I recently became curious about TPUs, a specialised hardware for training Machine- and Deep-Learning models, where TPU stands for Tensor Processing Unit. This fancy chip can provide very high gains for anyone aiming to perform really massive parallelisation of AI tasks such as training, fine-tuning, and inference.
In this blog post, I will touch on what a TPU is, why it could be useful for AI applications when compared to GPUs and briefly discuss associated opportunity costs.
OPIG recently celebrated its 20th year; and on 10 January 2023 I gave a talk just a day before the 10th anniversary of BLOPIG’s first blog post. It’s worth reflecting on what’s stayed the same and what’s changed since then.
Continue readingAs promised, I will tell you about another Linux Horror Story: The Nvidia driver automatic update that breaks your machine. This is a recurrent problem that I have suffered so many times that I tend to disable all Nvidia updates just to avoid it. Unfortunately, I forgot to do so on my new laptop, so it happened once more.
It all started when I tried to connect my dual monitor to my laptop, as I have been doing for the last 8 months. But the SO did not recognize the monitor. After unplugging and plugging my monitor a few times and rebooting my machine several times, I started thinking that it may be a drivers-related problem, so I just executed the command nvidia-smi to check if the GPU drivers were working. A familiar error message confirmed my fears:
NVIDIA-SMI has failed because it couldn’t communicate with the Nvidia driver.
Make sure that the latest NVIDIA driver is installed and running. If you are lucky enough, this is a consequence of the driver update and rebooting the machine will make it work again. Unfortunately, it was not my case, so I started the process of uninstalling and reinstalling the drivers. To do so, in an Ubuntu machine, you only need to use the following two commands.
Continue reading