[Kaldi-users] Error of Karel's script

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

Hi everyone,

I have another error when running to Karel's script. The log is:

============================================================================
               DNN Hybrid Training & Decoding (Karel's recipe)
============================================================================
steps/nnet/make_fmllr_feats.sh --nj 10 --cmd run.pl --transform-dir
exp/tri3/decode_test data-fmllr-tri3/test data/test exp/tri3
data-fmllr-tri3/test/log data-fmllr-tri3/test/data
steps/nnet/make_fmllr_feats.sh: feature type is lda_fmllr
steps/nnet/make_fmllr_feats.sh: Done!, type lda_fmllr, data/test -->
data-fmllr-tri3/test, using : raw-trans None, gmm exp/tri3, trans
exp/tri3/decode_test
steps/nnet/make_fmllr_feats.sh --nj 10 --cmd run.pl --transform-dir
exp/tri3/decode_dev data-fmllr-tri3/dev data/dev exp/tri3
data-fmllr-tri3/dev/log data-fmllr-tri3/dev/data
steps/nnet/make_fmllr_feats.sh: feature type is lda_fmllr
steps/nnet/make_fmllr_feats.sh: Done!, type lda_fmllr, data/dev -->
data-fmllr-tri3/dev, using : raw-trans None, gmm exp/tri3, trans
exp/tri3/decode_dev
steps/nnet/make_fmllr_feats.sh --nj 10 --cmd run.pl --transform-dir
exp/tri3_ali data-fmllr-tri3/train data/train exp/tri3
data-fmllr-tri3/train/log data-fmllr-tri3/train/data
steps/nnet/make_fmllr_feats.sh: feature type is lda_fmllr
steps/nnet/make_fmllr_feats.sh: Done!, type lda_fmllr, data/train -->
data-fmllr-tri3/train, using : raw-trans None, gmm exp/tri3, trans
exp/tri3_ali
utils/subset_data_dir_tr_cv.sh data-fmllr-tri3/train
data-fmllr-tri3/train_tr90 data-fmllr-tri3/train_cv10
/home/ming/kaldi-trunk/egs/timit/s5/utils/subset_data_dir.sh: reducing #utt
from 3696 to 3320
/home/ming/kaldi-trunk/egs/timit/s5/utils/subset_data_dir.sh: reducing #utt
from 3696 to 376
exp/dnn4_pretrain-dbn/log/rbm.3.log:progress: [6.69102 4.79589 4.59248
4.52796 4.49425 4.47364 4.45425 4.43291 4.41321 4.39089 4.37391 4.36182
4.33871 4.32305 4.30841 4.29437 4.27835 4.26338 4.24451 4.23261 4.22594
4.22015 ]
exp/dnn4_pretrain-dbn/log/rbm.4.log:progress: [5.08679 3.75421 3.60121
3.55219 3.52727 3.51 3.49744 3.48335 3.47008 3.45369 3.44215 3.43454
3.42081 3.40939 3.40078 3.39252 3.38437 3.3727 3.35988 3.35331 3.35172
3.34959 ]
exp/dnn4_pretrain-dbn/log/rbm.5.log:progress: [4.87049 3.32658 3.14669
3.10413 3.08638 3.0761 3.0662 3.05728 3.04658 3.03451 3.02772 3.02178
3.01145 3.00338 2.99626 2.99111 2.98376 2.97654 2.96586 2.96138 2.96174
2.95995 ]
exp/dnn4_pretrain-dbn/log/rbm.6.log:progress: [3.65318 2.68876 2.56733
2.52673 2.50761 2.49697 2.49113 2.4853 2.47522 2.46409 2.45972 2.45737
2.44803 2.4431 2.43837 2.43415 2.42968 2.42373 2.41349 2.41221 2.41144
2.41134 ]

Pre-training finished.
Removing features tmpdir /tmp/tmp.XQL9AHFHQJ @ ming
train.ark
# Accounting: time=5612 threads=1
# Ended (code 0) at Tue Nov 11 05:18:32 MST 2014, elapsed time 5612 seconds
# steps/nnet/pretrain_dbn.sh --hid-dim 1024 --rbm-iter 20
data-fmllr-tri3/train exp/dnn4_pretrain-dbn
# Started at Tue Nov 11 11:52:50 MST 2014
#
steps/nnet/pretrain_dbn.sh --hid-dim 1024 --rbm-iter 20
data-fmllr-tri3/train exp/dnn4_pretrain-dbn
# INFO
steps/nnet/pretrain_dbn.sh : Pre-training Deep Belief Network as a stack of
RBMs
 dir       : exp/dnn4_pretrain-dbn
 Train-set : data-fmllr-tri3/train
steps/nnet/pretrain_dbn.sh Skipping, already have
exp/dnn4_pretrain-dbn/6.dbn
# Accounting: time=0 threads=1
# Ended (code 0) at Tue Nov 11 11:52:50 MST 2014, elapsed time 0 seconds
ITERATION 13: TRAIN AVG.LOSS 0.7196, (lrate1.5625e-05), CROSSVAL AVG.LOSS
1.4502, nnet accepted
(nnet_6.dbn_dnn_iter13_learnrate1.5625e-05_tr0.7196_cv1.4502)
finished, too small rel. improvement .0006753171
Succeeded training the Neural Network : exp/dnn4_pretrain-dbn_dnn/final.nnet
Preparing feature transform with CNN layers for RBM pre-training.
steps/nnet/train.sh successfuly finished.. exp/dnn4_pretrain-dbn_dnn
Removing features tmpdir /tmp/tmp.rUdewHleyG @ ming
cv.ark
train.ark
# Accounting: time=1015 threads=1
# Ended (code 0) at Tue Nov 11 05:35:27 MST 2014, elapsed time 1015 seconds
# steps/nnet/train.sh --feature-transform
exp/dnn4_pretrain-dbn/final.feature_transform --dbn
exp/dnn4_pretrain-dbn/6.dbn --hid-layers 0 --learn-rate 0.008
data-fmllr-tri3/train_tr90 data-fmllr-tri3/train_cv10 data/lang
exp/tri3_ali exp/tri3_ali exp/dnn4_pretrain-dbn_dnn
# Started at Tue Nov 11 11:52:50 MST 2014
#
steps/nnet/train.sh --feature-transform
exp/dnn4_pretrain-dbn/final.feature_transform --dbn
exp/dnn4_pretrain-dbn/6.dbn --hid-layers 0 --learn-rate 0.008
data-fmllr-tri3/train_tr90 data-fmllr-tri3/train_cv10 data/lang
exp/tri3_ali exp/tri3_ali exp/dnn4_pretrain-dbn_dnn

# INFO
steps/nnet/train.sh : Training Neural Network
 dir       : exp/dnn4_pretrain-dbn_dnn
 Train-set : data-fmllr-tri3/train_tr90 exp/tri3_ali
 CV-set    : data-fmllr-tri3/train_cv10 exp/tri3_ali

SKIPPING TRAINING... (steps/nnet/train.sh)
nnet already trained : exp/dnn4_pretrain-dbn_dnn/final.nnet
(nnet/nnet_6.dbn_dnn_iter13_learnrate1.5625e-05_tr0.7196_cv1.4502_final_)

# Accounting: time=0 threads=1
# Ended (code 0) at Tue Nov 11 11:52:50 MST 2014, elapsed time 0 seconds
steps/nnet/decode.sh --nj 20 --cmd run.pl --acwt 0.2 exp/tri3/graph
data-fmllr-tri3/test exp/dnn4_pretrain-dbn_dnn/decode_test
Warning: run.pl ignoring options "-pe smp 2 "
steps/nnet/decode.sh --nj 20 --cmd run.pl --acwt 0.2 exp/tri3/graph
data-fmllr-tri3/dev exp/dnn4_pretrain-dbn_dnn/decode_dev
Warning: run.pl ignoring options "-pe smp 2 "
steps/nnet/align.sh --nj 20 --cmd run.pl data-fmllr-tri3/train data/lang
exp/dnn4_pretrain-dbn_dnn exp/dnn4_pretrain-dbn_dnn_ali
steps/nnet/align.sh: aligning data 'data-fmllr-tri3/train' using nnet/model
'exp/dnn4_pretrain-dbn_dnn', putting alignments in
'exp/dnn4_pretrain-dbn_dnn_ali'
steps/nnet/align.sh: done aligning data.
steps/nnet/make_denlats.sh --nj 20 --cmd run.pl --acwt 0.2 --lattice-beam
10.0 --beam 18.0 data-fmllr-tri3/train data/lang exp/dnn4_pretrain-dbn_dnn
exp/dnn4_pretrain-dbn_dnn_denlats
Making unigram grammar FST in exp/dnn4_pretrain-dbn_dnn_denlats/lang
Compiling decoding graph in exp/dnn4_pretrain-dbn_dnn_denlats/dengraph
Graph exp/dnn4_pretrain-dbn_dnn_denlats/dengraph/HCLG.fst already exists:
skipping graph creation.
steps/nnet/make_denlats.sh: generating denlats from data
'data-fmllr-tri3/train', putting lattices in
'exp/dnn4_pretrain-dbn_dnn_denlats'
Warning: run.pl ignoring options "-pe smp 2 "
steps/nnet/make_denlats.sh: done generating denominator lattices.
steps/nnet/train_mpe.sh --cmd run.pl --num-iters 6 --acwt 0.2 --do-smbr
true --use-silphones true data-fmllr-tri3/train data/lang
exp/dnn4_pretrain-dbn_dnn exp/dnn4_pretrain-dbn_dnn_ali
exp/dnn4_pretrain-dbn_dnn_denlats exp/dnn4_pretrain-dbn_dnn_smbr
steps/nnet/train_mpe.sh: invalid option --use-silphones

Then the script just exited without exiting error. I checked train_mpe.sh
and found nothing about the option "--use-silphones".

Now gcc is 4.9.1 because of ubuntu's update. I changed host_config.h of
CUDA driver so there is no error of "Unsupported gcc". Thanks.

-- 
Graduate Research Assistant
Signal Analysis, Representation and Perception Lab
Arizona State University, USA