featuretools
featuretools copied to clipboard
alteryx
An open source python library for automated feature engineering
`get_function` is repeated throughout every primitive. I purpose we explore the idea of making this available as a decorator instead which wraps the function needed to run in order to...
- https://github.com/alteryx/featuretools/pull/2111
- As a user of Featuretools, I want the library to support cuDF DataFrames, and take advantage of GPU acceleration.
The `And` Primitive has unneccessary inputs. Since it is commutative, the additional permutation is unneccessary. ```python [ ColumnSchema(logical_type=Boolean), ColumnSchema(logical_type=BooleanNullable), ], [ ColumnSchema(logical_type=BooleanNullable), ColumnSchema(logical_type=Boolean), ] ```
Both when I add Spark DataFrames to my EntitySet and when I call `.dfs()` on the Spark EntitySet, I see a flood of warnings: ```22/04/26 16:41:09 WARN WindowExec: No Partition...
There are a number of assumptions we make about the shape of Primitive `input_types` lists: - Its either a list of ColumnSchema objects or a list of lists of ColumnSchema...
* Use woodwork's `dependence` method to calculate correlations or logic behind the `dependence` function * Improve decision making for which feature to keep. Lowest depth feature with some other metric...
As a developer, I wish I could use Featuretools to split my EntitySet into train and test datasets. Sometimes all of the data will be included when an EntitySet is...
### Featuretools generates stacked features on top of output from multi-output primitives that cause error upon calculation #### Bug Description In some cases Featuretools will generate features that will result...
- Add Lag primitive, which can be helpful for time-series problems ```python from featuretools.primitives.base import TransformPrimitive from featuretools.variable_types import Variable class LagByPeriods(TransformPrimitive): """Shifts an array of values by a specified...