EasyNLP

EasyNLP copied to clipboard

复现时出现问题怎么回事儿？

5

您好~ 我按照您的要求配置对应环境，并且下载了相应的预训练模型和数据后还是执行不了，具体错误如下所示：怎么回事儿呢？ File "/home/.local/lib/python3.8/site-packages/datasets/dataset dict,py", line 472, in k: dataset.map(File "/home/local/lib/python3,8/site-packages/datasets/arrow dataset.py", line 1657, in mapreturn self.map singleFile "/home/.local/lib/python3,8/sitepackages/datasets/arrow dataset.py", line 185, in wrapperout: Union["DatasetuDatasetDictu] func(self,*args，**kwargs)File "/home/.local/lib/python3,8/site-packages/datasets/fingerprint,py", line 397, in...

Shajiu

Is "Unified Prompt Tuning (UPT) framework" source code available in this repo?

3

I read [https://arxiv.org/pdf/2205.05313.pdf](https://arxiv.org/pdf/2205.05313.pdf ) paper and I am interested in testing it, but the paper mentioned this. 

indunil-19

我看实例代码里用 train_steps = 100 inference_steps = 10 prompt = "亭台楼阁，曲径通幽，水墨绘画，中国风" olss 调度器，对 预训练的train_steps inference_steps promp 是否敏感？ 如果不同的train_steps ，inference_steps ，promp 差距很大，有无推荐；能复现论文的结果

seeyourcell

The current OLSS only supports diffusers=0.21.3. Can you update OLSS to support the updated version of diffusers? This way, many people can use OLSS. Thanks.

ZihaoW123

FreePromptEditing:TypeError: SelfAttentionControlEdit.__init__() got multiple values for argument 'self_replace_steps'

1

what's the problem

ersanliqiao

Diffsynth Fashion Video synthsis 里的模型 model_id

1

DyMessi