Wangman1

Results 20 comments of Wangman1

Hi, when I train the model in your mode, raise the follow error: ![image](https://user-images.githubusercontent.com/33320872/126311901-a34e9932-a19d-4f1d-8f65-a5ec845e5195.png)

And the Google link to JSON file cannot download the content. It does not have relevant download permission. ![image](https://user-images.githubusercontent.com/33320872/126320682-701fdca6-ffb7-41e7-a861-f26507d292ee.png)

@lhwcv Thank you so much for your help. My code has been successfully run!!! ![image](https://user-images.githubusercontent.com/33320872/126426744-0f3a9ca9-6030-4b22-82c2-e77334048085.png)

请问各位,这个问题有解决吗

好嘞,感谢您的回复,训练启动命令有可以参考的吗,相关介绍中好像没有找到pretrain的信息

好嘞,非常感谢您的回复~~

求问从头预训练qwen2-vl的话,能不能分别加载 qwen2-7b 语言模型和 vision_encoder 呢,如果可以的话大概需要怎么修改代码~

> 可以临时用 lmdeploy 部署,vllm要等后续的更新 您好,请问有计划支持vllm推理吗,目前使用lmdeploy和vllm推理和transformer的结果都有很大的gap,一致性无法解决