svolchkov
svolchkov
Looks like I'm having the exact same issue with TabularLIME. Tried reducing numSamples, no result
TabularSHAP also hangs indefinitely. Using pyspark's PipelineModel including a VectorAssembler and a RandomForestClassificationModel
I get a feeling this is not intended for a large number of features. At least TabularSHAP. Because https://github.com/microsoft/SynapseML/blob/master/core/src/main/scala/com/microsoft/azure/synapse/ml/explainers/KernelSHAPBase.scala#L134 . We have 38 input features and I suspect this will...