We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
我按照04-Qwen2-VL-2B Lora 微调.md的教程进行了单图的sft,是ok的。 然后因为个人需要,我想测试一下能否进行多图的sft。训练数据按照如下格式准备的。 messages = [ { "role": "user", "content": [ { "type": "image", "image": f"{file_path}", "resized_height": 280, "resized_width": 280, }, { "type": "image", "image": f"{file_path}", "resized_height": 280, "resized_width": 280, }, {"type": "text", "text": "COCO Yes:"}, ], } ] 测试结果,按照这个message格式,推理是ok的,但是训练会报错。 报错如下:
我定位了下,是image_grid_thw tensor这个的tensor维度发生了变化。(正常训练的tensor shape的是[2,3])
The text was updated successfully, but these errors were encountered:
让我看看
Sorry, something went wrong.
多图sft,这是什么场景?可以考虑开一个新的教程文档来适配这个场景
ds的vl它有一个推理场景是这样的,试了下推理的这种能力有限,想试着按照这种情况sft一下。
No branches or pull requests
我按照04-Qwen2-VL-2B Lora 微调.md的教程进行了单图的sft,是ok的。
然后因为个人需要,我想测试一下能否进行多图的sft。训练数据按照如下格式准备的。
messages = [
{
"role": "user",
"content": [
{
"type": "image",
"image": f"{file_path}",
"resized_height": 280,
"resized_width": 280,
},
{
"type": "image",
"image": f"{file_path}",
"resized_height": 280,
"resized_width": 280,
},
{"type": "text", "text": "COCO Yes:"},
],
}
]
测试结果,按照这个message格式,推理是ok的,但是训练会报错。
报错如下:
我定位了下,是image_grid_thw tensor这个的tensor维度发生了变化。(正常训练的tensor shape的是[2,3])
The text was updated successfully, but these errors were encountered: