From 61597039b4e5f4b28ff1762f67d4a79f93f9c3b2 Mon Sep 17 00:00:00 2001
From: 语帆 <yf352572@alibaba-inc.com>
Date: 星期三, 28 二月 2024 19:39:56 +0800
Subject: [PATCH] test

---
 examples/industrial_data_pretraining/lcbnet/demo_nj.sh |   88 ++++++++++++++++++++++----------------------
 1 files changed, 44 insertions(+), 44 deletions(-)

diff --git a/examples/industrial_data_pretraining/lcbnet/demo_nj.sh b/examples/industrial_data_pretraining/lcbnet/demo_nj.sh
index 5e634c3..9d7755f 100755
--- a/examples/industrial_data_pretraining/lcbnet/demo_nj.sh
+++ b/examples/industrial_data_pretraining/lcbnet/demo_nj.sh
@@ -16,56 +16,56 @@
 _logdir="${inference_dir}/logdir"
 echo "inference_dir: ${inference_dir}"
 
-mkdir -p "${_logdir}"
-key_file1=${file_dir}/wav.scp
-key_file2=${file_dir}/ocr.txt
-split_scps1=
-split_scps2=
-for JOB in $(seq "${nj}"); do
-    split_scps1+=" ${_logdir}/wav.${JOB}.scp"
-    split_scps2+=" ${_logdir}/ocr.${JOB}.txt"
-done
-utils/split_scp.pl "${key_file1}" ${split_scps1}
-utils/split_scp.pl "${key_file2}" ${split_scps2}
+# mkdir -p "${_logdir}"
+# key_file1=${file_dir}/wav.scp
+# key_file2=${file_dir}/ocr.txt
+# split_scps1=
+# split_scps2=
+# for JOB in $(seq "${nj}"); do
+#     split_scps1+=" ${_logdir}/wav.${JOB}.scp"
+#     split_scps2+=" ${_logdir}/ocr.${JOB}.txt"
+# done
+# utils/split_scp.pl "${key_file1}" ${split_scps1}
+# utils/split_scp.pl "${key_file2}" ${split_scps2}
 
-gpuid_list_array=(${CUDA_VISIBLE_DEVICES//,/ })
-for JOB in $(seq ${nj}); do
-    {
-        id=$((JOB-1))
-        gpuid=${gpuid_list_array[$id]}
+# gpuid_list_array=(${CUDA_VISIBLE_DEVICES//,/ })
+# for JOB in $(seq ${nj}); do
+#     {
+#         id=$((JOB-1))
+#         gpuid=${gpuid_list_array[$id]}
 
-        export CUDA_VISIBLE_DEVICES=${gpuid}
+#         export CUDA_VISIBLE_DEVICES=${gpuid}
 
-        python -m funasr.bin.inference \
-        --config-path=${file_dir} \
-        --config-name="config.yaml" \
-        ++init_param=${file_dir}/model.pb \
-        ++tokenizer_conf.token_list=${file_dir}/tokens.txt \
-        ++input=[${_logdir}/wav.${JOB}.scp,${_logdir}/ocr.${JOB}.txt] \
-        +data_type='["kaldi_ark", "text"]' \
-        ++tokenizer_conf.bpemodel=${file_dir}/bpe.model \
-        ++output_dir="${inference_dir}/${JOB}" \
-        ++device="${inference_device}" \
-        ++ncpu=1 \
-        ++disable_log=true  &> ${_logdir}/log.${JOB}.txt
+#         python -m funasr.bin.inference \
+#         --config-path=${file_dir} \
+#         --config-name="config.yaml" \
+#         ++init_param=${file_dir}/model.pb \
+#         ++tokenizer_conf.token_list=${file_dir}/tokens.txt \
+#         ++input=[${_logdir}/wav.${JOB}.scp,${_logdir}/ocr.${JOB}.txt] \
+#         +data_type='["kaldi_ark", "text"]' \
+#         ++tokenizer_conf.bpemodel=${file_dir}/bpe.model \
+#         ++output_dir="${inference_dir}/${JOB}" \
+#         ++device="${inference_device}" \
+#         ++ncpu=1 \
+#         ++disable_log=true  &> ${_logdir}/log.${JOB}.txt
 
-    }&
-done
-wait
+#     }&
+# done
+# wait
 
 
-mkdir -p ${inference_dir}/1best_recog
-for f in token score text; do
-    if [ -f "${inference_dir}/${JOB}/1best_recog/${f}" ]; then
-        for JOB in $(seq "${nj}"); do
-            cat "${inference_dir}/${JOB}/1best_recog/${f}"
-        done | sort -k1 >"${inference_dir}/1best_recog/${f}"
-    fi
-done
+#mkdir -p ${inference_dir}/1best_recog
+
+if [ -f "${inference_dir}/${JOB}/1best_recog/token" ]; then
+    for JOB in $(seq "${nj}"); do
+        cat "${inference_dir}/${JOB}/1best_recog/token" >> "${inference_dir}/1best_recog/token"
+    done  
+fi
 
 echo "Computing WER ..."
 echo "Computing WER ..."
-python utils/postprocess_text_zh.py ${inference_dir}/1best_recog/text ${inference_dir}/1best_recog/text.proc
-python utils/postprocess_text_zh.py  ${data_dir}/text ${inference_dir}/1best_recog/text.ref
-python utils/compute_wer.py ${inference_dir}/1best_recog/text.ref ${inference_dir}/1best_recog/text.proc ${inference_dir}/1best_recog/text.cer
-tail -n 3 ${inference_dir}/1best_recog/text.cer
\ No newline at end of file
+#python utils/postprocess_text_zh.py ${inference_dir}/1best_recog/text ${inference_dir}/1best_recog/text.proc
+
+#cp  ${data_dir}/text ${inference_dir}/1best_recog/text.ref
+#python utils/compute_wer.py ${inference_dir}/1best_recog/text.ref ${inference_dir}/1best_recog/text.proc ${inference_dir}/1best_recog/text.cer
+#tail -n 3 ${inference_dir}/1best_recog/text.cer
\ No newline at end of file

--
Gitblit v1.9.1