NimbusML
NimbusML copied to clipboard
microsoft
Python machine learning package providing simple interoperability between ML.NET and scikit-learn components.
**Describe the bug** When Transform is called without calling Fit, a ValueError is thrown. Something similar to sklearn.exceptions.NotFittedError would be more appropriate? **To Reproduce** Steps to reproduce the behavior: Create...
The docs on https://docs.microsoft.com/en-us/nimbusml/concepts/metrics#multiclass-classification-metrics are wrong since it states the following: ``` This corresponds to evaltype=’multiclass’. [...] (class N) - Accuracy of class N ``` Problem is that the `(class...
The nimbusml documentation says it's a quantile regression forest implementation, but I don't see where to input different quantiles for training or prediction and I don't see a quantiles arg...
Repro `from nimbusml.datasets import get_dataset from nimbusml.preprocessing import OnnxRunner, ToKey iris_df = get_dataset("iris").as_df() iris_df = iris_df.drop(['Label'], axis=1) transform = ToKey()
In production environment, we saw BridgeRuntimeException (bre) with bre.callstack property empty. This hasn't happened in the past. On talking to Gani, it appears that this could happens if c++ part...
**Describe the bug** Using `sys.version_info` to modify dependencies is not a recommended way according to sdispater/poetry#758. We should use [environment markers](https://www.python.org/dev/peps/pep-0508/#environment-markers) instead. If we use `sys.version_info` to modify dependencies, some...
I am interested in using LightLDA; however I am unable to find how to extract the topic word distribution after training the model (with fit_transform only the doc_topic distribution is...
**Describe the bug** Tests in test_wordembedding.py are being skipped on all versions of Python on CentOS7 and Ubuntu 14. Here is the blurb stating the "todo fix": https://github.com/microsoft/NimbusML/blob/b57cfccd3f493c21e6e61f4cb80c8ccef89e2505/src/python/nimbusml/tests/feature_extraction/text/test_wordembedding.py#L22-L40 In short,...
Currently if you DatasetTransformer with predictor model it outputs all the hidden fields. It needs to ouput only Score and optionally PredictedLabel if its classifier for ex, Probabilities if available.