pieths
pieths
The Windows unit tests are randomly crashing during the CI runs. It does not appear to happen locally (Windows + Python3.7). See [this pull request](https://github.com/microsoft/NimbusML/pull/331) for more information. The following...
When specifying `split_start='after_transforms'` in `CV.fit()`, the user defined transforms are not split up correctly. See the graph created by the `fit()` call in the code below. It seems like if...
NimbusML only has support for string based categorical columns. Numerical categorical columns (`KeyDataViewType`) which are returned from ML.Net are not converted back to their original representation even though Pandas does...
Using a transform, which only acts on a subset of the input columns, before a predictor and not explicitly specifying the features to the predictor will only pass the output...
Currently, each return value is transferred from managed code to native code one at a time. See `NativeDataInterop.cs`. ```csharp for (int i = 0; i < fillers.Length; i++) { fillers[i].Set();...
The inputs and outputs which are produced by `Pipeline.get_fit_info` are not valid. See `inputs`, `outputs` and `schema_after` in the `RangeFilter` section of the output: ```python train_data = {'c1': [2, 3,...
The online documentation for transforms and predictors does not currently show (or link to) the base class documentation. For instance, the docs for [`NGramFeaturizer`](https://docs.microsoft.com/en-us/python/api/nimbusml/nimbusml.feature_extraction.text.ngramfeaturizer?view=nimbusml-py-latest) do not show the `fit_transform()` and...
The tools which are used to build the documentation in `src\python\docs\sphinx` are no longer working. Here are a few of the errors which are produced when building the documentation: ```console...
Scikit-Learn has support for combining conceptually different machine learning predictors and returning the average/median/majority vote of the predicted values. See [VotingClassifer and VotingRegressor](https://scikit-learn.org/stable/modules/ensemble.html#voting-regressor). Could this functionality be added to NimbusML?
``` python import numpy as np import pandas as pd from nimbusml import FileDataStream from nimbusml.preprocessing.missing_values import Handler # 1 less than the maximum positive int32 value # See: https://docs.scipy.org/doc/numpy/user/basics.types.html...