[
https://issues.apache.org/jira/browse/SYSTEMML-1076?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mike Dusenberry updated SYSTEMML-1076:
--------------------------------------
Affects Version/s: SystemML 0.11
> Sparse Max Pooling Unsupported
> ------------------------------
>
> Key: SYSTEMML-1076
> URL: https://issues.apache.org/jira/browse/SYSTEMML-1076
> Project: SystemML
> Issue Type: Bug
> Affects Versions: SystemML 0.11
> Reporter: Mike Dusenberry
> Priority: Blocker
>
> Currently, our max pooling built-in operator does not support sparse
> matrices. However, sparse matrices are a common possibility in neural nets
> that make use of ReLU and Dropout layers, and are likely to happen at random
> depending on exact characteristics of intermediate matrices during training.
> Therefore, we should accept sparse matrices as input.
> {code}
> Caused by: org.apache.sysml.runtime.DMLRuntimeException: Sparse
> maxpooling_backward is not supported
> at
> org.apache.sysml.runtime.matrix.data.LibMatrixDNN.maxpooling_backward(LibMatrixDNN.java:526)
> at
> org.apache.sysml.runtime.instructions.cp.ConvolutionCPInstruction.processInstruction(ConvolutionCPInstruction.java:205)
> at
> org.apache.sysml.runtime.controlprogram.ProgramBlock.executeSingleInstruction(ProgramBlock.java:305)
> ... 26 more
> {code}
> Reproducible example (Python):
> {code}
> script = """
> N = 2
> C = 3
> Hin = 10
> Win = 10
> Hout = 5
> Wout = 5
> f = 2
> stride = 2
> X = matrix(0, rows=N, cols=C*Hin*Win)
> dout = matrix(0, rows=N, cols=C*Hout*Wout)
> out = max_pool_backward(X, dout, input_shape=[N,C,Hin,Win], pool_size=[f,f],
> stride=[stride,stride], padding=[0,0])
> """
> script = dml(script).output("out")
> out = ml.execute(script).get("out")
> l = out.toDF().collect()
> l
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
