Stuck in Building model on training data #76
Comments
sangyx
changed the title
|
Unfortunately, ARM chips are not supported by the old release of deeplearning4j that is used by WEKA. Updating the wekaDeepLearning4j package to use a more recent release of deeplearning4j is on the to-do list.
If you are happy to write Python code, another option for deep learning is the weka-pyscript package:
https://github.com/christopher-beckham/weka-pyscript
Cheers,
Eibe
From: Yunxin Sang ***@***.***>
Date: Monday, 6 May 2024 at 6:40 PM
To: Waikato/wekaDeeplearning4j ***@***.***>
Cc: Subscribed ***@***.***>
Subject: [Waikato/wekaDeeplearning4j] Stuck on Building model on training data (Issue #76)
Describe the bug
When I tried to follow the tutorial "Starting Simple - The Iris Dataset (Tabular)"<https://deeplearning.cms.waikato.ac.nz/examples/train_cnn/>. When I click the "Start", it stucks in the Status "Building model on training data"
Screenshots
If applicable, add screenshots to help explain your problem.
image.png (view on web)<https://github.com/Waikato/wekaDeeplearning4j/assets/32364921/b34b451b-5d9e-480c-83dd-4b278652ed64>
log:
01:29:03: Weka Explorer
01:29:03: (c) 1999-2022 The University of Waikato, Hamilton, New Zealand
01:29:03: web: https://www.cs.waikato.ac.nz/~ml/weka/
01:29:03: Started on Monday, 6 May 2024
01:29:19: Base relation is now iris (150 instances)
01:31:55: Started weka.classifiers.functions.Dl4jMlpClassifier
01:31:55: Command: weka.classifiers.functions.Dl4jMlpClassifier -S 1 -cache-mode MEMORY -early-stopping "weka.dl4j.earlystopping.EarlyStopping -maxEpochsNoImprovement 0 -valPercentage 0.0" -normalization "Standardize training data" -iterator "weka.dl4j.iterators.instance.DefaultInstanceIterator -bs 1" -iteration-listener "weka.dl4j.listener.EpochListener -eval true -n 5" -layer "weka.dl4j.layers.OutputLayer -lossFn \"weka.dl4j.lossfunctions.LossMCXENT \" -nOut 2 -activation \"weka.dl4j.activations.ActivationSoftmax \" -name \"Output layer\"" -logConfig "weka.core.LogConfiguration -append true -dl4jLogLevel WARN -logFile /Users/ys26425/wekafiles/wekaDeeplearning4j.log -nd4jLogLevel INFO -wekaDl4jLogLevel INFO" -config "weka.dl4j.NeuralNetConfiguration -biasInit 0.0 -biasUpdater \"weka.dl4j.updater.Sgd -lr 0.001 -lrSchedule \\\"weka.dl4j.schedules.ConstantSchedule -scheduleType EPOCH\\\"\" -dist \"weka.dl4j.distribution.Disabled \" -dropout \"weka.dl4j.dropout.Disabled \" -gradientNormalization None -gradNormThreshold 1.0 -l1 NaN -l2 NaN -minimize -algorithm STOCHASTIC_GRADIENT_DESCENT -updater \"weka.dl4j.updater.Adam -beta1MeanDecay 0.9 -beta2VarDecay 0.999 -epsilon 1.0E-8 -lr 0.001 -lrSchedule \\\"weka.dl4j.schedules.ConstantSchedule -scheduleType EPOCH\\\"\" -weightInit XAVIER -weightNoise \"weka.dl4j.weightnoise.Disabled \"" -numEpochs 10 -numGPUs 1 -averagingFrequency 10 -prefetchSize 24 -queueSize 0 -zooModel "weka.dl4j.zoo.CustomNet -channelsLast false -pretrained NONE"
Additional Information
* Weka version: 3.9.6
* wekaDeeplearning4j package version: Install in Package Manager, 1.7.2
* Operating System: macOS Sonoma Version 14.4.1 - M2 Pro
Additional context
Add any other context about the problem here.
I also can't see the result of "Is CPU Available?"
image.png (view on web)<https://github.com/Waikato/wekaDeeplearning4j/assets/32364921/ccbca9c6-bb1d-4356-a275-eeca20248db1>
—
Reply to this email directly, view it on GitHub<#76>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACS2HVLRR6SPY4RXSZVRICTZA4QVLAVCNFSM6AAAAABHIO6QAWVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI4DAMJVHE4DMNY>.
You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>
|
|
@eibe Thanks for your reply. Do you have a timeline for updating wekaDeeplearning4j? We're using Weka in an MBA class and trying to add more advanced content like CNN into our class. |
|
The plan is to have this update done by the end of June.
Cheers,
Eibe
From: Yunxin Sang ***@***.***>
Date: Thursday, 9 May 2024 at 2:29 PM
To: Waikato/wekaDeeplearning4j ***@***.***>
Cc: Eibe Frank ***@***.***>, Mention ***@***.***>
Subject: Re: [Waikato/wekaDeeplearning4j] Stuck in Building model on training data (Issue #76)
@eibe<https://github.com/eibe> Thanks for your reply. Do you have a timeline for updating wekaDeeplearning4j? We're using Weka in an MBA class and trying to add more advanced content like CNN into our class.
—
Reply to this email directly, view it on GitHub<#76 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACS2HVNKICAPH536TYKNQDDZBLNPVAVCNFSM6AAAAABHIO6QAWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMBRHAYTOMJYGU>.
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the bug
When I tried to follow the tutorial "Starting Simple - The Iris Dataset (Tabular)". It stucked in the Status "Building model on training data" after I clicked "start".
Screenshots
If applicable, add screenshots to help explain your problem.
log:
Additional Information
Additional context
Add any other context about the problem here.
I also can't see the result of "Is CPU Available?"
The text was updated successfully, but these errors were encountered: