运行infer.py报错IOError: [Errno 32] Broken pipe
Created by: windowxiaoming
----------- Configuration Arguments ----------- alpha: 2.6 beam_size: 300 beta: 5.0 cutoff_prob: 0.99 cutoff_top_n: 40 decoding_method: ctc_beam_search error_rate_type: cer infer_manifest: data/aishell/manifest.test lang_model_path: models/lm/zh_giga.no_cna_cmn.prune01244.klm mean_std_path: data/aishell/mean_std.npz model_path: checkpoints/aishell/params.latest.tar.gz num_conv_layers: 2 num_proc_bsearch: 8 num_rnn_layers: 3 num_samples: 10 rnn_layer_size: 512 share_rnn_weights: 0 specgram_type: linear trainer_count: 1 use_gpu: 0 use_gru: 1 vocab_path: data/aishell/vocab.txt
I0203 15:48:53.821841 16114 Util.cpp:166] commandline: --use_gpu=0 --rnn_use_batch=True --trainer_count=1 Process Process-2: Traceback (most recent call last): File "/home/hoge/anaconda2/envs/py27/lib/python2.7/multiprocessing/process.py", line 267, in _bootstrap self.run() File "/home/hoge/anaconda2/envs/py27/lib/python2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/storage/workspace/zhujieenv/DeepSpeech/data_utils/utility.py", line 120, in order_read_worker in_queue.put((order_id, sample)) File "", line 2, in put File "/home/hoge/anaconda2/envs/py27/lib/python2.7/multiprocessing/managers.py", line 759, in _callmethod kind, result = conn.recv() EOFError Process Process-3: Traceback (most recent call last): File "/home/hoge/anaconda2/envs/py27/lib/python2.7/multiprocessing/process.py", line 267, in _bootstrap self.run() File "/home/hoge/anaconda2/envs/py27/lib/python2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/storage/workspace/zhujieenv/DeepSpeech/data_utils/utility.py", line 134, in order_handle_worker while order_id != out_order[0]: File "", line 2, in getitem File "/home/hoge/anaconda2/envs/py27/lib/python2.7/multiprocessing/managers.py", line 758, in _callmethod conn.send((self._id, methodname, args, kwds)) IOError: [Errno 32] Broken pipe [INFO 2019-02-03 15:48:53,996 layers.py:2716] output for conv_0: c = 32, h = 81, w = 54, size = 139968 [INFO 2019-02-03 15:48:53,997 layers.py:3361] output for batch_norm_0: c = 32, h = 81, w = 54, size = 139968 [INFO 2019-02-03 15:48:53,997 layers.py:7533] output for scale_sub_region_0: c = 32, h = 81, w = 54, size = 139968 [INFO 2019-02-03 15:48:53,998 layers.py:2716] output for conv_1: c = 32, h = 41, w = 54, size = 70848 [INFO 2019-02-03 15:48:53,998 layers.py:3361] output for batch_norm_1: c = 32, h = 41, w = 54, size = 70848 [INFO 2019-02-03 15:48:53,999 layers.py:7533] output for scale_sub_region_1: c = 32, h = 41, w = 54, size = 70848 [INFO 2019-02-03 15:48:55,270 model.py:243] begin to initialize the external scorer for decoding [INFO 2019-02-03 15:48:55,341 model.py:253] language model: is_character_based = 1, max_order = 5, dict_size = 0 [INFO 2019-02-03 15:48:55,341 model.py:254] end initializing scorer [INFO 2019-02-03 15:48:55,341 infer.py:103] start inference ...
Target Transcription: 由于近期销售模式的调整 Output Transcription: 由于近期销售模式的调整 Current error rate [cer] = 0.000000
Target Transcription: 她表示最重要的就是诚恳 Output Transcription: 她表示虽重要的就是乘客 Current error rate [cer] = 0.272727
Target Transcription: 也助推了土地市场的火爆 Output Transcription: 也助推了土地市场的火爆 Current error rate [cer] = 0.000000
Target Transcription: 已经花费了一定的资源和成本 Output Transcription: 已经花费了一定的资源和成本 Current error rate [cer] = 0.000000
Target Transcription: 安徽铜陵结束了当地契税补贴政策 Output Transcription: 安徽同龄结束了当地契税补贴政策 Current error rate [cer] = 0.133333
Target Transcription: 研究者们希望年龄大的跑者能够注意脚踝的锻炼 Output Transcription: 研究者们希望也令大的跑者都能够注意脚踝的锻炼 Current error rate [cer] = 0.142857
Target Transcription: 但目前还存在服务架构不完善 Output Transcription: 但目前还存在服务价构不完善 Current error rate [cer] = 0.076923
Target Transcription: 可供客户为特定任务重新编程 Output Transcription: 可供客户为特定任务重新编程 Current error rate [cer] = 0.000000
Target Transcription: 脚踝的能力损失了大约百分之四十八 Output Transcription: 脚踝的能力损失了大约百分之四十八 Current error rate [cer] = 0.000000
Target Transcription: 稳增长措施需更全面地考虑化解楼市风险问题 Output Transcription: 为增长措施需更全面地考虑化解楼市风险问题 Current error rate [cer] = 0.050000 [INFO 2019-02-03 15:48:58,111 infer.py:124] finish inference