• Idea: Preventing getting stuck with variable number of mutations

  #84 opened a year ago by perib
  0

• add a scorer that counts selected number of features

  #83 opened a year ago by perib
  0

• Idea for Successive Halving to avoid repeated re-evaluations of the same pipeline.

  #85 opened a year ago by perib
  0

• edit/consolidate default configuration dictionary

  #76 opened a year ago by perib
  0

• replace instances of get_pareto_front with get_pareto_frontier

  #75 opened a year ago by perib
  0

• Avoiding repeated work : Cache node models with node hashing or manually.

  #73 opened a year ago by perib
  0

• add covariate adjustment

  #68 opened a year ago by perib
  0

• function that generates a learning curve

  #63 opened a year ago by perib
  0

• better plotting for graph pipeline

  #62 opened a year ago by perib
  0

• better thread efficiency for cv prunning with thresholds

  #72 opened a year ago by perib
  0

• Memory caching with GraphPipeline may miss some nodes where the ordering on inputs happens to be different between two nodes. - potential solution is to make graphs canonical - there is a C package that can be imported into TPOT that may do this

  #74 opened a year ago by perib
  0

• FeatureSetSelector module needs to implement a mask to be consistent with sklearn

  #71 opened a year ago by perib
  0