issues
search
yuanzhoulvpi2017
/
zero_nlp
中文nlp解决方案(大模型、数据、模型、训练、推理)
MIT License
2.85k
stars
355
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
合并Lora权重后的模型不生成回答了
#84
heccxixi
closed
1 year ago
0
量化后的模型没法分层
#83
aihaidong
closed
1 year ago
1
模型并行问题
#82
juemifuji
opened
1 year ago
4
改名
#81
yangliuIOC
closed
1 year ago
0
三个输入都有错误探讨
#80
onePlusOne111
opened
1 year ago
1
EOP TOKEN ID
#79
yangliuIOC
closed
1 year ago
2
显存
#78
yangliuIOC
closed
1 year ago
4
多卡并行的训练方法只用1张卡测试，max_seq_len 1024, batch_size 1还是会内存不够，用的3090~~~
#77
Rorschaaaach
closed
1 year ago
3
怎么让它不知道的就不要乱说。
#76
cywjava
closed
1 year ago
0
main_parallel.py 报错 print_dataset_example这里报错 preprocess_function_train错了
#75
hangzeli08
closed
1 year ago
1
chatglm不可以自动层并行吗？
#74
kevinuserdd
closed
1 year ago
3
Chatglm6b_ModelParallel_ptuning 编译错误
#73
online2311
closed
1 year ago
1
执行sh脚本报错IndexError: Out of range: piece id is out of range.
#72
janglichao
closed
1 year ago
1
Lora导致推理时长增加70%
#71
airsYuan
opened
1 year ago
4
关于modeling_chatglm.py
#70
ckqsars
opened
1 year ago
0
训练后的模型infer的时候报输入形状错误 RuntimeError: Tensors must have same number of dimensions: got 4 and 2
#69
xxyp
opened
1 year ago
4
作者你好，改名字只成功了一半
#68
YYGe01
opened
1 year ago
9
'ChatGLM Tokenizer' object has no attribute 'eos_token_id'
#67
OneStepAndTwoSteps
closed
1 year ago
8
infer的速度很慢
#66
OneStepAndTwoSteps
opened
1 year ago
1
有人训练“你是谁”成功了吗？
#65
BLAIR-wy
closed
1 year ago
5
关掉Lora微调大模型，模型并行训练报错：Expected all tensors to be on the same device, but found at least two devices, cuda:3 and cuda:0!
#64
huangcaiyun
opened
1 year ago
7
训练时，我想关掉fp16
#63
cywjava
closed
1 year ago
2
使用lora 微调后，调用生成报错 RuntimeError: expected scalar type Half but found Float
#62
cywjava
closed
1 year ago
3
报错: RuntimeError: Internal: [MASK] is already defined.
#61
EssentialCuber
closed
1 year ago
4
原封不动下载的代码和数据，在已经成功运行官方版本的环境里出错
#60
xianglei3
closed
1 year ago
2
使用lora 微调后，怎么所有的checkpoint 的大小都是一样的？
#59
cywjava
closed
1 year ago
1
wandb这里卡住了，怎么解决
#58
rucideyi
opened
1 year ago
2
最新版的多卡并行
#57
cywjava
closed
1 year ago
3
只训练大模型，并行出错
#56
safehumeng
opened
1 year ago
7
训练后的模型不能像chatglm-6B中生成的模型一样被加载成为接口。报输入形状错误。
#55
natureLanguageQing
closed
1 year ago
1
训练时数据是不是需要处理下？
#54
yzho0907
opened
1 year ago
0
ValueError: Unrecognized configuration class
#53
littlerookie
opened
1 year ago
3
instruction, input, output都代表什么意思，有相关的文档么？
#52
bh4ffu
opened
1 year ago
1
Mytrainer.py有15个错误，没有引入相关的包
#51
luieswww
opened
1 year ago
2
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`...报这个信息，保存不了模型文件
#50
Chenzongchao
closed
1 year ago
1
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`...
#49
Chenzongchao
closed
1 year ago
1
如何给ChatGLM-6B增加特定领域的知识，然后根据这些知识来问答
#48
Mr-IT007
closed
1 year ago
4
可以增量训练么？
#47
bh4ffu
closed
1 year ago
3
模型对预训练数据集拟合效果很差
#46
yanchaoguo
opened
1 year ago
6
我用了BELLE的0.5M语料训练
#45
1079863482
opened
1 year ago
4
多卡并行训练报错
#44
cywjava
closed
1 year ago
5
'ChatGLMForConditionalGeneration' object has no attribute 'model_parallel' 大佬这个是因为没开多卡吗
#43
Chenzongchao
closed
1 year ago
2
训练的epoch数
#42
xiaoweiweixiao
opened
1 year ago
2
使用训练后的模型报错
#41
bh4ffu
closed
1 year ago
5
closed
#40
xiaosimao
closed
1 year ago
0
大佬修改名称有什么经验呢
#39
Chenzongchao
closed
1 year ago
4
训练后没有效果，我换了data2里面的内容后，又报如下错误 。。
#38
cywjava
opened
1 year ago
10
微调后怎么启动一个api server供外部调用？
#37
bh4ffu
closed
1 year ago
2
微调后的checkpoint 能保存为原来的bin格式的文件吗？
#36
cywjava
closed
1 year ago
1
交友贴
#35
PKQ1688
opened
1 year ago
3
Previous
Next