Spaces:
Runtime error
Runtime error
Joshua Sundance Bailey
commited on
Commit
·
241c14d
1
Parent(s):
5e6fcec
README
Browse files
README.md
CHANGED
|
@@ -26,7 +26,7 @@ tags: [langchain, streamlit, docker]
|
|
| 26 |
|
| 27 |
This project shows how to build a simple chatbot UI with [Streamlit](https://streamlit.io) and [LangChain](https://langchain.com).
|
| 28 |
|
| 29 |
-
This `README` was written by [Claude 2](https://www.anthropic.com/index/claude-2), an LLM from [Anthropic](https://www.anthropic.com/).
|
| 30 |
|
| 31 |
# Features
|
| 32 |
- Chat interface for talking to AI assistant
|
|
@@ -41,19 +41,17 @@ This `README` was written by [Claude 2](https://www.anthropic.com/index/claude-2
|
|
| 41 |
- `meta-llama/Llama-2-7b-chat-hf`
|
| 42 |
- `meta-llama/Llama-2-13b-chat-hf`
|
| 43 |
- `meta-llama/Llama-2-70b-chat-hf`
|
|
|
|
| 44 |
- Streaming output of assistant responses
|
| 45 |
- Leverages LangChain for dialogue and memory management
|
| 46 |
- Integrates with [LangSmith](https://smith.langchain.com) for tracing conversations
|
| 47 |
- Allows giving feedback on assistant's responses
|
| 48 |
- Tries reading API keys and default values from environment variables
|
| 49 |
- Parameters in sidebar can be customized
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
-
|
| 53 |
-
-
|
| 54 |
-
- `Dockerfile`, `docker-compose.yml`: Docker deployment
|
| 55 |
-
- `kubernetes/`: Kubernetes deployment files
|
| 56 |
-
- `.github/workflows/`: CI/CD workflows
|
| 57 |
|
| 58 |
# Deployment
|
| 59 |
`langchain-streamlit-demo` is deployed as a [Docker image](https://hub.docker.com/r/joshuasundance/langchain-streamlit-demo) based on the [`python:3.11-slim-bookworm`](https://github.com/docker-library/python/blob/81b6e5f0643965618d633cd6b811bf0879dee360/3.11/slim-bookworm/Dockerfile) image.
|
|
@@ -71,7 +69,7 @@ CI/CD workflows in `.github/workflows` handle building and publishing the image
|
|
| 71 |
|
| 72 |
or
|
| 73 |
|
| 74 |
-
`docker run -p 7860:7860
|
| 75 |
|
| 76 |
3. Open http://localhost:7860 in your browser
|
| 77 |
|
|
@@ -82,10 +80,6 @@ or
|
|
| 82 |
|
| 83 |
`docker compose up`
|
| 84 |
|
| 85 |
-
or
|
| 86 |
-
|
| 87 |
-
`docker compose up --env-file .env`
|
| 88 |
-
|
| 89 |
4. Open http://localhost:7860 in your browser
|
| 90 |
|
| 91 |
## Kubernetes
|
|
|
|
| 26 |
|
| 27 |
This project shows how to build a simple chatbot UI with [Streamlit](https://streamlit.io) and [LangChain](https://langchain.com).
|
| 28 |
|
| 29 |
+
This `README` was originally written by [Claude 2](https://www.anthropic.com/index/claude-2), an LLM from [Anthropic](https://www.anthropic.com/).
|
| 30 |
|
| 31 |
# Features
|
| 32 |
- Chat interface for talking to AI assistant
|
|
|
|
| 41 |
- `meta-llama/Llama-2-7b-chat-hf`
|
| 42 |
- `meta-llama/Llama-2-13b-chat-hf`
|
| 43 |
- `meta-llama/Llama-2-70b-chat-hf`
|
| 44 |
+
- `codellama/CodeLlama-34b-Instruct-hf`
|
| 45 |
- Streaming output of assistant responses
|
| 46 |
- Leverages LangChain for dialogue and memory management
|
| 47 |
- Integrates with [LangSmith](https://smith.langchain.com) for tracing conversations
|
| 48 |
- Allows giving feedback on assistant's responses
|
| 49 |
- Tries reading API keys and default values from environment variables
|
| 50 |
- Parameters in sidebar can be customized
|
| 51 |
+
- Includes various forms of document chat
|
| 52 |
+
- Question/Answer Pair Generation
|
| 53 |
+
- Summarization
|
| 54 |
+
- Standard retrieval chains
|
|
|
|
|
|
|
|
|
|
| 55 |
|
| 56 |
# Deployment
|
| 57 |
`langchain-streamlit-demo` is deployed as a [Docker image](https://hub.docker.com/r/joshuasundance/langchain-streamlit-demo) based on the [`python:3.11-slim-bookworm`](https://github.com/docker-library/python/blob/81b6e5f0643965618d633cd6b811bf0879dee360/3.11/slim-bookworm/Dockerfile) image.
|
|
|
|
| 69 |
|
| 70 |
or
|
| 71 |
|
| 72 |
+
`docker run -p 7860:7860 --env-file .env joshuasundance/langchain-streamlit-demo:latest`
|
| 73 |
|
| 74 |
3. Open http://localhost:7860 in your browser
|
| 75 |
|
|
|
|
| 80 |
|
| 81 |
`docker compose up`
|
| 82 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 83 |
4. Open http://localhost:7860 in your browser
|
| 84 |
|
| 85 |
## Kubernetes
|