Created by: wshhja

I ran the infer.py as follow: python -u infer.py
--use_gru=True
--use_gpu=False
--share_rnn_weights=False
--infer_manifest='data/aishell/manifest.test'
--mean_std_path='mean_std.npz'
--vocab_path='vocab.txt'
--model_path='params.tar.gz'
--lang_model_path='zh_giga.no_cna_cmn.prune01244.klm' \

The params above are: --infer_manifest='data/aishell/manifest.test' \ --------created by run example/aishell/run_data.sh --mean_std_path='mean_std.npz' \ -------unzip from the speech model's compressed package(baidu_cn1.2k_model.tar.gz). --vocab_path='vocab.txt' \ -------unzip from the speech model's compressed package(baidu_cn1.2k_model.tar.gz). --model_path='params.tar.gz' \ -------unzip from the speech model's compressed package(baidu_cn1.2k_model.tar.gz). --lang_model_path='zh_giga.no_cna_cmn.prune01244.klm' ` -------the downloading language model

But it doesn't work and the console output remains as follow: ----------- Configuration Arguments ----------- alpha: 2.5 beam_size: 500 beta: 0.3 cutoff_prob: 1.0 cutoff_top_n: 40 decoding_method: ctc_beam_search error_rate_type: wer infer_manifest: data/aishell/manifest.test lang_model_path: zh_giga.no_cna_cmn.prune01244.klm mean_std_path: mean_std.npz model_path: params.tar.gz num_conv_layers: 2 num_proc_bsearch: 8 num_rnn_layers: 3 num_samples: 10 rnn_layer_size: 2048 share_rnn_weights: 0 specgram_type: linear trainer_count: 8 use_gpu: 0 use_gru: 1 vocab_path: vocab.txt

I0423 22:42:26.985620 154393 Util.cpp:166] commandline: --use_gpu=0 --rnn_use_batch=True --trainer_count=8

How can i fixed this problems?Thanks.

Created by: wshhja

My paddle version is: PaddlePaddle 0.11.0, compiled with with_avx: ON with_gpu: OFF with_mkl: OFF with_mkldnn: OFF with_double: OFF with_python: ON with_rdma: OFF with_timer: OFF

Created by: wshhja

Today,i run the infer.py via paddlepaddle-gpu(0.11.0).The script is:python -u infer.py
--use_gru=True
--use_gpu=True
--share_rnn_weights=False
--infer_manifest='data/aishell/manifest.test'
--mean_std_path='mean_std.npz'
--vocab_path='vocab.txt'
--model_path='baidu_cn1.2k_model.tar.gz'
--lang_model_path='zh_giga.no_cna_cmn.prune01244.klm' \

But the console output is the same as yesterday.

And,i want to know weather my model_path param is rigth. Thanks

Created by: AshishKarel

Maybe try reducing the batch size