update

嘉渊

2023-04-25 4b7dff9c7147c8ab8b66dedceee3d2b8ee485f10

 egs/aishell/conformer/run.sh

@@ -3,17 +3,17 @@
. ./path.sh || exit 1;

# machines configuration
CUDA_VISIBLE_DEVICES="0,1"
CUDA_VISIBLE_DEVICES="2,3"
gpu_num=2
count=1
gpu_inference=true  # Whether to perform gpu decoding, set false for cpu decoding
# for gpu decoding, inference_nj=ngpu*njob; for cpu decoding, inference_nj=njob
njob=8
njob=5
train_cmd=utils/run.pl
infer_cmd=utils/run.pl

# general configuration
feats_dir="../DATA" #feature output dictionary, for large data
feats_dir="/nfs/wangjiaming.wjm/Funasr_data/aishell-1-fix-cmvn" #feature output dictionary
exp_dir="."
lang=zh
dumpdir=dump/fbank
@@ -21,7 +21,7 @@
token_type=char
scp=feats.scp
type=kaldi_ark
stage=0
stage=3
stop_stage=4

# feature configuration
@@ -34,7 +34,7 @@
data_aishell=

# exp tag
tag=""
tag="exp1"

. utils/parse_options.sh || exit 1;

@@ -52,7 +52,7 @@
model_dir="baseline_$(basename "${asr_config}" .yaml)_${feats_type}_${lang}_${token_type}_${tag}"

inference_config=conf/decode_asr_transformer.yaml
inference_asr_model=valid.acc.ave_10best.pth
inference_asr_model=valid.acc.ave_10best.pb

# you can set gpu num for decoding here
gpuid_list=$CUDA_VISIBLE_DEVICES  # set gpus for decoding, the same as training stage by default
@@ -161,7 +161,8 @@
            rank=$i
            local_rank=$i
            gpu_id=$(echo $CUDA_VISIBLE_DEVICES | cut -d',' -f$[$i+1])
            asr_train.py \
            train.py \
                --task_name asr \
                --gpu_id $gpu_id \
                --use_preprocessor true \
                --token_type char \
@@ -177,7 +178,6 @@
                --resume true \
                --output_dir ${exp_dir}/exp/${model_dir} \
                --config $asr_config \
                --input_size $feats_dim \
                --ngpu $gpu_num \
                --num_worker_count $count \
                --multiprocessing_distributed true \