Hunter McGushion
Hunter McGushion
- Add regression example script/notebook to "examples/sklearn_examples/" - Format should mirror that of examples in other subdirs
- Using HyperparameterHunter with the recently released Keras version 2.3.0 raises metaclass conflict error, breaking Keras compatibility - **TEMPORARY WORKAROUND: Use Keras 2.2.5** ## Error Traceback - From executing `examples/keras_examples/experiment_example.py`...
- Fix bug matching Experiment Descriptions including hyperparameter types not handled by JSON - Accomplish above by switching to YAML Descriptions - Can revert back to JSON Descriptions via `settings.G.description_format`
- Introduce :class:`feature_engineering.FeatureEngineer` in project README as new core library feature - Brief overview, which links to the extended notebook: "examples/advanced_examples/feature_engineering.ipynb"
- Convert it and its abstract definition at :meth:`BaseCVExperiment._initialize_folds` to static methods - The new static method should receive `cv_type` as input, and it should return `folds` - Add doctests
## Note - This issue (along with several others) was originally opened by @caprone in #111, but it is being moved here for easier tracking - If this issue is...
- Unable to supply `validation_data` to a Keras `CVExperiment` via `model_extra_params[“fit”]` - This is because HyperparameterHunter automatically sets `validation_data` to be the OOF data produced by the cross validation scheme...
- `Embedding` layer can’t be given initialized weights - `Embedding` alone works if it initializes own weights - Needs function parameters to be given weight initialization ## Note - This...
- Experiment matching incorrectly considers a string `kernel_initializer` value to be different from a callable value for the same function, with the default arguments - This problem may be encountered...