ChatLLM - GitLite

KO | EN

gitlite — search

#ai #llm #golang #go #android #machine-learning #rust #linux #docker #nodejs #awesome-list #vue #security #pytorch

ChatLLM

(★ 55)

Test script of LLMs

File Explorer

Download Latest Version (.zip)

- .devcontainer.json
- Dockerfile
- .devcontainer.json
- Dockerfile
- .devcontainer.json
- Dockerfile
.gitignore
chat_calm.py
chat_calm2.py
chat_elyza.py
chat_gemma.py
chat_gemma_ja.py
chat_llama2.py
chat_rinna.py
chat_rwkv.py
chat_stablelm.py
chat_weblab.py
chat_weblab_gptq.py
docker-compose.yaml
Dockerfile
LICENSE
README.md

# Installation Guide

1. Get the code

git clone https://github.com/karaage0703/ChatLLM

Downloads the entire project code from GitHub to your computer.

cd ChatLLM

Moves into the project folder you just downloaded.

2. Docker

Easy Recommended

Prerequisites

Git Needed to download the project code from GitHub.
Docker Desktop Needed to build and run containers. Install it and keep it running in the background.

$ docker compose up base

Builds and starts all defined containers (server, database, etc.) at once, in the background.

$ docker compose up gpu

Builds and starts all defined containers (server, database, etc.) at once, in the background.

$ docker compose up nvidia

Builds and starts all defined containers (server, database, etc.) at once, in the background.

$ docker ps

Type this command into your terminal and run it.

$ docker exec -it <image name> /bin/bash

Type this command into your terminal and run it.

✅ Run docker compose ps to check the containers are Up. If the README mentions a port, open http://localhost:PORT in your browser.

Pulled directly from this repo's README.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# ChatLLM
Chat test scripts for LLM

## Setup
Git clone
```sh
$ cd && git clone https://github.com/karaage0703/ChatLLM
```

### Devcontainer(VS Code)

Setup devcontainer

[Reference(Japanese)](https://zenn.dev/karaage0703/books/80b6999d429abc8051bb/viewer/6ebae8)

### Docker compose

#### Docker build
CPU

```sh
$ docker compose up base
```

GPU

```sh
$ docker compose up gpu
```

GPU(Nvidia container)

```sh
$ docker compose up nvidia
```
#### Docker run

Check docker image name

```sh
$ docker ps
```

```sh
$ docker exec -it <image name> /bin/bash

# example
# $ docker exec -it chatllm-nvidia-1 /bin/bash
```

#### Run app

```sh
root@hostname:/# cd /root
root@hostname:~# python3 chat_calm.py
root@hostname:~# python3 chat_rinna.py
root@hostname:~# python3 chat_rwkv.py
root@hostname:~# python3 chat_llama2.py
root@hostname:~# python3 chat_weblab.py
root@hostname:~# python3 chat_elyza.py
```

### Docker

#### Docker build

```
$ cd ~/ChatLLM
$ docker build -t ubuntu:ChatLLM .
```

#### Run docker
Use GPU

```sh
$ cd ~/ChatLLM
$ docker run -it --rm -v $(pwd):/root --gpus all ubuntu:ChatLLM
```

Use CPU

```sh
$ cd ~/ChatLLM
$ docker run -it --rm -v $(pwd):/root ubuntu:ChatLLM
```

#### Run app

```sh
root@hostname:~# python3 chat_calm.py
root@hostname:~# python3 chat_rinna.py
root@hostname:~# python3 chat_rwkv.py
root@hostname:~# python3 chat_llama2.py
```

for stablelm

```sh
root@hostname:/# huggingface-cli login
root@hostname:~# python3 chat_stablelm.py
```

## References
- https://huggingface.co/cyberagent/open-calm-7b
- https://huggingface.co/rinna/japanese-gpt-neox-3.6b-instruction-sft
- https://nowokay.hatenablog.com/entry/2023/05/17/144518
- https://note.com/npaka/n/n401dccfadedc
- https://note.com/npaka/n/n0ad63134fbe2
- https://zenn.dev/karaage0703/articles/2b753b4dc26471
- https://zenn.dev/karaage0703/articles/d58d79d8e77ab8
- https://zenn.dev/karaage0703/articles/d3893b551c68fa
- https://note.com/npaka/n/nfacbeb1ae709
- https://colab.research.google.com/github/mkshing/notebooks/blob/main/stabilityai_japanese_stablelm_alpha_7b.ipynb
- https://gitlab.com/nvidia/container-images/cuda/blob/master/doc/supported-tags.md
- https://note.com/npaka/n/nd05b0334cb75
- https://zenn.dev/tsuzukia/articles/f886a93fc1f2fa
- https://note.com/npaka/n/nbb94b45f47a5
- https://huggingface.co/google/gemma-7b