jbilcke-hf's picture
jbilcke-hf HF staff
added a clarification about which machine you need
36f130e
|
raw
history blame
2.1 kB
metadata
title: Template Node WizardCoder Express
emoji: 🧙
colorFrom: blue
colorTo: purple
sdk: docker
pinned: false
app_port: 7860

A minimalist Docker space to help people getting started with Node, WizardCoder (through CTransformers and Pythonia), Express and TypeScript. Ready to be used in a Hugging Face Space.

Installation

Prerequisites

CTransformers

This project relies on CTransformers called through Pythonia.

To install ctransformers:

pip install ctransformers
# or this, depending on your Python environment:
# pip3 install ctransformers

For GPU (CUDA) support set environment variable CT_CUBLAS=1 and install from source using:

CT_CUBLAS=1 pip install ctransformers --no-binary ctransformers
# or this, depending on your Python environment:
# CT_CUBLAS=1 pip3 install ctransformers --no-binary ctransformers

Building and run without Docker

nvm use
npm i
npm run start

Building and running with Docker

npm run docker

This script is a shortcut executing the following commands:

docker build -t template-node-wizardcoder-express .
docker run -it -p 7860:7860 template-node-wizardcoder-express

Attention! If you have a Mac, you may have trouble running the project on your machine.

You will see the following error message because Docker won't be able to use the pre-generated binaries for libctransformers.so due to architecture incompatibility:

🌉 OSError: /home/user/.local/lib/python3.11/site-packages/ctransformers/lib/avx2/libctransformers.so: cannot open shared object file: No such file or directory]

However if you run your project on a Hugging Face space, you should be just fine :)

Deployment to Hugging Face

The standard free CPU instance (16 Gb) will not be enough for WizardCoder, you should use the upgraded CPU instance (32 Gb)

I haven't upgraded mine yet, so it will probably crash when you try it: https://huggingface.co/spaces/jbilcke-hf/template-node-wizardcoder-express