ann from the Spell team We’d excited to introduce our cloud GPU offering for fast.ai students.
We spent some time getting the fast.ai notebooks to work easily with Spell. We’re fast.ai learners ourselves and wanted to make sure to make it super simple for learners to get up and running with the fast.ai notebooks.
We’re really proud to offer what we think is a simple, fast, and reliable cloud GPU service that’ll keep up with your machine learning needs.
To run the notebooks, sign up for a Spell account and install the CLI tool in terminal. You’ll automatically activate your $100 credit once you sign up and enter your billing information. We email you when you have $20 credit left (and you can set additional custom notifications) so you’ll never be hit with unexpected charges.
To get started, just run the following commands in terminal (more detail in our docs):
$ pip install spell $ spell login
Then grab the course materials and start your fast.ai Jupyter notebook:
$ git clone https://github.com/fastai/fastai.git && cd fastai $ spell jupyter --machine-type K80 \ --conda-env fastai \ --mount public/tutorial/fast.ai:data
Jupyter will be running locally, but be wired to a kernel with a K80 and the fast.ai conda environment loaded. Once Jupyter starts, switch to the “Spell - fastai on K80” kernel by clicking “Change kernel” under the Kernel menu (you only have to do this once).
Remember to stop your session (
^C ^C in terminal) so you don’t get charged for an open notebook you’re not actively using!
Spell is a command line tool for super simple remote execution - like the bash & operator, but for sending work off to a cloud instance. It’s great for developing for deep learning locally but seamlessly running on a cloud GPU.
Questions/feedback welcome - we’ll be hanging out here and in slack!