search

zyang1580 / CoLLM

The implementation for the work "CoLLM: Integrating Collaborative Embeddings into Large Language Models for Recommendation".
BSD 3-Clause "New" or "Revised" License
47 stars 6 forks source link

running time #7

Closed NicholasEinstein closed 4 months ago

NicholasEinstein commented 4 months ago

作者你好，请问3090能不能跑你的代码呢，你用的两张A100的训练时间大概是多久呢

NicholasEinstein commented 4 months ago

如果采用llama2-7b模型 alpaca lora的方式会不会更快一点

zyang1580 commented 4 months ago

3090 没有尝试过，A40应该可以，A100的运行时间我记得阶段一的范围在6-20小时，阶段二比较快，具体多少不记得了。（collm模型架构部分的代码可以简化下，等一段时间更新简化后的版本）

alpaca lora的方式没有尝试，过段时间空闲的时候我尝试下。

NicholasEinstein @.***> 于2024年3月9日周六 00:17写道：

如果采用llama2-7b模型 alpaca lora的方式会不会更快一点

— Reply to this email directly, view it on GitHub https://github.com/zyang1580/CoLLM/issues/7#issuecomment-1985981651, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMO3KU5ZJ6WF5NMGZNIYF4DYXHQBDAVCNFSM6AAAAABENBHAUKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSOBVHE4DCNRVGE . You are receiving this because you are subscribed to this thread.Message ID: @.***>