nbox 0.9.6a0

ML Inference 🥶

🏖️ Nbox

pypi nbox is stable for SSH and other APIs. Check out v1 for the latest code and APIs!

A library that makes using a host of models provided by the opensource community a lot more easier.

The entire purpose of this package is to make using models 🥶.

pip install nbox

Current LoC

SLOC	Directory	SLOC-by-Language (Sorted)
996     top_dir         python=996
88      framework       python=88

Totals grouped by language (dominant language first):
python:        1084 (100.00%)

🔥 Usage

import nbox

# As all these models come from the popular frameworks you use such as 
# torchvision, efficient_pytorch or hf.transformers
model = nbox.load("torchvision/mobilenetv2", pretrained = True)

# nbox makes inference the priority so you can use it
# pass it a list for batch inference 
out_single = model('cat.jpg')
out = model([Image.open('cat.jpg'), np.array(Image.open('cat.jpg'))])
tuple(out.shape) == (2, 1000)

# deploy the model to a managed kubernetes cluster on NimbleBox.ai
url_endpoint, nbx_api_key = model.deploy()

# or load a cloud infer model and use seamlessly
model = nbox.load(
  model_key_or_url = url_endpoint,
  nbx_api_key = nbx_api_key,
  category = "image"
)

# Deja-Vu!
out_single = model('cat.jpg')
out = model([Image.open('cat.jpg'), np.array(Image.open('cat.jpg'))])
tuple(out.shape) == (2, 1000)

⚙️ CLI

Just add this to your dockerfile or github actions.

NBX_AUTH=1 python -m nbox deploy --model_path=my/model.onnx --deployment_type="nbox"

# or for more details

python -m nbox --help

✏️ Things for Repo

• Using poetry for proper package management as @cshubhamrao says.
  • Add new packages with poetry add <name>. Do not add torch, tensorflow and others, useless burden to manage those.
  • When pushing to pypi just do poetry build && poetry publish this manages all the things around
• Install pytest and then run pytest tests/ -v.
• Using black for formatting, VSCode to the moon.
• To make the docs:

  # from current folder
  sphinx-apidoc -o docs/source/ ./nbox/ -M -e
  cd docs && make html
  cd ../build/html && python3 -m http.server 80
  

🧩 License

The code in thist repo is licensed as BSD 3-Clause. Please check for individual repositories for licenses.