16

one of the limitations is that we can get only 12 continuous hours per session. Is there any limitations for the usage for GPU and TPU?

Vineetha Vijayan
  • 188
  • 1
  • 1
  • 13
  • The limitations are in terms of RAM, GPU RAM and HBM, dependent on Google Colab hardware, at the moment is respectively ≈25GB, ≈12GB and ≈64GB. This will limit the dataset you can load in memory and the batch size in your training process. – Hichame Yessou Jan 17 '20 at 11:01
  • Hello! Can I upload my whole project to google colab and run it from there please or it only accepts notebooks? – Avv Nov 25 '21 at 20:03

2 Answers2

14

Yes, you can only use 1 GPU with a limited memory of 12GB and TPU has 64 GB High Bandwidth Mmeory.You can read here in this article. So, if you want to use large dataset then I would recommend you to use tf.data.Dataset for preparing it before training. If you want to use GPUs you can use any TF version. But for TPU I would recommend using TF1.14.

Rishabh Sahrawat
  • 2,437
  • 1
  • 15
  • 32
  • Is there any way to get the SIFT , SURF libraries in colab? – Vineetha Vijayan Jan 24 '20 at 10:35
  • 1
    I think you can use it directly from `cv2`. Check this Colab [file](https://colab.research.google.com/drive/1p7McfxQdXSD32KO3jPFigejiiTQjKivp). – Rishabh Sahrawat Jan 24 '20 at 10:42
  • Can I upload my project to Google colab please? I have a project with files and dataset, so how to load it there and run project? – Avv Nov 25 '21 at 20:03
  • @Avra Yes, you can: https://stackoverflow.com/questions/48376580/google-colab-how-to-read-data-from-my-google-drive – Dahn Dec 04 '21 at 09:30
5

From Colab's documentation,

In order to be able to offer computational resources for free, Colab needs to maintain the flexibility to adjust usage limits and hardware availability on the fly. Resources available in Colab vary over time to accommodate fluctuations in demand, as well as to accommodate overall growth and other factors.

In a nutshell, Colab has dynamic resource provisioning. So they can change the hardware, it it is being taxed too much automatically. Google giveth and Google taketh away.

Link

Anshuman Kumar
  • 464
  • 1
  • 6
  • 20