环境安装成功，但是不懂为什么运行demo报错，求解决。

ssh://kailing@10.134.35.150:22/home/kailing/anaconda3/envs/kailing3.7/bin/python3 -u /home/kailing/qiu/ABSA/Senta-master/demo.py
ModuleNotFoundError: No module named 'numpy.core._multiarray_umath'
ModuleNotFoundError: No module named 'numpy.core._multiarray_umath'
W0704 17:25:20.889032 8592 device_context.cc:236] Please NOTE: device: 0, CUDA Capability: 75, Driver API Version: 11.2, Runtime API Version: 10.0
W0704 17:25:20.892278 8592 device_context.cc:244] device: 0, cuDNN Version: 7.6.
Traceback (most recent call last):
File "/home/kailing/qiu/ABSA/Senta-master/demo.py", line 24, in
result = my_senta.predict(texts, aspects)
File "/home/kailing/qiu/ABSA/Senta-master/senta/train.py", line 271, in predict
result = self.inference.run(inputs)
paddle.fluid.core_avx.EnforceNotMet:

--------------------------------------------
C++ Call Stacks (More useful to developers):
--------------------------------------------
0 std::string paddle::platform::GetTraceBackString(char const*&&, char const*, int)
1 paddle::platform::EnforceNotMet::EnforceNotMet(std::__exception_ptr::exception_ptr, char const*, int)
2 void paddle::operators::math::Blas::MatMul(paddle::framework::Tensor const&, paddle::operators::math::MatDescriptor const&, paddle::framework::Tensor const&, paddle::operators::math::MatDescriptor const&, float, paddle::framework::Tensor*, float) const
3 paddle::operators::MatMulKernel::Compute(paddle::framework::ExecutionContext const&) const
4 std::_Function_handler, paddle::operators::MatMulKernel, paddle::operators::MatMulKernel >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1}>::_M_invoke(std::_Any_data const&, paddle::framework::ExecutionContext const&)
5 paddle::framework::OperatorWithKernel::RunImpl(paddle::framework::Scope const&, paddle::platform::Place const&, paddle::framework::RuntimeContext*) const
6 paddle::framework::OperatorWithKernel::RunImpl(paddle::framework::Scope const&, paddle::platform::Place const&) const
7 paddle::framework::OperatorBase::Run(paddle::framework::Scope const&, paddle::platform::Place const&)
8 paddle::framework::Executor::RunPreparedContext(paddle::framework::ExecutorPrepareContext*, paddle::framework::Scope*, bool, bool, bool)
9 paddle::NativePaddlePredictor::Run(std::vector > const&, std::vector >*, int)

------------------------------------------
Python Call Stacks (More useful to users):
------------------------------------------
File "/root/gaocan01/0-bin/4-paddle_1.6.3_py3.7/miniconda3/lib/python3.7/site-packages/paddle/fluid/framework.py", line 2488, in append_op
attrs=kwargs.get("attrs", None))
File "/root/gaocan01/0-bin/4-paddle_1.6.3_py3.7/miniconda3/lib/python3.7/site-packages/paddle/fluid/layer_helper.py", line 43, in append_op
return self.main_program.current_block().append_op(*args, **kwargs)
File "/root/gaocan01/0-bin/4-paddle_1.6.3_py3.7/miniconda3/lib/python3.7/site-packages/paddle/fluid/layers/nn.py", line 7072, in matmul
'alpha': float(alpha),
File "/mnt/du/gaocan01/1-textone/senta/senta/modules/ernie.py", line 152, in _build_model
x=input_mask, y=input_mask, transpose_y=True)
File "/mnt/du/gaocan01/1-textone/senta/senta/modules/ernie.py", line 85, in __init__
input_mask)
File "/mnt/du/gaocan01/1-textone/senta/senta/models/ernie_two_sent_classification_ch.py", line 132, in make_embedding
use_fp16=use_fp16
File "/mnt/du/gaocan01/1-textone/senta/senta/models/ernie_two_sent_classification_ch.py", line 51, in forward
emb_dict = self.make_embedding(fields_dict, phase)
File "/mnt/du/gaocan01/1-textone/senta/senta/training/base_trainer.py", line 192, in init_save_inference_net
forward_output_dict = self.model_class.forward(fields_dict, phase=InstanceName.SAVE_INFERENCE)
File "/mnt/du/gaocan01/1-textone/senta/senta/training/base_trainer.py", line 141, in init_net
self.init_save_inference_net()
File "/mnt/du/gaocan01/1-textone/senta/senta/training/base_trainer.py", line 48, in __init__
self.init_net()
File "/mnt/du/gaocan01/1-textone/senta/senta/training/custom_trainer.py", line 31, in __init__
BaseTrainer.__init__(self, params, data_set_reader, model_class)
File "./train.py", line 71, in build_trainer
trainer = trainer_class(params=params_dict, data_set_reader=dataset_reader, model_class=model)
File "./train.py", line 91, in
trainer = build_trainer(trainer_params_dict, dataset_reader, model, num_train_examples)

----------------------
Error Message Summary:
----------------------
Error: Paddle internal Check failed. (Please help us create a new issue, here we need to find the developer to add a user friendly error message)
[CUBLAS: execution failed.] at (/paddle/paddle/fluid/operators/math/blas_impl.cu.h:51)
[operator < matmul > error]

Process finished with exit code 1

Start with demo.py line 24 and follow the predict call into senta/train.py line 271. Run the demo in the reported environment, then investigate the NumPy import failures and the Paddle matmul/CUBLAS error; done means the demo completes and returns a prediction without these errors.

Written by the indexing model from the issue text.