[TTS] TextExecutor().text_punc(text="xx")异常退出

[liuyang77886]

没有任何错误输出，直接退出，gpu有2G，只用到了几百兆
win11，NVIDIA Quadro M620， device: 0, GPU Compute Capability: 5.0, Driver API Version: 11.7, Runtime API Version: 11.7，cuDNN Version: 8.5，cuda-toolkit-cuda_11.7.1_516.94_windows，cuda-driver-517.40-quadro-rtx-desktop-notebook-win10-win11-64bit-international-dch-whql，nvidia cuda 11.7.102 driver
paddle-bfloat 0.1.7
paddle2onnx 1.0.1
paddleclas 2.4.3
paddlefsl 1.1.0
paddlehub 2.3.0
paddlenlp 2.4.0
paddleocr 2.6
paddlepaddle-gpu 2.4.0rc0.post117
paddleslim 2.2.1
paddlespeech 1.1.3
paddlespeech-feat 0.1.0
paddlex 2.1.0

(paddle) G:\work_py\notebook\paddle\PaddleSpeech>python
Python 3.7.12 | packaged by conda-forge | (default, Oct 26 2021, 05:35:01) [MSC v.1916 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.

from paddlespeech.cli.text.infer import TextExecutor
text_punc = TextExecutor()
result = text_punc(text="今天的天气真不错啊你下午有空吗我想约你一起去吃饭")
E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\image_utils.py:213: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
resample=Image.BILINEAR,
E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\image_utils.py:379: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.
resample=Image.NEAREST,
E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\ernie_vil\feature_extraction.py:65: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.
resample=Image.BICUBIC,
E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\clip\feature_extraction.py:64: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.
resample=Image.BICUBIC,
W1008 20:29:08.600579 3124 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 5.0, Driver API Version: 11.7, Runtime API Version: 11.7
W1008 20:29:08.613807 3124 gpu_resources.cc:91] device: 0, cuDNN Version: 8.5.
(paddle) G:\work_py\notebook\paddle\PaddleSpeech>

[yt605155624]

2G 应该是不够的
参考 #2387 修改 log 等级获取更多信息，此外，可在代码

PaddleSpeech/paddlespeech/cli/text/infer.py

Line 308 in 0359c3f

def __call__(

中 print 更多信息进行调试

[liuyang77886]

我更换版本了 paddlepaddle-gpu 2.3.2.post116
现在报其他错额

File "test.py", line 13, in
device=paddle.get_device())
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlespeech\cli\utils.py", line 328, in _warpper
return executor_func(self, *args, **kwargs)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlespeech\cli\text\infer.py", line 259, in call
self._init_from_path(task, model, lang, config, ckpt_path, punc_vocab)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlespeech\cli\text\infer.py", line 135, in _init_from_path
cfg_path=self.cfg_path, ckpt_path=self.ckpt_path)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlespeech\text\models\ernie_linear\ernie_linear.py", line 40, in init
os.path.dirname(cfg_path))
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\model_utils.py", line 348, in from_pretrained
base_model = cls.base_model_class(*base_args, **base_kwargs)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\utils.py", line 159, in impl
init_func(self, *args, **kwargs)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\ernie\modeling.py", line 846, in init
task_id, use_task_id)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\ernie\modeling.py", line 62, in init
weight_attr=weight_attr)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\nn\layer\common.py", line 1457, in init
is_bias=False)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\fluid\dygraph\layers.py", line 424, in create_parameter
default_initializer)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\fluid\layer_helper_base.py", line 381, in create_parameter
**attr._to_kwargs(with_initializer=True))
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\fluid\framework.py", line 3572, in create_parameter
initializer(param, self)
File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\fluid\initializer.py", line 474, in call
'std', self._std_dev, 'seed', self._seed)
SystemError: (Fatal) Operator truncated_gaussian_random raises an class thrust::system::system_error exception.
The exception content is
:parallel_for failed: cudaErrorNoKernelImageForDevice: no kernel image is available for execution on the device. (at ..\paddle\fluid\imperative\tracer.cc:307)

[liuyang77886]

2G 应该是不够的 参考 #2387 修改 log 等级获取更多信息，此外，可在代码

PaddleSpeech/paddlespeech/cli/text/infer.py

Line 308 in 0359c3f

def __call__(

中 print 更多信息进行调试

2G够了，我以前都能运行哦

[yt605155624]

cudaErrorNoKernelImageForDevice 应该是 paddle 的 cuda 版本和你机器的 cuda 版本不一致导致的（你自己机器的 cuda 有达到 11.6 么），在 paddle 的 issue 区可以搜到

[liuyang77886]

driver最新版，CUDA 10.2.89 >= 440.33 >= 441.22，cudnn-》for CUDA

[abcdbosh]

我更换版本了 paddlepaddle-gpu 2.3.2.post116 现在报其他错额

File "test.py", line 13, in device=paddle.get_device()) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlespeech\cli\utils.py", line 328, in _warpper return executor_func(self, *args, **kwargs) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlespeech\cli\text\infer.py", line 259, in call self._init_from_path(task, model, lang, config, ckpt_path, punc_vocab) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlespeech\cli\text\infer.py", line 135, in _init_from_path cfg_path=self.cfg_path, ckpt_path=self.ckpt_path) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlespeech\text\models\ernie_linear\ernie_linear.py", line 40, in init os.path.dirname(cfg_path)) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\model_utils.py", line 348, in from_pretrained base_model = cls.base_model_class(*base_args, **base_kwargs) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\utils.py", line 159, in impl init_func(self, *args, **kwargs) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\ernie\modeling.py", line 846, in init task_id, use_task_id) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddlenlp\transformers\ernie\modeling.py", line 62, in init weight_attr=weight_attr) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\nn\layer\common.py", line 1457, in init is_bias=False) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\fluid\dygraph\layers.py", line 424, in create_parameter default_initializer) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\fluid\layer_helper_base.py", line 381, in create_parameter **attr._to_kwargs(with_initializer=True)) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\fluid\framework.py", line 3572, in create_parameter initializer(param, self) File "E:\pfile\anaconda3\envs\paddle\lib\site-packages\paddle\fluid\initializer.py", line 474, in call 'std', self._std_dev, 'seed', self._seed) SystemError: (Fatal) Operator truncated_gaussian_random raises an class thrust::system::system_error exception. The exception content is :parallel_for failed: cudaErrorNoKernelImageForDevice: no kernel image is available for execution on the device. (at ..\paddle\fluid\imperative\tracer.cc:307)

您好，后续是如何解决的呢