明明是有GPU的，但是报错：

W1121 07:43:47.606920 587 init.cc:128] Compiled with WITH_GPU, but no GPU found in runtime.

----------- Configuration Arguments -----------
batch_size: 10
checkpoints: checkpoints
data_dir: ./data
do_lower_case: True
epoch: 100
init_checkpoint: ./models/50
learning_rate: 0.01
max_seq_len: 128
output: ./output/score.txt
save_inference_model_path: ./models/inference_model
save_steps: 10000
skip_steps: 10
task_name: match_kn_gene
use_cuda: True
use_fast_executor: False
validation_steps: 1000
voc_size: 14373
vocab_path: ./dict/gene.dict
warmup_proportion: 0.1
weight_decay: 0.01
------------------------------------------------
Traceback (most recent call last):
File "predict.py", line 101, in
main(args)
File "predict.py", line 62, in main
place = fluid.CUDAPlace(0)
paddle.fluid.core_avx.EnforceNotMet: cudaGetDeviceCount failed in paddle::platform::GetCUDADeviceCount: no CUDA-capable device is detected at [/paddle/paddle/fluid/platform/gpu_info.cc:97]
PaddlePaddle Call Stacks:
0 0x7fb8d60eaad0p void paddle::platform::EnforceNotMet::Init(char const*, char const*, int) + 352
1 0x7fb8d60eae49p paddle::platform::EnforceNotMet::EnforceNotMet(std::__exception_ptr::exception_ptr, char const*, int) + 137
2 0x7fb8d810c45cp paddle::platform::GetCUDADeviceCount() + 1036
3 0x7fb8d60ca1c7p
4 0x7fb8d611d936p
5 0x7fb936abf973p PyObject_Call + 67
6 0x7fb936ace80dp
7 0x7fb936abf973p PyObject_Call + 67
8 0x7fb936b18764p
9 0x7fb936b1501bp
10 0x7fb936abf973p PyObject_Call + 67
11 0x7fb936b55d49p PyEval_EvalFrameEx + 15289
12 0x7fb936b5a17ep PyEval_EvalFrameEx + 32750
13 0x7fb936b5b6c9p PyEval_EvalCodeEx + 2025
14 0x7fb936b5b8eap PyEval_EvalCode + 26
15 0x7fb936b74a6dp
16 0x7fb936b75be8p PyRun_FileExFlags + 120
17 0x7fb936b76e08p PyRun_SimpleFileExFlags + 232
18 0x7fb936b8901cp Py_Main + 2972
19 0x7fb935d59830p __libc_start_main + 240
20 0x559a0e69507fp

Traceback (most recent call last):
File "./tools/extract_predict_utterance.py", line 68, in
main()
File "./tools/extract_predict_utterance.py", line 63, in main
sys.argv[3])
File "./tools/extract_predict_utterance.py", line 27, in extract_predict_utterance
score_list = [line.strip() for line in open(score_file, 'r')]
IOError: [Errno 2] No such file or directory: './output/score.txt'
Traceback (most recent call last):
File "./tools/eval.py", line 167, in
f1 = calc_f1(sents)
File "./tools/eval.py", line 151, in calc_f1
p = hit_char_total / pred_char_total
ZeroDivisionError: float division by zero