Run Deep Learning Notebooks on Free GPUs !

Original article can be found here (source): Deep Learning on Medium

Run Deep Learning Notebooks on Free GPUs !

AI communities like Google Colab and Paperspace are providing free GPUs ,CPUs and TPUs

Google Colab is providing free TPUs,GPUs and CPU service but PaperSpace is providing free GPUs and CPus. With the announcement of Gradient Community Notebooks, a free cloud GPU service based on Jupyter notebooks that radically simplifies the process of ML/AI development. Now, any developer working with popular deep learning frameworks such as PyTorch, TensorFlow, and Keras, can easily launch powerful free GPU instances and collaborate on their ML projects.

Launch a Free Notebook

Gradient Community Notebooks are public, shareable Jupyter Notebooks that run on free cloud GPUs and CPUs. Notebooks can be run on any DL or ML framework, pre-configured to work out of the box. Use your own container or choose from a wide selection of templates complete with popular drivers and dependencies, like CUDA and cuDNN.

Gradient let you focus on building your models, not troubleshooting your environment.

Instance types available for the Free Notebooks include:

  • Free-CPU — C3 CPU instance
  • Free-GPU+ — NVIDIA M4000 GPU
  • Free-P5000 — NVIDIA P5000 GPU

How to create Free GPU Notebook and run

Getting started with your first Free GPU Notebook is incredibly easy.

  1. Select your instance (M4000 or P5000 cloud GPU, or C3 cloud CPU)
  2. Launch your own Notebook, or fork a ready-to-run project from the ML showcase or Community

You can run an unlimited number of sessions for up to 6 hours at a time. Your Notebook will remain fully versioned, and you can restart your instance to run for another 6 hours as many times as you like.

Gradient Community Notebooks are focused on just that — the Community. All notebooks are set to public, and can be shared and forked by other Community members. Sample GPU Note created below and can be shared to anyone

Now Data Scientists/Researchers no need to look for any clouds like AWS ,GCP,Azure etc for GPUs and TPUs.