Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/ryosukedtomita/langchaintest
LangChainをとりあえず使ってみるためにJupyter notebookとVSCocdeの環境を整える。
https://github.com/ryosukedtomita/langchaintest
docker jupyter-extension langchain-python vscode
Last synced: about 2 months ago
JSON representation
LangChainをとりあえず使ってみるためにJupyter notebookとVSCocdeの環境を整える。
- Host: GitHub
- URL: https://github.com/ryosukedtomita/langchaintest
- Owner: RyosukeDTomita
- License: unlicense
- Created: 2024-03-12T10:25:21.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2024-03-14T13:07:22.000Z (10 months ago)
- Last Synced: 2024-05-11T05:55:24.980Z (8 months ago)
- Topics: docker, jupyter-extension, langchain-python, vscode
- Language: Jupyter Notebook
- Homepage:
- Size: 6.84 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Setting up a Jupyter Notebook using a Docker container to use Lang Chain and access it through VS Code Jupyter extensions.
## INDEX
- [ABOUT](#about)
- [LICENSE](#license)
- [ENVIRONMENT](#environment)
- [PREPARING](#preparing)
- [HOW TO USE](#how-to-use)
- [ERROR LOG](#error-log)
- [REFERENCE](#reference)
---
## ABOUT
- [Lang Chain のチュートリアル](https://python.langchain.com/docs/expression_language/get_started) をやってみるための環境構築をしました。
- VSCode の Extensions で Jupyter Notebook を立ち上げ，Docker Container 内の Jupyter サーバーにアクセスして使用します。
---
## LICENSE
---
[un license](./LICENSE)
---
## ENVIRONMENT
- Python: see [Dockerfile](./Dockerfile)
- VSCode with [Jupyter Notebook Extensions](https://marketplace.visualstudio.com/items?itemName=ms-toolsai.jupyter)
---
## PREPARING
- install [Jupyter Notebook Extensions](https://marketplace.visualstudio.com/items?itemName=ms-toolsai.jupyter)
- Buy OpenAi API [here](https://platform.openai.com/account/api-keys)
---
## HOW TO USE
1. edit [.env](./.env) to add your OpenAi API KEY
```shell
cat <> .env
OPENAI_API_KEY="hogehoge"
EOF
```
Azure OpenAIを使う場合
```shell
cat <> .env
OPENAI_API_VERSION=2023-12-01-preview
AZURE_OPENAI_ENDPOINT=https://example.com
AZURE_OPENAI_API_KEY=hogehoge
EOF
```
2. run Docker container
```shell
docker compose up
lcel-container | [I 2024-03-13 04:54:57.732 ServerApp] http://127.0.0.1:8888/tree?token=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
```
3. Open [./test.ipybn](./test.ipybn) in VSCode Jupyter Extensions
4. Select Kernel --> Select Another Kernel --> token 付きの URL を docker compose up の標準出力に表示されている URL を入力
---
## ERROR LOG
### raise HTTPError(403, "'\_xsrf' argument missing from POST"
- token を空にすると VSCode の Jupyter の Extension から kernel 指定時に xsrf エラーが出る。
```
# compose.yml
#command: jupyter notebook --ip=0.0.0.0 --port=8888 --allow-root --NotebookApp.password='' --NotebookApp.token=''
command: jupyter notebook --ip=0.0.0.0 --port=8888 --allow-root --NotebookApp.password=''
```
### Azure OpenAI 使用時に BadRequestError: Error code: 400 OperationNotSupported The completion operation does not work が出る
- [model の種類](https://learn.microsoft.com/ja-jp/azure/ai-services/openai/concepts/models#gpt-35-models)を見ると，gpt-35-turbo は completion をサポートしていないので gpt-35-turbo-instruct に変更する。これは 2024/03/14 時点ではアメリカリージョンでしか使えない。
---
## REFERENCE