G.O.D
a java interface, hope it help https://github.com/gameofdimension/jni-faiss.git
use tensorboard to inspect the graph
到 docker 容器里把 jar 扒出来加到你自己的项目
this function ensure you passed the params (lambda and eta) exactly as you trained the xgboost model, if you are sure about this you can safely comment out this line....
https://github.com/vllm-project/vllm/pull/1558
你的 gpu 显存多大
试下这个分支 https://github.com/gameofdimension/vllm/tree/13b ```python from vllm import LLM, SamplingParams llm = LLM(model="baichuan-inc/baichuan-13B", trust_remote_code=True) prompts = [ "Hello, my name is", "The president of the United States is", "The capital of France...
跟非 vllm 的比较一下呢
那就是这个模型本身的问题了，跟 vllm 适配无关哈
据说 7B 的 tp 也有问题，13B tp 部分是沿用 7B 的。可以参考下这里的讨论 https://github.com/vllm-project/vllm/pull/530#issuecomment-1650859711