OiO.lk Community platform!

Oio.lk is an excellent forum for developers, providing a wide range of resources, discussions, and support for those in the developer community. Join oio.lk today to connect with like-minded professionals, share insights, and stay updated on the latest trends and technologies in the development field.
  You need to log in or register to access the solved answers to this problem.
  • You have reached the maximum number of guest views allowed
  • Please register below to remove this limitation

Nvidia Cudatoolkit vs Conda Cudatoolkit

  • Thread starter Thread starter Rajdeep Dutta
  • Start date Start date
R

Rajdeep Dutta

Guest
Till date, I have been using Tensorflow-GPU by installing it using pip and the Cuda related software and Nvidia softwares/drivers from Nvidia's website. Recently, I found that using conda install tensorflow-gpu also installs cudatoolkit and cudnn.

So, how are these(the ones provided by conda) different from the ones that I downloaded from Nvidia's website?

In my first (previous) environment, conda list showed that I have installed only TensorFlow(from PyPi) and no cudnn/cudatoolkit, but still everything worked.  This is my base environment where I installed Tensorflow-GPU using pip

Also, in a new environment in which I ran conda install tensorflow-gpu, conda list showed me tensorflow-gpu has been installed along with cudatoolkit and cudnn by Anaconda. And in this environment also, everything worked fine.This is a new environment where I used conda to install tensorflow-gpu

So does this mean, that downloading and installing Cuda from Nvidia's website is only necessary if I use pip to install TensorFlow?
<p>Till date, I have been using <strong>Tensorflow-GPU</strong> by installing it using <strong>pip</strong> and the Cuda related software and Nvidia softwares/drivers from <strong>Nvidia's website</strong>. Recently, I found that using <code>conda install tensorflow-gpu</code> also installs <strong>cudatoolkit</strong> and <strong>cudnn</strong>. </p>

<p>So, how are these(the ones provided by conda) different from the ones that I downloaded from Nvidia's website?</p>

<p>In my first (previous) environment, <code>conda list</code> showed that I have installed only TensorFlow(from PyPi) and no cudnn/cudatoolkit, but still everything worked. <a href="https://i.sstatic.net/DVAIZ.png" rel="noreferrer"><img src="https://i.sstatic.net/DVAIZ.png" alt=" This is my base environment where I installed Tensorflow-GPU using pip"></a></p>

<p>Also, in a new environment in which I ran <code>conda install tensorflow-gpu</code>, <code>conda list</code> showed me <strong>tensorflow-gpu</strong> has been installed along with <strong>cudatoolkit</strong> and <strong>cudnn</strong> by <strong>Anaconda</strong>. And in this environment also, everything worked fine.<a href="https://i.sstatic.net/49OaA.png" rel="noreferrer"><img src="https://i.sstatic.net/49OaA.png" alt="This is a new environment where I used conda to install tensorflow-gpu"></a> </p>

<p>So does this mean, that downloading and installing Cuda from Nvidia's website is <strong>only</strong> necessary if I use pip to install TensorFlow?</p>
 

Online statistics

Members online
0
Guests online
4
Total visitors
4
Top