Renamed Vset.modules to Vset.vfuncs and Vset.out to Vset.fitted_vfuncs

[ssaxena00]

• Still have to change docs (how to do this?)
• Can still change:
  • ‘apply_modules’ -> ‘apply_vfuncs’ (in utils.py),
  • ‘modules’ -> ‘funcs’ (in vset.py)
  • 'module_keys' -> 'func_keys' (in vset.py)

[jpdunc23]

Let's also change the init argument names from modules and module_keys to funcs and func_keys.

[ssaxena00]

Should I also change apply_modules -> apply_vfuncs (in utils.py)?

[jpdunc23]

Yes, please. Let's replace all usage of module with vfunc.

[ssaxena00]

Discovered errors in p_check.ipynb:

• removed compute_Interval ?
• in executing filtered_rf_set.fit(X_train_clean, y_train):
  • got TypeError: float() argument must be a string or a number, not 'dict'

[ssaxena00]

Discovered errors in build_Vset.ipynb:

• call to build_vset is not resulting in correct .vfuncs attribute
  • length is 1 and vfunc looks like:
    RandomForestRegressor(criterion='absolute_error', n_estimators={'max_features': ['sqrt', 'log2'], 'min_samples_split': [2, 10], 'n_estimators': [100, 200, 300]})

[ssaxena00]

Discovered errors in 03_computer_vision_dnn:

• while executing modeling_set.fit(train_data):
  • got TypeError: train() got multiple values for argument 'loss_fn'

[ssaxena00]

Causing error in 04_feat_importance_stability.ipynb because build_vset not setting .vfuncs attribute correctly

[ssaxena00]

Discovered error in mlflow_logging_hardcoded:

• while executing train(C, min_samples_leaf:
  • got KeyError: ('X_test', (X_train, X_train, LR, 'X_test'), 'Acc')

[jpdunc23]

Discovered error in mlflow_logging_hardcoded:

• while executing train(C, min_samples_leaf:

  • got KeyError: ('X_test', (X_train, X_train, LR, 'X_test'), 'Acc')

@ssaxena00 We can delete that notebook. Also feel free to replace compute_interval with perturbation_stats and let me know if usage of the latter isn't clear.