crazyCrab
获得徽章 6
赞了这篇文章
赞了这篇文章
赞了这篇文章
赞了这篇文章
赞了这篇文章
赞了这篇文章
赞了这篇文章
赞了这篇文章
赞了这篇文章
#豆包MarsCode AI练中学# 补充一下10_记忆/05_ConversationSummaryBufferMemory 由于统计token的方法失败所以得自己重载一下汇总token的方法 ```get_num_tokens_from_messages```。粗略的实现，本质是避免这个脚本报错。
```
class CustomChatOpenAI(ChatOpenAI):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
def get_num_tokens_from_messages(self, messages):
"""Returns the number of tokens used by a list of messages."""
print(">>> messages: ", messages)
num_tokens = 0
for message in messages:
num_tokens += 4 # every message follows <im_start>{role/name}\n{content}<im_end>\n
num_tokens += len(message.content)
num_tokens += 2 # every reply is primed with <im_start>assistant
return num_tokens
```
```
class CustomChatOpenAI(ChatOpenAI):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
def get_num_tokens_from_messages(self, messages):
"""Returns the number of tokens used by a list of messages."""
print(">>> messages: ", messages)
num_tokens = 0
for message in messages:
num_tokens += 4 # every message follows <im_start>{role/name}\n{content}<im_end>\n
num_tokens += len(message.content)
num_tokens += 2 # every reply is primed with <im_start>assistant
return num_tokens
```
展开
评论
点赞
#豆包MarsCode AI练中学# 15_RAG应用/03_VectorstoreIndex VectorstoreIndexCreator中的Qdrant参数原来是在vectorstore_kwargs这里写![[发呆]](//lf-web-assets.juejin.cn/obj/juejin-web/xitu_juejin_web/img/jj_emoji_4.28b310a.png)
![[发呆]](http://lf-web-assets.juejin.cn/obj/juejin-web/xitu_juejin_web/img/jj_emoji_4.28b310a.png)
评论
4
赞了这篇文章