Distributed Training On NVIDIA DGX Station A100 | Deep Learning Tutorial 43 (Tensorflow & Python)

Using tensorflow mirrored strategy we will perform distributed training on NVIDIA DGX Station A100 System. Distributed training is used to split the training workload on different GPUs on a multi GPU system. We will see how performance can be optimized and training times can be reduced using this approach.

🔖Hashtags🔖

#deeplearningmultigpu #deeplearninggpusetup #tensorflowdistributedtraining #tensorflowmirroredstratergy #distributedtraining #dgxa100 #nvidiadgxa100

#️⃣ Social Media #️⃣

❗❗ DISCLAIMER: All opinions expressed in this video are of my own and not that of my employers'.

[jayantbhatia]

your videos doesn't need any background music they are already awesome

[sachavanweeren]

Thank you very much. In my view this is one of the best explained and most complete series of videos on tensorflow ....

[HashanDananjaya]

DGX must be amazing!! waiting for the TensorFlow data pipeline video ...

[ajoynambiar]

Can you show what's the best way to run ML on distributed computing on cloud for beginners. I am not able to run my code in memory.

[saurabharbal]

Thanh You Very Much for all this efforts. I have some issues here how to use same on cloud?
Please help me.

[hanslanda]

The a100s i have Are only visible in tensorflow when i have a mig instance, but then i can only use one. What can i do to use all gpus in tensorflow without mig? Its Running on a ubuntu server

[minhducvu]

We are waiting for the GANs explanations!!!

[dulangikanchana]

hi can you do a video on what is feature space?

[TuntaiBuri]

The IP address of the notebook is which machine (DGX) ?

[salikmalik]

Sir, continue you maths series for data science plz

[maloukemallouke]

Hi, thank's for video .
How much cost the Nvide stations like your's please ?

[imranmehraj]

sir we have here AI - DGX A100 and we are facing frequent shutdown(like sleep mode) of this server and we have to manually restart it again and again, we contact NVidia as well they send some service engineers but its happening again. we have good cooling system for it and there is not too much load because its being used by the research scholars. we have also checked power supply problem but we are finding any solution to fix it. please if you can suggest some tips to it that will most kind of you.

[architaray]

Can upload my dataset here? I uploaded zip files but could not unzip it.

[shafagh_projects]

thank you so much for your great tutorials. However, Can you provide a tutorial on TPU (Tensor Processing Unit) and how it works, and its competency with CPU and GPU?
I know I can google it but would like to know the opinion of experts like you.

[blasttrash]

hey just googled that nvidia a100 dgx station price is almost $150, 000. How did you buy it? Did you spend your own money on it? just curious, or did you rent it somehow?

[kunalroy]

How many videos are left to be premeired in this deep learning series??

[miholeus]

Where did you buy dgx? What's the price for it?

[punnarahul]

want some more practice exercises on ml

[shubhamsuryawanshi]

Sir i need a video on your journey ..
Pls
I m doing data scientist course

[divyasingh]

I came across Walmart coding challenge, please make a video on it.