zh794390558
commented
2 years ago 看log是数据类型错误，麻烦打印下输入的数据类型并查看paddle api 文档，看下是否一致。
Closed Doubledongli closed 2 years ago
zh794390558
commented
2 years ago 看log是数据类型错误，麻烦打印下输入的数据类型并查看paddle api 文档，看下是否一致。
Doubledongli
commented
2 years ago 打印下输入的数据类型并查看paddle api 文档，看下是否一致
数据类型默认是int32，并且完全一样的代码，我在ubuntu上面运行是正常的，win10上就报错，并且我通过paddle.to_tensor转为int64以后，仍旧会报其他的错误，错误类型如下：
File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 915, in _dygraph_call_func outputs = self.forward(*inputs, **kwargs) File "D:\conda\envs\asr\lib\site-packages\paddle\nn\layer\loss.py", line 1130, in forward norm_by_times=norm_by_times) File "D:\conda\envs\asr\lib\site-packages\paddle\nn\functional\loss.py", line 1126, in ctc_loss input_lengths, label_lengths) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\layers\loss.py", line 613, in warpctc norm_by_times, ) RuntimeError: (NotFound) There are no kernels which are registered in the warpctc operator. [Hint: Expected kernels_iter != all_op_kernels.end(), but received kernels_iter == all_op_kernels.end().] (at C:\home\workspace\Paddle_release\paddle\fluid\imperative\prepared_operator.cc:327) [operator < warpctc > error]
个人怀疑，这是一个系统底层数据转换的bug。
zh794390558
commented
2 years ago 
确定是和这个api类型一致？确定的话有可能是win上的bug了。可以在paddle那边儿问问看。
Doubledongli
commented
2 years ago 话有可能是win上的bug了。可以在paddle那边儿问问看。
已经找到原因了，windows无法安装pip的ppspeech的ctc_decoders，这个只有mac和linux版本，所以导致底层调用失败，啥时候paddle那边能出windows版本啊，自己想编译一直出问题
yt605155624
commented
2 years ago
2022-07-05 11:07:06.922 | INFO | paddlespeech.s2t.training.trainer:do_train:271 - Train Total Examples: 6245 Traceback (most recent call last): File "D:\develop\workspace\PaddleSpeech\paddlespeech\s2t\training\trainer.py", line 342, in run self.do_train() File "D:\develop\workspace\PaddleSpeech\paddlespeech\s2t\training\trainer.py", line 307, in do_train raise e File "D:\develop\workspace\PaddleSpeech\paddlespeech\s2t\training\trainer.py", line 286, in do_train self.train_batch(batch_index, batch, msg) File "D:\develop\workspace\PaddleSpeech\paddlespeech\s2t\exps\deepspeech2\model.py", line 55, in train_batch loss = self.model(audio, audio_len, text, text_len) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 930, in call return self._dygraph_call_func(*inputs, kwargs) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 915, in _dygraph_call_func outputs = self.forward(*inputs, *kwargs) File "D:\develop\workspace\PaddleSpeech\paddlespeech\s2t\models\ds2\deepspeech2.py", line 292, in forward loss = self.decoder(eouts, eouts_len, text, text_len) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 930, in call return self._dygraph_call_func(inputs, kwargs) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 915, in _dygraph_call_func 2022-07-05 11:07:18.467 | ERROR | paddlespeech.s2t.training.trainer:do_train:306 - (InvalidArgument) The type of data we are trying to retrieve does not match the type of data currently contained in the container. (at C:\home\workspace\Paddle_release\paddle\phi\core\dense_tensor.cc:148) [operator < warpctc > error] 2022-07-05 11:07:18.468 | INFO | paddlespeech.s2t.training.timer:exit:44 - Epoch-Train Time Cost: 0:00:11.545535 2022-07-05 11:07:18.468 | INFO | paddlespeech.s2t.training.timer:exit:44 - Training Done: 0:00:12.690650 outputs = self.forward(*inputs, kwargs) File "D:\develop\workspace\PaddleSpeech\paddlespeech\s2t\modules\ctc.py", line 95, in forward loss = self.criterion(logits, ys_pad, hlens, ys_lens) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 930, in call return self._dygraph_call_func(*inputs, *kwargs) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 915, in _dygraph_call_func outputs = self.forward(inputs, kwargs) File "D:\develop\workspace\PaddleSpeech\paddlespeech\s2t\modules\loss.py", line 93, in forward loss = self.loss(logits, ys_pad, hlens, ys_lens, self._kwargs) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 930, in call return self._dygraph_call_func(*inputs, *kwargs) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\dygraph\layers.py", line 915, in _dygraph_call_func outputs = self.forward(inputs, kwargs) File "D:\conda\envs\asr\lib\site-packages\paddle\nn\layer\loss.py", line 1130, in forward norm_by_times=norm_by_times) File "D:\conda\envs\asr\lib\site-packages\paddle\nn\functional\loss.py", line 1126, in ctc_loss input_lengths, label_lengths) File "D:\conda\envs\asr\lib\site-packages\paddle\fluid\layers\loss.py", line 613, in warpctc norm_by_times, ) ValueError: (InvalidArgument) The type of data we are trying to retrieve does not match the type of data currently contained in the container. (at C:\home\workspace\Paddle_release\paddle\phi\core\dense_tensor.cc:148) [operator < warpctc > error]