GithubHelp home page GithubHelp logo

关于推理预测 about qwen-vl HOT 15 OPEN

elesun2018 avatar elesun2018 commented on August 16, 2024
关于推理预测

from qwen-vl.

Comments (15)

elesun2018 avatar elesun2018 commented on August 16, 2024

请问huggingface中多出的这三个文件作用,应该如何使用。
image

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

用lora微调qwen-vl模型,用peft merge_and_unload save_pretrained保存成huggingface模型文件
用Qwen-VL# python web_demo_mm.py加载这个huggingface模型文件进行推理,报错:
assert generation_config.chat_format == 'chatml', _ERROR_BAD_CHAT_FORMAT
AssertionError: We detect you are probably using the pretrained model (rather than chat model) for chatting, since the chat_format in generation_config is not "chatml".
If you are directly using the model downloaded from Huggingface, please make sure you are using our "Qwen/Qwen-7B-Chat" Huggingface model (rather than "Qwen/Qwen-7B") when you call model.chat().
我们检测到您可能在使用预训练模型(而非chat模型)进行多轮chat,因为您当前在generation_config指定的chat_format,并未设置为我 们在对话中所支持的"chatml"格式。
如果您在直接使用我们从Huggingface提供的模型,请确保您在调用model.chat()时,使用的是"Qwen/Qwen-7B-Chat"模型(而非"Qwen/Qwen-7B"预训练模型)。

请问如何修改,谢谢!

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

image
image
这三个文件是要直接拷贝近huggingface模型文件output_qwen_hf ?

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

请问按照readme lora合并保存模型
image
没看到加载预训练路径的代码,AutoPeftModelForCausalLM这个会自动下载huggingface上的 QwenVL预训练模型吗,他是根据哪个配置选项自动下载QwenVL还是Qwen chat模型文件的?

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

请问什么是ChatML格式

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

推理时发现跟没训练几乎没区别
lora训练后数参数,如何知道正在合并到了新的整体模型中了
image
谢谢

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

能否帮忙解答一下,谢谢

from qwen-vl.

fanshuaiyao avatar fanshuaiyao commented on August 16, 2024

image image这三个文件是要直接拷贝近huggingface模型文件output_qwen_hf ?

我是这个干的 也成功了 但是预测的效果很不好

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

请问xi现在是否支持vllm加速
https://docs.vllm.ai/en/latest/models/supported_models.html
image
如何才能支持VLLM加速,谢谢

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

@fanshuaiyao 跟这个有关系么?
image

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

请问如何将LLM(lm_head)输出概率值shift_logits转换成文本答案text。
经过:
predict_ids = np.argmax(results.shift_logits, axis=-1)
text = tokenizer.batch_decode(predict_ids, skip_special_tokens=True)
发现text 绝大部分为乱码显示。
谢谢!

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

能否解答一下上述几个问题,谢谢!

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

发现huggingface预训练模型中tokenizer的tokenizer_config.json与fientune tokenizer.save_pretrained保存的内容不太一致!
image
clean_up和model_max_length
请问这些参数不同有没有影响,谢谢!

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

能否解答一下上述几个问题,谢谢!

from qwen-vl.

elesun2018 avatar elesun2018 commented on August 16, 2024

能否解答一下上述几个问题,谢谢!

from qwen-vl.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.