[SYSTEMDS-1566] Possible regression from 0.13 -> 0.14 for MNIST LeNet script

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: SystemML 0.15
• Assignee: Mike Dusenberry
• Reporter: Mike Dusenberry
• Created: 29/Apr/17 01:13
• Updated: 09/Sep/17 05:05
• Resolved: 23/May/17 19:00

Description

For the 0.14 release testing, I tried out the MNIST LeNet example on both 0.13 and 0.14 and noticed a possible regression. Basically, on 0.14 the script took longer to run and had 2513 Spark instructions executed, while on 0.13 only 864 Spark instructions were executed. This was run locally on a laptop using the 2 instructions at the top of the script (and copied below). I've also attached the stats and runtime explain logs.

1. Download data

nn/examples/get_mnist_data.sh

2. Execute from the scripts directory.

spark-submit --master local[*] --driver-memory 10G --conf spark.driver.maxResultSize=0 --conf spark.rpc.message.maxSize=128 SystemML.jar -f nn/examples/mnist_lenet-train.dml -stats -explain -nvargs train=nn/examples/data/mnist/mnist_train.csv test=nn/examples/data/mnist/mnist_test.csv C=1 Hin=28 Win=28 epochs=1 out_dir=nn/examples/model/mnist_lenet

Issue Links

• relates to: SYSTEMDS-540 Deep Learning
• relates to: SYSTEMDS-618 Deep Learning DML Library
• relates to: SYSTEMDS-1554 IPA Scalar Transient Read Replacement
• relates to: SYSTEMDS-1561 Improve constant folding during compilation