OSUM-EChat / conf /data_s2s.yaml
xlgeng's picture
开始部署
841f290
# ===========================副语言 s2s thinking ===================================
# age gender,
age_gender_common:
path: /home/A02_tmpdata3/osum_s2s/gender/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 1511
weight: 2
gender_xianshi:
path: /home/A02_tmpdata3/osum_s2s/sex_xianshi_cosyvoice2_by_cywang_added_by_20250625/raw_data/s2s_handle/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 30
weight: 2
gender_yinshi_3k:
path: /home/A02_tmpdata3/osum_s2s/S2SChat_sex_yinshi_7_4_osum_by_cywang_added_by_20250708/raw_data/s2s_handle/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 3
weight: 2
gender_yinshi_5k:
path: /home/A02_tmpdata3/osum_s2s/S2SChat_sex_yinshi_5000_6_13_data_by_gjli_added_by_20250622/raw_data/s2s_handle/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 6
weight: 2
age_xianshi:
path: /home/A02_tmpdata3/osum_s2s/age_xianshi_cosyvoice2_by_cywang_added_by_20250625/raw_data/s2s_handle/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 25
weight: 2
# caption
caption_common_7label:
path: /home/A02_tmpdata3/osum_s2s/caption/raw_data/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 162
weight: 2
caption_common_50_label:
path: /home/A02_tmpdata3/osum_s2s/caption_add_2025_1_6/raw_data/s2s_data_with_gender/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 395 # 实际是196k
weight: 2
caption_xianshi:
path: /home/A02_tmpdata3/osum_s2s/caption_s2s_xianshi_20250806/raw_data/s2s_data/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 6
weight: 10
# emotion
emotion_100K_sensevoice:
path: /home/A02_tmpdata3/osum_s2s/emotion_yinshi_zxzhao_with_q_emo_by_cywang_added_by_20250701/handle_data/s2s_handle/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 107
weight: 10
emotion_30K_sensevoice:
path: /home/A02_tmpdata3/emotion/中英混多音色情感数据库/s2s_handle/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
tar_num: 33
weight: 10
S2SChat_osum_setting_qa_527_updated_by_cywang_added_by_20250616_think:
path: /home/A02_tmpdata3/osum_s2s/S2SChat_osum_setting_qa_527_updated_by_cywang_added_by_20250616/raw_data/s2s_handle/xlgeng_new_data/s2s_thinking/doubao/combines_list.txt
shard_num: 8
weight: 10
# ======================================s2s 副语言 thinking end=====================================
# ===========================副语言 s2s no thinking ===================================
# age gender,
age_gender_common_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/gender/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 1511
weight: 2
gender_xianshi_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/sex_xianshi_cosyvoice2_by_cywang_added_by_20250625/raw_data/s2s_handle/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 30
weight: 2
gender_yinshi_3k_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/S2SChat_sex_yinshi_7_4_osum_by_cywang_added_by_20250708/raw_data/s2s_handle/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 3
weight: 2
gender_yinshi_5k_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/S2SChat_sex_yinshi_5000_6_13_data_by_gjli_added_by_20250622/raw_data/s2s_handle/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 6
weight: 2
age_xianshi_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/age_xianshi_cosyvoice2_by_cywang_added_by_20250625/raw_data/s2s_handle/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 25
weight: 2
# caption
caption_common_7label_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/caption/raw_data/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 162
weight: 2
caption_common_50_label_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/caption_add_2025_1_6/raw_data/s2s_data_with_gender/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 395 # 实际是196k
weight: 2
caption_xianshi_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/caption_s2s_xianshi_20250806/raw_data/s2s_data/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 6
weight: 2
# emotion
emotion_100K_sensevoice_no_thinking:
path: /home/A02_tmpdata3/osum_s2s/emotion_yinshi_zxzhao_with_q_emo_by_cywang_added_by_20250701/handle_data/s2s_handle/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 107
weight: 10
emotion_30K_sensevoice_no_thinking:
path: /home/A02_tmpdata3/emotion/中英混多音色情感数据库/s2s_handle/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
tar_num: 33
weight: 10
S2SChat_osum_setting_qa_527_updated_by_cywang_added_by_20250616:
path: /home/A02_tmpdata3/osum_s2s/S2SChat_osum_setting_qa_527_updated_by_cywang_added_by_20250616/raw_data/s2s_handle/xlgeng_new_data/s2s_no_thinking/doubao/combines_list.txt
shard_num: 8
weight: 10
# -------------------------------------------s2s 副语言 no thinking end-------------------------------------------
S2SChat_syndata_merged_by_300W_zhguo_added_by_20250616:
path: /home/A02_tmpdata3/osum_s2s/S2SChat_syndata_merged_by_300W_zhguo_added_by_20250616/combines_data_s2s/combines_list.txt
tar_num: 3000
weight: 1
S2SChat_osum_total_data_lst_check_final_100W_by_zhguo_added_by_20250616:
path: /home/A02_tmpdata3/osum_s2s/S2SChat_osum_total_data_lst_check_final_100W_by_zhguo_added_by_20250616/combines_data_s2s/combines_list.txt
tar_num: 1000
weight: 1
gaozhiliang_gbma:
path: /home/A02_tmpdata3/osum_s2s/gaozhiliang_gbma/shards_list.txt
new_data_list: /home/node44_tmpdata3/netease/gbma/workspace/osum/data/process/0803/all_data_info.jsonl
new_lab_path: /home/work_nfs23/asr_data/data/osum_chat/s2s/gaozhiliang_gbma/shards_list.txt
shard_num: 24
weight: 1
# ======================================s2s no thinking end=====================================
# emotion explicit;
S2SChat_0628_E1_shard_by_kxxia_added_by_20250630:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_0628_E1_shard_by_kxxia_added_by_20250630/shards_list.txt
new_data_list: /home/work_nfs16/cywang/workspace/OSUM/E1/0628_E1_shard.jsonl
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/0628_E1_shard/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_0628_E1_shard_by_kxxia_added_by_20250630/shards_list.txt
shard_num: 147
description: "E1 shard, 情感显示数据"
weight: 1
S2SChat_eng_e1_by_cywang_added_by_20250711:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_eng_e1_by_cywang_added_by_20250711/shards_list.txt
new_data_list: /home/work_nfs16/kxxia/work/common/eng_e1.jsonl1752154262.3374825
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/eng_e1/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_eng_e1_by_cywang_added_by_20250711/shards_list.txt
shard_num: 50
weight: 2
# 下面一共才200多个
S2SChat_0630_trans_en2zh_by_cywang_added_by_20250704:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_0630_trans_en2zh_by_cywang_added_by_20250704/shards_list.txt
new_data_list: /home/work_nfs16/cywang/workspace/OSUM/trans_emotion/0630_trans_en2zh.jsonl
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/0630_trans_en2zh/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_0630_trans_en2zh_by_cywang_added_by_20250704/shards_list.txt
shard_num: 128
weight: 0.5
S2SChat_0630_trans_zh2en_by_cywang_added_by_20250704:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_0630_trans_zh2en_by_cywang_added_by_20250704/shards_list.txt
new_data_list: /home/work_nfs16/cywang/workspace/OSUM/trans_emotion/0630_trans_zh2en.jsonl
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/0630_trans_zh2en/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_0630_trans_zh2en_by_cywang_added_by_20250704/shards_list.txt
shard_num: 128
weight: 0.5
S2SChat_pachong_part1_filter_author_data_by_gjli_added_by_20250622:
shard_num: 28
huawei_path: /mnt/sfs/asr/update_data/S2SChat_pachong_part1_filter_author_data_by_gjli_added_by_20250622/shards_list.txt
new_data_list: /home/work_nfs16/gjli/workspaces/poem/6-16_shigepachong/pachong_part1_filter_content_data.list
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/pachong_part1_filter_author_data/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_pachong_part1_filter_author_data_by_gjli_added_by_20250622/shards_list.txt
weight: 1
S2SChat_pachong_part1_filter_content_data_by_gjli_added_by_20250622:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_pachong_part1_filter_content_data_by_gjli_added_by_20250622/shards_list.txt
new_data_list: /home/work_nfs16/gjli/workspaces/poem/6-16_shigepachong/pachong_part1_filter_author_data.list
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/pachong_part1_filter_content_data/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_pachong_part1_filter_content_data_by_gjli_added_by_20250622/shards_list.txt
shard_num: 68
weight: 1
S2SChat_poem_1_2_6_3_author_data_150num_by_gjli_added_by_20250622:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_poem_1_2_6_3_author_data_150num_by_gjli_added_by_20250622/shards_list.txt
new_data_list: /home/work_nfs16/gjli/workspaces/poem/6.3/poem_1_2_6-3_author_data.list
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/poem_1_2_6-3_author_data/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_poem_1_2_6_3_author_data_150num_by_gjli_added_by_20250622/shards_list.txt
shard_num: 2
weight: 1
S2SChat_poem_1_2_6_3_content_data_150num_by_gjli_added_by_20250622:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_poem_1_2_6_3_content_data_150num_by_gjli_added_by_20250622/shards_list.txt
new_data_list: /home/work_nfs16/gjli/workspaces/poem/6.3/poem_1_2_6-3_content_data.list
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/poem_1_2_6-3_content_data/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_poem_1_2_6_3_content_data_150num_by_gjli_added_by_20250622/shards_list.txt
shard_num: 9
weight: 1
S2SChat_poem_500_author_data_new_by_gjli_added_by_20250622:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_poem_500_author_data_new_by_gjli_added_by_20250622/shards_list.txt
new_data_list: /home/work_nfs16/gjli/workspaces/poem/poem_500_author_data_new.list
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/poem_500_author_data_new/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_poem_500_author_data_new_by_gjli_added_by_20250622/shards_list.txt
shard_num: 4
weight: 1
S2SChat_poem_500_content_data_new_by_gjli_added_by_20250622:
huawei_path: /mnt/sfs/asr/update_data/S2SChat_poem_500_content_data_new_by_gjli_added_by_20250622/shards_list.txt
new_data_list: /home/work_nfs16/gjli/workspaces/poem/poem_500_content_data_new.list
new_lab_path: /home/work_nfs11/cywang/data/shard/S2Chat/poem_500_content_data_new/shards_list.txt
path: /home/A02_tmpdata3/osum_s2s/S2SChat_poem_500_content_data_new_by_gjli_added_by_20250622/shards_list.txt
shard_num: 4
weight: 1