You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
想法背景:
1、按照部署文档,在服务器部署了,并尝试使用azure open api ,只成功了webserver,使用azure openapi的完整chat流程 还未成功。
2、想着本地debug看看,按照同样的源码安装步骤,下载源码,安装依赖后,又报错如下:
File "/Users/caoyuan/githubProject/DB-GPT/pilot/server/llm_manage/api.py", line 8, in <module> from pilot.model.cluster import WorkerStartupRequest, WorkerManagerFactory File "/Users/caoyuan/githubProject/DB-GPT/pilot/model/cluster/__init__.py", line 10, in <module> from pilot.model.cluster.worker.default_worker import DefaultModelWorker File "/Users/caoyuan/githubProject/DB-GPT/pilot/model/cluster/worker/default_worker.py", line 6, in <module> from pilot.model.model_adapter import get_llm_model_adapter, LLMModelAdaper File "/Users/caoyuan/githubProject/DB-GPT/pilot/model/model_adapter.py", line 450, in <module> register_conv_template( File "/Users/caoyuan/opt/anaconda3/envs/db_gpt_dev/lib/python3.10/site-packages/fastchat/conversation.py", line 290, in register_conv_template template.name not in conv_templates AssertionError: aquila-legacy has been registered.
目前尚未解决,不知道是不是和版本有关系。
问题,讨论点是: 我只想在基于azure open api进行部分操作或使用的情况下,fastchat对我的必要性不大,但是我却需要把fschat下的东西,进行启动,管理,安装,报错,进而阻塞。
但是fschat本身也是将基于gradio的webdemo,cli,controller,woker以及openai api wapper拆分的,可以各自启动。
当然我理解可能是基于需要使用私有 emebdding 的考虑?
不这一块不知道是怎么考虑的? 有考虑进行拆分吗,这样比如我使用azure 或者 其他国内如文心服务的时候,我其实是不太需要这些模块的。
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
想法背景:
1、按照部署文档,在服务器部署了,并尝试使用azure open api ,只成功了webserver,使用azure openapi的完整chat流程 还未成功。
2、想着本地debug看看,按照同样的源码安装步骤,下载源码,安装依赖后,又报错如下:
File "/Users/caoyuan/githubProject/DB-GPT/pilot/server/llm_manage/api.py", line 8, in <module> from pilot.model.cluster import WorkerStartupRequest, WorkerManagerFactory File "/Users/caoyuan/githubProject/DB-GPT/pilot/model/cluster/__init__.py", line 10, in <module> from pilot.model.cluster.worker.default_worker import DefaultModelWorker File "/Users/caoyuan/githubProject/DB-GPT/pilot/model/cluster/worker/default_worker.py", line 6, in <module> from pilot.model.model_adapter import get_llm_model_adapter, LLMModelAdaper File "/Users/caoyuan/githubProject/DB-GPT/pilot/model/model_adapter.py", line 450, in <module> register_conv_template( File "/Users/caoyuan/opt/anaconda3/envs/db_gpt_dev/lib/python3.10/site-packages/fastchat/conversation.py", line 290, in register_conv_template template.name not in conv_templates AssertionError: aquila-legacy has been registered.
目前尚未解决,不知道是不是和版本有关系。
问题,讨论点是: 我只想在基于azure open api进行部分操作或使用的情况下,fastchat对我的必要性不大,但是我却需要把fschat下的东西,进行启动,管理,安装,报错,进而阻塞。
但是fschat本身也是将基于gradio的webdemo,cli,controller,woker以及openai api wapper拆分的,可以各自启动。
当然我理解可能是基于需要使用私有 emebdding 的考虑?
不这一块不知道是怎么考虑的? 有考虑进行拆分吗,这样比如我使用azure 或者 其他国内如文心服务的时候,我其实是不太需要这些模块的。
Beta Was this translation helpful? Give feedback.
All reactions