Deep Learning Time Series

vinhdiesal
vinhdiesal Registered Posts: 11 ✭✭✭✭
edited July 16 in Using Dataiku

I'm following the deep learning time-series tutorial and get the following errors when training the models.

[2020-05-07 15:41:21,123] [9793/MainThread] [WARNING] [tensorflow] From /Users/vinhtran/Library/DataScienceStudio/dss_home/code-envs/python/Python27/lib/python2.7/site-packages/tensorflow_core/python/ops/resource_variable_ops.py:1630: calling __init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.
Instructions for updating:
If using Keras pass *_constraint arguments to layers.
[2020-05-07 15:41:21,339] [9793/MainThread] [INFO] [dataiku.doctor.utils.listener] END -  Fitting model
Traceback (most recent call last):
  File "/Applications/DataScienceStudio.app/Contents/Resources/kit/python/dataiku/doctor/server.py", line 47, in serve
    ret = api_command(arg)
  File "/Applications/DataScienceStudio.app/Contents/Resources/kit/python/dataiku/doctor/dkuapi.py", line 45, in aux
[2020/05/07-15:41:21.339] [MRT-1879] [INFO] [dku.block.link.interaction]  - Check result for nullity exceptionIfNull=true result=null
    return api(**kwargs)
  File "/Applications/DataScienceStudio.app/Contents/Resources/kit/python/dataiku/doctor/commands.py", line 453, in train_prediction_keras
    pipeline.generated_features_mapping)
  File "/Applications/DataScienceStudio.app/Contents/Resources/kit/python/dataiku/doctor/prediction_entrypoints.py", line 437, in prediction_train_model_keras
    generated_features_mapping)
  File "/Applications/DataScienceStudio.app/Contents/Resources/kit/python/dataiku/doctor/deep_learning/keras_support.py", line 86, in get_keras_model
    check_model_output_dimension(model, prediction_type, target_map, modeling_params)
  File "/Applications/DataScienceStudio.app/Contents/Resources/kit/python/dataiku/doctor/deep_learning/keras_support.py", line 154, in check_model_output_dimension
    "It is currently: {}".format(len(target_map), output_shape[-1]))
ValueError: For this multiclass classification problem, output of Deep Learning Architecture must have a dimension equal to 224 (number of classes). It is currently: 1
[2020/05/07-15:41:21.499] [KNL-python-single-command-kernel-monitor-1883] [INFO] [dku.kernels]  - Process done with code 0
[2020/05/07-15:41:21.500] [KNL-python-single-command-kernel-monitor-1883] [INFO] [dip.tickets]  - Destroying API ticket for analysis-ml-TEMPERATURE-afTttfu on behalf of admin
[2020/05/07-15:41:21.500] [MRT-1879] [INFO] [dku.kernels]  - Getting kernel tail
[2020/05/07-15:41:21.501] [MRT-1879] [INFO] [dku.kernels]  - Trying to enrich exception: com.dataiku.dip.io.SocketBlockLinkKernelException: Failed to train : <type 'exceptions.ValueError'> : For this multiclass classification problem, output of Deep Learning Architecture must have a dimension equal to 224 (number of classes). It is currently: 1 from kernel com.dataiku.dip.analysis.coreservices.AnalysisMLKernel@313b0caf process=null pid=?? retcode=0
[2020/05/07-15:41:21.501] [MRT-1879] [WARN] [dku.analysis.ml.python]  - Training failed
com.dataiku.dip.io.SocketBlockLinkKernelException: Failed to train : <type 'exceptions.ValueError'> : For this multiclass classification problem, output of Deep Learning Architecture must have a dimension equal to 224 (number of classes). It is currently: 1
    at com.dataiku.dip.io.SocketBlockLinkInteraction.throwExceptionFromPython(SocketBlockLinkInteraction.java:302)
    at com.dataiku.dip.io.SocketBlockLinkInteraction$AsyncResult.checkException(SocketBlockLinkInteraction.java:215)
    at com.dataiku.dip.io.SocketBlockLinkInteraction$AsyncResult.get(SocketBlockLinkInteraction.java:190)
    at com.dataiku.dip.io.SingleCommandKernelLink$1.call(SingleCommandKernelLink.java:208)
    at com.dataiku.dip.analysis.ml.prediction.PredictionTrainAdditionalThread.process(PredictionTrainAdditionalThread.java:74)
    at com.dataiku.dip.analysis.ml.shared.PRNSTrainThread.run(PRNSTrainThread.java:137)

Before I trained the models, I noticed warning of classes greater than 50.

Answers

  • Mattsco
    Mattsco Dataiker, Registered Posts: 125 Dataiker

    Hi,
    The error says that for this multiclass classification problem, output of Deep Learning Architecture must have a dimension equal to 224 (number of classes). And in your code it is currently: 1
    I think it's because you need to select Regression instead of Multiclass model in Basic > Target settings.

  • vinhdiesal
    vinhdiesal Registered Posts: 11 ✭✭✭✭

    I still get the same error when selecting the Multiclass model .

    I also noticed that it errors out with the windowprocessor.py where nan is not defined. I tried something where I deleted the windowprocessor.py and when I hit the train button, I get the same exact error even when the file is deleted. I think there is something corrupted in the memory is there a way to clear it?

Setup Info
    Tags
      Help me…