Created by: suntingfeng

你好，我在进行模型训练的时候出现了以下的错误，请问这是什么原因引起的呢？该如何解决呢？谢谢！！！ ----------- Configuration Arguments ----------- augment_conf_path: conf/augmentation.configbatch_size: 256dev_manifest: ../test_manifest/manifest.dev init_model_path: ../resource/params.tar.gz is_local: True learning_rate: 0.0005 max_duration: 27.0 mean_std_path: ../resource/mean_std.npz min_duration: 0.0 num_conv_layers: 2 num_iter_print: 100 num_passes: 200 num_proc_data: 16 num_rnn_layers: 3 output_model_dir: ../output rnn_layer_size: 2048 share_rnn_weights: True shuffle_method: batch_shuffle_clipped specgram_type: linear test_off: False train_manifest: ../test_manifest/manifest.train trainer_count: 1 use_gpu: True use_gru: False use_sortagrad: True vocab_path: ../resource/vocab.txt

I1014 19:49:58.496965 127202 Util.cpp:166] commandline: --use_gpu=True --rnn_use_batch=True --log_clipping=True --trainer_count=1 [INFO 2019-10-14 19:50:00,083 layers.py:2716] output for conv_0: c = 32, h = 81, w = 54, size = 139968 [INFO 2019-10-14 19:50:00,085 layers.py:3361] output for batch_norm_0: c = 32, h = 81, w = 54, size = 139968 [INFO 2019-10-14 19:50:00,086 layers.py:7533] output for scale_sub_region_0: c = 32, h = 81, w = 54, size =139968 [INFO 2019-10-14 19:50:00,088 layers.py:2716] output for conv_1: c = 32, h = 41, w = 54, size = 70848 [INFO 2019-10-14 19:50:00,089 layers.py:3361] output for batch_norm_1: c = 32, h = 41, w = 54, size = 70848 [INFO 2019-10-14 19:50:00,090 layers.py:7533] output for scale_sub_region_1: c = 32, h = 41, w = 54, size =70848 I1014 19:50:21.294397 127202 GradientMachine.cpp:94] Initing parameters.. I1014 19:50:25.766219 127202 GradientMachine.cpp:101] Init parameters done. /usr/lib64/python2.7/site-packages/resampy/core.py:90: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type. if not np.issubdtype(x.dtype, np.float): /usr/lib64/python2.7/site-packages/resampy/core.py:90: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type. if not np.issubdtype(x.dtype, np.float): /usr/lib64/python2.7/site-packages/resampy/core.py:90: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type. if not np.issubdtype(x.dtype, np.float):

------- Time: 1 sec, Pass: 0, ValidationCost: 2138.80664062 Traceback (most recent call last): File "train.py", line 133, in main() File "train.py", line 129, in main train() File "train.py", line 120, in train test_off=args.test_off) File "/home/rec/local_disk/template/stf/items/deepspeech2/src/model.py", line 156, in train feeding=adapted_feeding_dict) File "/usr/lib64/python2.7/site-packages/paddle/v2/trainer.py", line 214, in train gm=self.gradient_machine)) File "/home/rec/local_disk/template/stf/items/deepspeech2/src/model.py", line 149, in event_handler trainer.save_parameter_to_tar(f) File "/usr/lib64/python2.7/site-packages/paddle/v2/trainer.py", line 134, in save_parameter_to_tar self.parameters.to_tar(f) File "/usr/lib64/python2.7/site-packages/paddle/v2/parameters.py", line 343, in to_tar self.serialize(nm, buf) File "/usr/lib64/python2.7/site-packages/paddle/v2/parameters.py", line 304, in serialize param = self.get(name) File "/usr/lib64/python2.7/site-packages/paddle/v2/parameters.py", line 242, in get return self.getitem(key=parameter_name) File "/usr/lib64/python2.7/site-packages/paddle/v2/parameters.py", line 186, in getitem return self.__getter_inner(key, api.PARAMETER_VALUE) File "/usr/lib64/python2.7/site-packages/paddle/v2/parameters.py", line 164, in __getter_inner each_gradient_machine, key) File "/usr/lib64/python2.7/site-packages/paddle/v2/parameters.py", line 418, in get_parameter_in_gradient_machine raise ValueError("No such parameter") ValueError: No such parameter