Comments (2)
Hi @FieldRen, The current memory implementations in langchain are heuristic-based, meaning you are using all the conversation history or the last k messages. Hence, such approaches are not adaptive if the user changes the topic which was part of earlier conversation history, the heuristic approaches might miss it because of token overflow or it could be outside the last k messages. We in our case try to make the history component adaptive hence, it only brings the previous k messages which are relevant to the current message from the entire history. Hence, adding more relevant context in the prompt and never running out of token length. Below is an example:
If you want to delve deeper into current memory implementations of langchain, here is a great article.
from chatgpt-memory.
I got it, It's a very good idea.Thanks for your explanation!
from chatgpt-memory.
Related Issues (20)
- Implement adaptive memory
- Add ChatGPTClient
- have consistent docstrings & type hints for entire repo.
- add dotenv template
- Add detailed Docs HOT 2
- Installing the current project: chatgpt-memory (0.0.1) HOT 1
- Important Question / History by user HOT 2
- Generates irrelevant / too much conversation? HOT 5
- Error loading ASGI app. Could not import module "rest_api" HOT 2
- conversation id returns error: redis.exceptions.ResponseError: idx: no such index HOT 2
- ModuleNotFoundError: No module named 'redis' HOT 10
- getting error redis.exceptions.ResponseError: Unknown Index name HOT 1
- gpt-4 model not supported HOT 2
- Retaining conversation id's between process executions? HOT 5
- Do I need to initialize the Redis database first? HOT 4
- Model name change HOT 5
- Add memory interface
- add datastore interface
- Implement memory manager
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
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.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from chatgpt-memory.