libc++abi: terminating with uncaught exception of type NSException

Machine Learning & AI General
DDHH OP
Created Oct ’21
Replies 2
Boosts 0
Views 2.8k
Participants 3 
creating model "model3"
Metal device set to: Apple M1

systemMemory: 16.00 GB
maxCacheSize: 5.33 GB

Episode 1 of 1
2021-10-14 17:46:13.626 python[10722:54381] -[MPSGraph adamUpdateWithLearningRateTensor:beta1Tensor:beta2Tensor:epsilonTensor:beta1PowerTensor:beta2PowerTensor:valuesTensor:momentumTensor:velocityTensor:maximumVelocityTensor:gradientTensor:name:]: unrecognized selector sent to instance 0x16acdc810
2021-10-14 17:46:13.640 python[10722:54381] *** Terminating app due to uncaught exception 'NSInvalidArgumentException', reason: '-[MPSGraph adamUpdateWithLearningRateTensor:beta1Tensor:beta2Tensor:epsilonTensor:beta1PowerTensor:beta2PowerTensor:valuesTensor:momentumTensor:velocityTensor:maximumVelocityTensor:gradientTensor:name:]: unrecognized selector sent to instance 0x16acdc810'
*** First throw call stack:
(
        0   CoreFoundation                      0x0000000182887838 __exceptionPreprocess + 240
        1   libobjc.A.dylib                     0x00000001825b10a8 objc_exception_throw + 60
        2   CoreFoundation                      0x0000000182918694 -[NSObject(NSObject) __retain_OA] + 0
        3   CoreFoundation                      0x00000001827e8cd4 ___forwarding___ + 1444
        4   CoreFoundation                      0x00000001827e8670 _CF_forwarding_prep_0 + 96
        5   libmetal_plugin.dylib               0x000000015630a290 _ZN12metal_plugin14MPSApplyAdamOpIfEC2EPNS_20OpKernelConstructionE + 656
        6   libmetal_plugin.dylib               0x0000000156309ebc _ZN12metal_pluginL14CreateOpKernelINS_14MPSApplyAdamOpIfEEEEPvP23TF_OpKernelConstruction + 52
        7   libtensorflow_framework.2.dylib     0x000000012a00c5d4 _ZN10tensorflow12_GLOBAL__N_120KernelBuilderFactory6CreateEPNS_20OpKernelConstructionE + 88
        8   libtensorflow_framework.2.dylib     0x000000012a08e158 _ZN10tensorflow14CreateOpKernelENS_10DeviceTypeEPNS_10DeviceBaseEPNS_9AllocatorEPNS_22FunctionLibraryRuntimeEPNS_11ResourceMgrERKNSt3__110shared_ptrIKNS_14NodePropertiesEEEiPPNS_8OpKernelE + 784
        9   libtensorflow_framework.2.dylib     0x000000012a2692b8 _ZN10tensorflow21CreateNonCachedKernelEPNS_6DeviceEPNS_22FunctionLibraryRuntimeERKNSt3__110shared_ptrIKNS_14NodePropertiesEEEiPPNS_8OpKernelE + 272
        10  libtensorflow_framework.2.dylib     0x000000012a213c20 _ZN10tensorflow26FunctionLibraryRuntimeImpl12CreateKernelERKNSt3__110shared_ptrIKNS_14NodePropertiesEEEPNS_22FunctionLibraryRuntimeEPPNS_8OpKernelE + 600
        11  libtensorflow_framework.2.dylib     0x000000012a27e430 _ZN10tensorflow22ImmutableExecutorState10InitializeERKNS_5GraphE + 1192
        12  libtensorflow_framework.2.dylib     0x000000012a269064 _ZN10tensorflow16NewLocalExecutorERKNS_19LocalExecutorParamsERKNS_5GraphEPPNS_8ExecutorE + 304
        13  libtensorflow_framework.2.dylib     0x000000012a276e6c _ZN10tensorflow12_GLOBAL__N_124DefaultExecutorRegistrar7Factory11NewExecutorERKNS_19LocalExecutorParamsERKNS_5GraphEPNSt3__110unique_ptrINS_8ExecutorENS9_14default_deleteISB_EEEE + 48
        14  libtensorflow_framework.2.dylib     0x000000012a2777e8 _ZN10tensorflow11NewExecutorERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERKNS_19LocalExecutorParamsERKNS_5GraphEPNS0_10unique_ptrINS_8ExecutorENS0_14default_deleteISG_EEEE + 92
        15  libtensorflow_framework.2.dylib     0x000000012a216278 _ZN10tensorflow26FunctionLibraryRuntimeImpl10CreateItemEPPNS0_4ItemE + 2676
        16  libtensorflow_framework.2.dylib     0x000000012a21706c _ZN10tensorflow26FunctionLibraryRuntimeImpl3RunERKNS_22FunctionLibraryRuntime7OptionsEyN4absl12lts_202103244SpanIKNS_6TensorEEEPNSt3__16vectorIS8_NSB_9allocatorIS8_EEEENSB_8functionIFvRKNS_6StatusEEEE + 676
        17  libtensorflow_framework.2.dylib     0x000000012a2250c0 _ZNK10tensorflow29ProcessFunctionLibraryRuntime14RunMultiDeviceERKNS_22FunctionLibraryRuntime7OptionsEyPNSt3__16vectorIN4absl12lts_202103247variantIJNS_6TensorENS_11TensorShapeEEEENS5_9allocatorISC_EEEEPNS6_INS5_10unique_ptrINS0_11CleanUpItemENS5_14default_deleteISI_EEEENSD_ISL_EEEENS5_8functionIFvRKNS_6StatusEEEENSP_IFSQ_RKNS0_21ComponentFunctionDataEPNS0_12InternalArgsEEEE + 2640
        18  libtensorflow_framework.2.dylib     0x000000012a228098 _ZNK10tensorflow29ProcessFunctionLibraryRuntime3RunERKNS_22FunctionLibraryRuntime7OptionsEyN4absl12lts_202103244SpanIKNS_6TensorEEEPNSt3__16vectorIS8_NSB_9allocatorIS8_EEEENSB_8functionIFvRKNS_6StatusEEEE + 2012
        19  libtensorflow_framework.2.dylib     0x000000012a228868 _ZNK10tensorflow29ProcessFunctionLibraryRuntime7RunSyncERKNS_22FunctionLibraryRuntime7OptionsEyN4absl12lts_202103244SpanIKNS_6TensorEEEPNSt3__16vectorIS8_NSB_9allocatorIS8_EEEE + 160
        20  _pywrap_tensorflow_internal.so      0x000000013c175554 _ZN10tensorflow19KernelAndDeviceFunc3RunEPNS_19ScopedStepContainerERKNS_15EagerKernelArgsEPNSt3__16vectorIN4absl12lts_202103247variantIJNS_6TensorENS_11TensorShapeEEEENS6_9allocatorISD_EEEEPNS_19CancellationManagerERKNS9_8optionalINS_25EagerRemoteFunctionParamsEEERKNSK_INS_17ManagedStackTraceEEE + 516
        21  _pywrap_tensorflow_internal.so      0x000000013c13fd60 _ZN10tensorflow18EagerKernelExecuteEPNS_12EagerContextERKN4absl12lts_2021032413InlinedVectorIPNS_12TensorHandleELm4ENSt3__19allocatorIS6_EEEERKNS3_8optionalINS_25EagerRemoteFunctionParamsEEERKNS7_10unique_ptrINS_15KernelAndDeviceENS_4core15RefCountDeleterEEEPNS_14GraphCollectorEPNS_19CancellationManagerENS3_4SpanIS6_EERKNSD_INS_17ManagedStackTraceEEE + 372
       ...
       ...
        80  python                              0x00000001021f5ddc main + 56
        81  libdyld.dylib                       0x0000000182729430 start + 4
)
libc++abi: terminating with uncaught exception of type NSException

I have built a deep Q network using python. It can work on Windows, but I cannot run it on MacBook with M1. I had already checked the issue about matplotlib, it is not the case.

Hope to get a response ASAP, thanks in advance.

Replies  2
Boosts  0
Views  2.8k
Participants  3
Shakshi OP
Oct ’21

This is probably a bug in tensorflow-metal 0.2.0. The same error occurred in the environment where tensorflow-metal 0.2.0 was installed.

If you install tensorflow-metal 0.1.2 as shown below, the error will no longer occur.

pip install tensorflow-metal==0.1.2
0
Frameworks Engineer OP
Apple
Nov ’21

We are unable to get this issue reproduced. Please try updating to the latest version of the OS and let us know if this issue still persists, and provide a short script we can use to reproduce the issue.

0
libc++abi: terminating with uncaught exception of type NSException
First post date Last post date
 
 
Q