jspr commited on
Commit
ee8cd9e
·
1 Parent(s): ac3b3f0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -32
README.md CHANGED
@@ -1,32 +1 @@
1
- # PaperChat
2
-
3
- This repo is an implementation of a chatbot specifically focused on question answering over the [LangChain documentation](https://langchain.readthedocs.io/en/latest/).
4
-
5
- ## 🚀 Important Links
6
-
7
- Website: [chat.langchain.dev](https://chat.langchain.dev)
8
-
9
- Hugging Face Spage: [huggingface.co/spaces/hwchase17/chat-langchain](https://huggingface.co/spaces/hwchase17/chat-langchain)
10
-
11
- Blog Post: [blog.langchain.dev/langchain-chat/](https://blog.langchain.dev/langchain-chat/)
12
-
13
- ## 📚 Technical description
14
-
15
- There are two components: ingestion and question-answering.
16
-
17
- Ingestion has the following steps:
18
-
19
- 1. Pull html from documentation site
20
- 2. Parse html with BeautifulSoup
21
- 3. Split documents with LangChain's [TextSplitter](https://langchain.readthedocs.io/en/latest/modules/utils/combine_docs_examples/textsplitter.html)
22
- 4. Create a vectorstore of embeddings, using LangChain's [vectorstore wrapper](https://langchain.readthedocs.io/en/latest/modules/utils/combine_docs_examples/vectorstores.html) (with OpenAI's embeddings and Weaviate's vectorstore)
23
-
24
- Question-Answering has the following steps:
25
-
26
- 1. Given the chat history and new user input, determine what a standalone question would be (using GPT-3)
27
- 2. Given that standalone question, look up relevant documents from the vectorstore
28
- 3. Pass the standalone question and relevant documents to GPT-3 to generate a final answer
29
-
30
- ## 🧠 How to Extend to your documentation
31
-
32
- Coming soon.
 
1
+ # paperchat