作者发布的aishell 模型无法去restore,retrain
Created by: fangmingbnu
当我尝试去restore主页上的那个aishell模型的时候,出现错误。这是我的命令,CUDA_VISIBLE_DEVICES=0 python -u train.py --batch_size=16 --trainer_count=1 --num_passes=10 --num_proc_data=16 --num_conv_layers=2 --num_rnn_layers=3 --rnn_layer_size=1024 --num_iter_print=100 --learning_rate=4e-5 --max_duration=27.0 --min_duration=0.0 --test_off=False --use_sortagrad=True --use_gru=True --use_gpu=True --is_local=True --share_rnn_weights=False --train_manifest='data/aishell/manifest.train' --dev_manifest='data/aishell/manifest.dev' --mean_std_path='data/aishell/mean_std.npz' --vocab_path='data/aishell/vocab.txt' --output_model_dir='diymodel/aishell/' --augment_conf_path='conf/augmentation.config' --specgram_type='linear' --shuffle_method='batch_shuffle_clipped' --init_model_path='models/aishell/params.tar.gz'。
错误提示如下: ----------- Configuration Arguments ----------- augment_conf_path: conf/augmentation.config batch_size: 16 dev_manifest: data/aishell/manifest.dev init_model_path: models/aishell/params.tar.gz is_local: 1 learning_rate: 4e-05 max_duration: 27.0 mean_std_path: data/aishell/mean_std.npz min_duration: 0.0 num_conv_layers: 2 num_iter_print: 100 num_passes: 10 num_proc_data: 16 num_rnn_layers: 3 output_model_dir: diymodel/aishell/ rnn_layer_size: 1024 share_rnn_weights: 0 shuffle_method: batch_shuffle_clipped specgram_type: linear test_off: 0 train_manifest: data/aishell/manifest.train trainer_count: 1 use_gpu: 1 use_gru: 1 use_sortagrad: 1 vocab_path: data/aishell/vocab.txt
I0327 18:03:09.554020 23876 Util.cpp:166] commandline: --use_gpu=1 --rnn_use_batch=True --log_clipping=True --trainer_count=1 [INFO 2018-03-27 18:03:11,308 layers.py:2714] output for conv_0: c = 32, h = 81, w = 54, size = 139968 [INFO 2018-03-27 18:03:11,309 layers.py:3282] output for batch_norm_0: c = 32, h = 81, w = 54, size = 139968 [INFO 2018-03-27 18:03:11,310 layers.py:7454] output for scale_sub_region_0: c = 32, h = 81, w = 54, size = 139968 [INFO 2018-03-27 18:03:11,310 layers.py:2714] output for conv_1: c = 32, h = 41, w = 54, size = 70848 [INFO 2018-03-27 18:03:11,310 layers.py:3282] output for batch_norm_1: c = 32, h = 41, w = 54, size = 70848 [INFO 2018-03-27 18:03:11,311 layers.py:7454] output for scale_sub_region_1: c = 32, h = 41, w = 54, size = 70848 I0327 18:03:15.383965 23876 GradientMachine.cpp:94] Initing parameters.. I0327 18:03:17.461269 23876 GradientMachine.cpp:101] Init parameters done. F0327 18:03:20.693872 23876 TensorApply.h:126] Check failed: lhs_.getWidth() == rhs_.getWidth() (8876032 vs. 8804352) *** Check failure stack trace: *** @ 0x7fc5aa6570bd google::LogMessage::Fail() @ 0x7fc5aa659408 google::LogMessage::SendToLog() @ 0x7fc5aa656bcb google::LogMessage::Flush() @ 0x7fc5aa65a2de google::LogMessageFatal::~LogMessageFatal() @ 0x7fc5aa7b49c8 paddle::adamApply() @ 0x7fc5aa7a3b59 paddle::AdamParameterOptimizer::update() @ 0x7fc5aa7a3f82 paddle::OptimizerWithGradientClipping::update() @ 0x7fc5aa78fe1e paddle::SgdThreadUpdater::updateImpl() @ 0x7fc5aa62c7a1 ParameterUpdater::update() @ 0x7fc5aa0d5ff7 _wrap_ParameterUpdater_update @ 0x4bc3fa PyEval_EvalFrameEx @ 0x4c136f PyEval_EvalFrameEx @ 0x4b9ab6 PyEval_EvalCodeEx @ 0x4c16e7 PyEval_EvalFrameEx @ 0x4b9ab6 PyEval_EvalCodeEx @ 0x4c16e7 PyEval_EvalFrameEx @ 0x4b9ab6 PyEval_EvalCodeEx @ 0x4c1e6f PyEval_EvalFrameEx @ 0x4b9ab6 PyEval_EvalCodeEx @ 0x4c1e6f PyEval_EvalFrameEx @ 0x4b9ab6 PyEval_EvalCodeEx @ 0x4eb30f (unknown) @ 0x4e5422 PyRun_FileExFlags @ 0x4e3cd6 PyRun_SimpleFileExFlags @ 0x493ae2 Py_Main @ 0x7fc5e416e830 __libc_start_main @ 0x4933e9 _start @ (nil) (unknown) Aborted (core dumped) 不知道是不是什么参数设置错了,请教大家,谢谢!