mleap
mleap copied to clipboard
combust
MLeap: Deploy ML Pipelines to Production
Hi, we would like to build mleap-serving images using a different base image (RedHat UBI-based rather than Ubuntu), and so were wondering if the Dockerfile and any other build logic...
Deserialization of scikit pipelines seems not implemented see https://github.com/combust/mleap/blob/d276c1c800277a1edcf30dda6e8559c7413b21f8/python/mleap/sklearn/pipeline.py#L121
In `transform row stream` of TransformServiceSpec: https://travis-ci.org/combust/mleap/builds/590427424#L4478-L4493 [info] transform row stream 4479[info] - transforms rows in a stream 4480[info] throttling 4481[info] - throttles elements through the flow *** FAILED ***...
See https://github.com/combust/mleap/pull/686#discussion_r434005552
I'm attempting to rebuild our ML pipeline using MLeap to serve our models. Our transformers are written in Java and mostly extend Spark's `UnaryTransformer` class. They all also start with...
Currently sklearn logistic classifier only supports predictions. https://github.com/neilsummers/mleap/blob/master/python/mleap/sklearn/logistic.py ``` def serialize_to_bundle(self, transformer, path, model_name): # compile tuples of model attributes to serialize attributes = list() attributes.append(('intercept', transformer.intercept_.tolist()[0])) attributes.append(('coefficients', transformer.coef_.tolist()[0])) attributes.append(('num_classes',...
Since there doesn't seem to be a mention of the reverse_string_indexer in mleap/python/mleap/sklearn/preprocessing/, I don't believe the ReverseStringIndexer is currently supported for sklearn.
# Standardizing ML Pipeline Serialization Currently there is a large array of serialization formats for machine learning models: 1. PMML is an XML-based format primarily targeting the JVM for executing...
LDA uses a random sampling to score data. This causes slightly different values to to be generated between each run, which makes testing outright equality of tensors a no-go for...