INVALID_ARGUMENT error in the Unsupervised GraphSAGE example

[hoangdzung]

I'm trying to run the Unsupervised GraphSAGE example.
The environment is python 3.8 , stellargraph 1.2.1, tensorflow 2.12.0.
However, I've got the following error

2023-05-04 11:32:00.921911: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'Placeholder/_0' with dtype int32
	 [[{{node Placeholder/_0}}]]
2023-05-04 11:32:02.759185: E tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:417] Loaded runtime CuDNN library: 8.2.4 but source was compiled with: 8.6.0.  CuDNN library needs to have matching major version and equal or higher minor version. If using a binary install, upgrade your CuDNN library.  If building from sources, make sure the library loaded at runtime is compatible with the version specified during compile configuration.
2023-05-04 11:32:02.782678: E tensorflow/compiler/xla/status_macros.cc:57] INTERNAL: RET_CHECK failure (tensorflow/compiler/xla/service/gpu/gpu_compiler.cc:618) dnn != nullptr 
*** Begin stack trace ***
	tsl::CurrentStackTrace[abi:cxx11]()
	
	xla::status_macros::MakeErrorStream::Impl::GetStatus()
	xla::gpu::GpuCompiler::OptimizeHloModule(xla::HloModule*, stream_executor::StreamExecutor*, stream_executor::DeviceMemoryAllocator*, xla::gpu::GpuTargetConfig const&, xla::AutotuneResults const*)
	xla::gpu::GpuCompiler::RunHloPasses(std::unique_ptr<xla::HloModule, std::default_delete<xla::HloModule> >, stream_executor::StreamExecutor*, xla::Compiler::CompileOptions const&)
	xla::Service::BuildExecutable(xla::HloModuleProto const&, std::unique_ptr<xla::HloModuleConfig, std::default_delete<xla::HloModuleConfig> >, xla::Backend*, stream_executor::StreamExecutor*, xla::Compiler::CompileOptions const&, bool)
	xla::LocalService::CompileExecutables(xla::XlaComputation const&, absl::lts_20220623::Span<xla::Shape const* const>, xla::ExecutableBuildOptions const&)
	xla::LocalClient::Compile(xla::XlaComputation const&, absl::lts_20220623::Span<xla::Shape const* const>, xla::ExecutableBuildOptions const&)
	tensorflow::XlaDeviceCompilerClient::BuildExecutable(tensorflow::XlaCompiler::Options const&, tensorflow::XlaCompilationResult const&)
	tensorflow::DeviceCompiler<xla::LocalExecutable, xla::LocalClient>::CompileStrict(tensorflow::DeviceCompilationClusterSignature const&, tensorflow::XlaCompiler::CompileOptions const&, tensorflow::XlaCompiler::Options const&, std::vector<tensorflow::XlaArgument, std::allocator<tensorflow::XlaArgument> > const&, tensorflow::NameAttrList const&, tensorflow::DeviceCompilationCache<xla::LocalExecutable>::Value, tensorflow::DeviceCompiler<xla::LocalExecutable, xla::LocalClient>::CompileScope, tensorflow::OpKernelContext*, tensorflow::DeviceCompilationProfiler*, tsl::mutex*)
	tensorflow::DeviceCompiler<xla::LocalExecutable, xla::LocalClient>::CompileImpl(tensorflow::XlaCompiler::CompileOptions const&, tensorflow::XlaCompiler::Options const&, tensorflow::NameAttrList const&, std::vector<tensorflow::XlaArgument, std::allocator<tensorflow::XlaArgument> > const&, tensorflow::DeviceCompiler<xla::LocalExecutable, xla::LocalClient>::CompileScope, tensorflow::DeviceCompileMode, tensorflow::OpKernelContext*, tensorflow::DeviceCompilationProfiler*, tensorflow::XlaCompilationResult const**, xla::LocalExecutable**)
	
	tensorflow::XlaLocalLaunchBase::ComputeAsync(tensorflow::OpKernelContext*, std::function<void ()>)
	tensorflow::BaseGPUDevice::ComputeAsync(tensorflow::AsyncOpKernel*, tensorflow::OpKernelContext*, std::function<void ()>)
	
	
	Eigen::ThreadPoolTempl<tsl::thread::EigenEnvironment>::WorkerLoop(int)
	std::_Function_handler<void (), tsl::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&)
	
	
	clone
*** End stack trace ***

Any idea how to fix it? Since I didn't change any code of the example notebook. I guess it might due to the versions.

[cloudmadeofcandy]

Anh ơi cái library này hình như chết 3 năm rồi, anh migrate sang PyTorch Geometric thui