TomSuen

> @ecnuycxie As you said, I can generate new images now. But the result seems not always fine.  I use Song as the content image, and write chinese image...

> 你好，请问可以问一下您的第二阶段的数据准备是怎么准备的吗，我的准备过程一直在出问题： 运行train后会报错：Traceback (most recent call last): File "F:\python_project\VQ-Font-main\VQ-Font-main\train.py", line 221, in main() File "F:\python_project\VQ-Font-main\VQ-Font-main\train.py", line 217, in main train(args, cfg) File "F:\python_project\VQ-Font-main\VQ-Font-main\train.py", line 140, in train drop_last=True) File "F:\python_project\VQ-Font-main\VQ-Font-main\datasets__init__.py",...

> Thanks for your suggestion. Voice blending might be possible but not sure it works or not. We may conduct some experiments to find out. Hi, I have met with...

> Sounds fine to me. Do you mean the strong emotion was not reflected from the converted audio? Yes, I would like to ask how to better preserve the emotion...

> Currently it is an issue, we will study more on that Look forward to your good news😀

> I have successfully tested the flash checkpoint using the following command with 4 A100 nodes in the [forked repo](https://github.com/workingloong/Megatron-LM-CKPT) whose commit id is [cb995d5](https://github.com/NVIDIA/Megatron-LM/tree/cb995d571faea19d01a1bf55ed0fd89523b9ce64). > > ``` > dlrover-run...

您好，请问你只是用了flash checkpoint功能吗？我想知道dlrover-run可以跟deepspeed一起用吗 @liangxuZhang

> When using **dlrover version 0.3.7** in an environment set to **nvcr.io/nvidia/pytorch:24.01-py3**, and when used in conjunction with Megatron-LM, if the rank value is **set only in the script and...

闲鱼购买，但是里边会有木马病毒，注意查杀。

> > 闲鱼购买，但是里边会有木马病毒，注意查杀。 > > 请问里面是这样的图片。如何做成数据集放到ttf_folder目录下了？ @TomSuen 用项目中的font2img.py脚本把.ttf转换为png就可以了