issues
search
TsinghuaAI
/
CPM-2-Finetune
Finetune CPM-2
MIT License
83
stars
21
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Cpm2 open
#50
Suleymanlizad
opened
2 months ago
0
change mp size 后,训练会出现 size missmatch 的错误
#49
samulew
opened
1 year ago
2
训好的模型如何转化成huggingface的模型格式呢
#48
Tron1994
opened
1 year ago
0
官方提供预训练模型参数是4个模型并行的文件,这限定模型并行必须是4?
#47
Tron1994
opened
1 year ago
1
加载100000模型,_load_zero_checkpoint失败,提示没有相关zero_pp_rank*文件
#46
Tron1994
opened
1 year ago
1
CPM2Datasets.py 中的T5Dataset报错
#45
dingtine
closed
1 year ago
1
在A100上加载FusedAdam报错
#44
giter000
opened
1 year ago
1
CPM2的文本生成example怎么没有,prompt训练完了之后不知道咋推理
#43
touwenameng
opened
2 years ago
1
Docker DeepSpeed error: ssh: Could not resolve hostname node0: Name or service not known
#42
sebastian-nehrdich
closed
1 year ago
4
promt adgen文件缺失
#41
zhu1090093659
closed
1 year ago
3
内部做了古诗翻译和菜谱生成的demo可以提供数据和demo吗?
#40
jiangliqin
closed
1 year ago
0
CPM2如何做few-shot的文本生成任务
#39
zhihao-chen
closed
2 years ago
0
请教:使用中英文双语模型报了一下的错误:
#38
Chunhui-Zou
closed
2 years ago
1
请教:模型在跑prompt的的脚本时,并没有用到test的数据,是为什么呢?还有prompt训练好的模型参数保存在哪里?
#37
Chunhui-Zou
closed
2 years ago
5
救助:模型支持的最长输入序列是多少
#36
Chunhui-Zou
closed
2 years ago
2
想看模型生成的结果,该修改代码那一块
#35
Chunhui-Zou
closed
2 years ago
0
Math23K 没有公开test_private.json 文件吗?
#34
XiaoqingNLP
closed
2 years ago
3
RuntimeError: Unable to proceed, no GPU resources available
#33
louxingrui
opened
2 years ago
2
数据集怎么处理,我下载了LSCTS数据集,运行程序后报错。
#32
Chunhui-Zou
closed
2 years ago
2
Create finetune_cpm2_sogou-log.sh
#31
xcjthu
closed
2 years ago
0
CPM2模型推理代码
#30
Bournet
closed
2 years ago
1
CPM2在生成任务上的微调策略
#29
XiaoqingNLP
closed
2 years ago
3
怎么加入新词再finetune
#28
LinglingGreat
closed
2 years ago
3
A100-8卡环境cublas报错
#27
linjianz
closed
2 years ago
1
用deepspeed工具,将cpm2.0的pt模型文件转化为fp32_state_dict失败
#26
linjianz
closed
2 years ago
0
How to use BMInf to inference 100000.tar 11B model?
#25
linjianz
closed
2 years ago
1
模型并行度修改的切割问题
#24
leelinglin
closed
1 year ago
1
模型fine-tune显存溢出
#23
leelinglin
closed
2 years ago
4
how to use 32000.tar?
#22
AdamBear
closed
1 year ago
0
双机8卡分布式训练
#21
forrestbing
closed
2 years ago
2
the decoder input in evaluate_gen()
#20
GMago-LeWay
closed
2 years ago
1
显存占用
#19
2020zyc
closed
1 year ago
2
请问可以不用deepspeed吗
#18
2020zyc
closed
2 years ago
7
bug? save_zero
#17
2020zyc
closed
3 years ago
1
attention.dense.weight not found when prompt fine tuning
#16
luotongml
closed
2 years ago
17
请问2张A100-40G能跑吗
#15
2020zyc
closed
3 years ago
11
prompt tunning问题
#14
zirui
closed
3 years ago
4
请问sentinel id的作用什么?
#13
wakafengfan
closed
3 years ago
1
Can CPM-2 run in playground model, any prompt hint?
#12
qhduan
closed
3 years ago
1
Finetune loss and acc is pool
#11
k15201363625
closed
3 years ago
10
docker 运行失败
#10
windflee
closed
3 years ago
1
MoE Finetune
#9
lizy14
closed
3 years ago
2
用两个机器,启动的时候,报错
#8
lonelydancer
closed
3 years ago
3
deepspeed init hang住
#7
lonelydancer
closed
3 years ago
3
CPM-2-Finetuning的推理速度有多快(在V100上)
#6
lonelydancer
closed
1 year ago
2
adgen数据集
#5
zhenhao-huang
closed
3 years ago
4
finetune的最小配置是?
#4
eshaoliu
closed
3 years ago
10
V100单卡能inference吗
#3
zuowang
closed
3 years ago
1
报错
#2
superqing001
closed
3 years ago
0
是否可以提供docker运行的脚本参考?
#1
superqing001
closed
3 years ago
2
Next