How to Install TensorFlow GPU on Windows 11 | GPU Setup on Windows | TensorFlow GPU Setup
HTML-код
- Опубликовано: 5 окт 2024
- ✍️🏆🏅🎁🎊🎉✌️👌⭐⭐⭐⭐⭐
ENROLL in My Highest Rated Udemy Courses
to 🔑 Crack Data Science Interviews and Jobs
🏅🎁 Python for Machine Learning: A Step-by-Step Guide | Udemy
Course Link: bit.ly/ml-ds-p...
🎁🎊 Deep Learning for Beginners with Python
Course Link: bit.ly/dl-with...
📚 📗 Natural Language Processing ML Model Deployment at AWS
Course Link: bit.ly/bert_nlp
📊 📈 Data Visualization in Python Masterclass: Beginners to Pro
Course Link: bit.ly/udemy95...
📘 📙 Natural Language Processing (NLP) in Python for Beginners
Course Link: bit.ly/intro_nlp
🎉✌️ Advanced Natural Language and Image Processing Projects | Udemy
Course Link: bit.ly/kgptalk...
📈 📘 2021 Python for Linear Regression in Machine Learning
Course Link: bit.ly/regress...
📙📊 2021 R 4.0 Programming for Data Science || Beginners to Pro
Course Link: bit.ly/r4-ml
✍️🏆 Introduction to Spacy 3 for Natural Language Processing
Course Link: bit.ly/spacy-i...
How to Install TensorFlow GPU on Windows PC. You will learn how to install CUDA Toolkit and tensorflow GPU.
Install Anaconda and Jupyter Notebook on Windows 10/11 | Machine Learning with Anaconda and Jupyter
• Install Anaconda and J...
Install TensorFlow in Windows WSL2
• How to Install TensorF...
** TensorFlow 2 Setup Link **
www.tensorflow...
** INSTALL **
1. conda create -n py310 python=3.10
2. conda activate py310
3. conda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0
4. python -m pip install "tensorflow=2.10"
** Test GPU **
import tensorflow as tf
tf.config.list_physical_devices('GPU')
tf.test.is_gpu_available()
==================================
🔊 Watch till last for a detailed description
💯 Read Full Blog with Code
kgptalkie.com
💬 Leave your comments and doubts in the comment section
📌 Save this channel and video for watch later
👍 Like this video to show your support and love ❤️
~~~~~~~~
🆓 Watch My Top Free Data Science Videos
👉🏻 Python for Data Scientist
bit.ly/3dETtFb
👉🏻 Machine Learning for Beginners
bit.ly/2WOVh7N
👉🏻 Feature Selection in Machine Learning
bit.ly/2YW6ZQH
👉🏻 Text Preprocessing and Mining for NLP
bit.ly/31sYMUN
👉🏻 Natural Language Processing (NLP)
Tutorials bit.ly/3dF1cTL
👉🏻 Deep Learning with TensorFlow 2.0
and Keras bit.ly/3dFl09G
👉🏻 COVID 19 Data Analysis and Visualization
Masterclass bit.ly/31vNC1U
👉🏻 Machine Learning Model Deployment Using
Flask at AWS bit.ly/3b1svaD
👉🏻 Make Your Own Automated Email Marketing
Software in Python bit.ly/2QqLaDy
***********
🤝 BE MY FRIEND
🌍 Check Out ML Blogs: kgptalkie.com
🐦Add me on Twitter: / laxmimerit
📄 Follow me on GitHub: github.com/lax...
📕 Add me on Facebook: / kgptalkie
💼 Add me on LinkedIn: / laxmimerit
👉🏻 Complete Udemy Courses: bit.ly/32taBK2
⚡ Check out my Recent Videos: bit.ly/3ldnbWm
🔔 Subscribe me for Free Videos: bit.ly/34wN6T6
🤑 Get in touch for Promotion: info@kgptalkie.com
I need to thank you financially.
I wasted 4 full days looking for tutorials on RUclips.
None worked, just yours!
Thanks
🤣😂🤣same 😭😭 wasted 4 days
i wasted 5 days to connect my GPU with tf , from your video it took me few minutes to get the job done .salute
thank you,
more than 1 week of trying and failing, your tutorial was simple, direct and effective
Thanks for watching
"I can't express how relieved I am right now. I've been wrestling with this problem for days, burning the midnight oil, and it seemed insurmountable. But thanks to your guidance, the fog has lifted, and my problem is finally solved. Your help has been a game-changer for me!"
Thank you Sir,
You are welcome! Please watch my other videos as well.
Wow. They made it a nightmare for Windows users.
Agreed 👍
Just as annoying for Linux as I’m running into issues installing CUDA.
@@KGPTalkie
how can you install Tensorflow inside the Python 🐍 console inside Paraview, Salome or FreeCAD?
Is there a way to install AutoGen AI agents 🤖?
just wonderful, I tried to get this up and running 2 years ago, wasted lot of time and gave up, today i decided to give it a shot by following your instructions and it WORKED!!! You are Amazing.
This is the only tutorial so far that's simple but effective compared to other videos that solves my month-long problem. Thank you for this sir
Thank you so much! It worked. I spent the whole day searching for a solution and no one worked until I reached yours!
Bro......thank you sooo much.... ive been trying to do it for days ... i watched a lot of videos and none worked... this is simple and its working bro.. Thank You
you are a hero, I was struggling for 2 months to fix this issue
Thanks for watching
I usually don't comment on ytb video but this one is exceptional. This video ends my 13 consecutive working hours. You truly are my savior! Thank you kind sir!!!
You are better than Chat GPT at this!
😍
@@KGPTalkieAbsolutely ❤️❤️
I swear to god! Thank you very very much! I hate setting this up. Always struggling. Tested so many variations. WSL2, native, conda, docker... All sht.... always cpu or failed.
But your specific combination hit the brick for me!
Sad it is not working for the newest stuff, but ok. Its new enough. I killed all my IDE's and everything on my pc to get it running. With your tutorial and 10minutes of time I have managed to pull it over!
Thank you very very much! You should defnitely get a donator link in the description!
i am going to join the ppl here who are eternally grateful for this. I've spend days! installing CUDA toolkits, cuDNN libraries, downgrading, setting env variables, create thing difference mixes on several conda env... nothing worked until this!. THANKS A LOT!!!!!
Thanks a lot for appreciation 😍
I have been trying for months to fix this problem. Thank you for this clear and concise tutorial! It worked on my end. (GTX)
to bypass environement error while doing :
python -m pip install "tensorflow
thanks! this works for me! only extra step for me is to uninstall numpy and reinstall numpy 2.0.0
how do do that? i tried, but i have error : module compiled using Numpy 1.0 cannot be run in Numpy 2.1... non of these tensorflow with gpu videos work for me... what a waste of time
@@realracing3specter295 I also faced. It's simple execute "python -m pip uninstall numpy" and then 'python -m pip install "numpy
Hi sir, i am so grateful and thank you for your video. i followed lot of videos and pages, only thing that helped me is your video. i cannot believe it worked. i installed cudnn, cuda and visual studio etc and messed up in lap memory. Thank you so much for your amazing video.
Your welcome ✨
It's 3 am and I just found you. Thank you so much.
thank you alot, i wasted many hours and finally you saved me, good work, i don't subscribe to any youtube channel, in months, you are the first channel to subscribe. thanks again
Thanks ❤️
Hello, I've encountered a problem following your tutorial. I followed everything but at the end when I tried to import tensorflow in jupyter notebook, I got an error about how I'm trying to use numpy 1.x instead of 2.x or something like that, which is confusing as I'm new to this whole thing, but my guess is that since you have put this on youtube, there has been some changes to anaconda distribution.
did you got the solution?
uninstall numpy first, then: pip install numpy==1.26.4
@@anushkakeshri4436 uninstall numpy first, then: pip install numpy==1.26.4
@@4d_sculpture It worked! Thanks!
@@4d_sculpturethanks
thanks, I was cussing the fuck out everyone with lame blogs and poor youtube videos. But you my friend , you are a savior. thanks a ton man.
Btw are you from IIT KGP ?
Thanks
Yes
@@KGPTalkie Bro I have installed as you said in the video but while training my ANN model it's still using CPU and not my Nvidia GPU ( i'm running the ipynb file on the py310 environment only) Any more steps required ?
Like I guess I'm missing some in between steps. Like after creating py310 environment and doing all you mentioned , i got my nvidia gpu name printed in the command prompt.
Now how to actually use this new enviroment in jupyter is my new challenge.
Like when i print ( in jupyter notebook which has model code)
import tensorflow as tf
tf.config.list_physical_devices('GPU')
tf.test.is_gpu_available()
it's not shwoing anything just empy list
Thank you very much! , short and informative cuda is now working on my computer
Really Really thank you so much !!!!
from last 2 weeks I have been trying to install it for gpu
followed lots of videos then finally tried with your instructions and its working !!!!!!!!!!!
Subscribed 😁😁😁
same :(
Thank my you so much I’ve been struggling with this for months and your method worked
i really want to thank you from the bottom of my heart very nice and simple video thank you again
Thanks a lot for watching this
ty bro, was a hard journey to achieve that for me
All the best 🎉
Heartfelt gratitude to you. Thank you. Question: do I need to install keras separately to use it?
God bless you man. you save me. i spent months for this .
So helpfulll thankyouuuu very much liked the video and subscribed to channel
IT WORKS!!!! GOODJOB!!!!
Amazing video brother thanks.🎆🔥
you are great! now is working!! Question have you tried with the other option using that WLS with a Linux distribution?
6 months prior i just droped the idea of ever ever using local gpu for dL because other tutorials just confused me..... I am that I came across your Channel
Thanks a lot for guiding properly
Great crystal clear, Thank You!!
Thanks
Thanks you so much. It worked for me.
bro thank you so much i have been trying to do this for a long time , thank you so much
Sir it is working for tensorflow 2.10.1 and 2.10 but when i testing the model for installation then math_ops import error facing and it says you need to upgrade tensorflow version but if i upgrade then gpu is not detected what should i do for tensorflow 2.16.1 and 2.17??
Finally this work for me, though I still got 500 Internal Server Error when I open Jupyter from the Anaconda env (tried re-installing but still got the same problem, I use VSC now).
I have Nvidia GTX 1650 Mobile, and I followed all the instruction except step 4 where I use tensorflow-gpu==2.9 instead of what shown there. I also matched the python version, cuDNN, and CUDA with the web version (tf-gpu 2.9.0 match with cuDNN 8.1, CUDA 11.2, and I use Python 3.7 instead of 3.10 shown in this video). If I follow the instruction as it is, I got tensorflow version not matching with the version that can utilize GPU.. Thank you.
Hey, needed a bit of help, so after doing the steps in the video, how will i use it in jupyter lab. I had installed tensorflow latest version before and it did not detect gpu, now after downloading the 2.10 version, in this new environment, how do i make my jupyter lab use this tensorflow and not the previous one.
I have been working on this for more that 1 whole week...I cant thank you enough for this man..!! { for everyone facing issues with tf not detecting GPU.. its the damn tensorflow version that is the issue!. *TensorFlow 2.10 was the last TensorFlow release that supported GPU on native-Windows*, Download CUDA and cuDNN respectively.
its my 4th day, still struggling, (unlike countless other videos) even this video does not work for me... tell me, do i have to download lower versions of cuda and cudnn as well? after considering tensorflow 2.10?
@realracing3specter295 did you use python 3.9 or less? Otherwise it won't work.
@@KGPTalkie i have same python version as in video : python 3.10.4 and now tensorflow works, but it cant locate cudnn files, and shows cant locate dll file errors. And thus show null value in detecting gpu.
@@realracing3specter295 so is your error solved cuz currently i'm also facing this issue
@@yashkancharwale8350 i finally installed the same version of cuda toolkits and cudnn as in this video, and it worked. 😇 i didnt know i had to match all those packages with their compatible versions as well.
Thanks a lot, this is really great video which solved my problem.
Great. Thanks for watching
Why don't you use WLS2. Its really great to use instead of sticking with older version of TF.
sadly wsl method does not work with my computer.
@IamNishanKhan what is the issue?
GOD. FINALLY. THANK YOU
Welcome 🥂
YOU ARE AMAZING!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
IT WORKS
it worked for me. thanks for this easy step . 3rd command to install tensoflow is python -m pip install "tensorflow
then what to do?
@@ankush9056 use this one please : python -m pip install "tensorflow
straight forward. Thanks!
It shows true in command prompt but shows false in jupyter notebook. I have opened notebook in py310 environment
Change the environment in anaconda to py310 and then open Jupiter. You are running base root environment
@@rajandeepsinghbhullar4894 Thank you my friend
Indian guy on IT, you can trust!!!
ty, straight to the point
bro really tq for the useful video
Thanks a ton bhai😀
Thank you so much for this video brother
Thanks a lot bro u r a life saviour
hello mate thank you for the video i managed to do all you have said but now how can i use that environment in spyder ? I tried conda activate py310 and then typed spyder, but there my gpu is not detected (it was detected in anaconda prompt)
same problem for me
Muchas gracias, espere meses por un video asi😊
Wooh!! Man, You Saved me dude, Thank you so much....
i cant thank you enough for this
thank you so much. saved my life.
Thanks, helped me a lot.
If you get >TimeoutError: The read operation timed out< when running >python -m pip install "tensorflow
Thank you man .. really great full
in the anaconda prompt it shows gpu, but when i do the same in jupyter it shows False, HELP!!!!!!
ok i figured it out i had to do pip install jupyter after activating the python
Working. Got the following error
Traceback (most recent call last):
File "C:\Users\ankit\anaconda3\envs\venv2\lib\site-packages
equests\compat.py", line 11, in
import chardet
ModuleNotFoundError: No module named 'chardet'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
blah blah blah
This issue gets resolved by the following steps
pip install chardet
pip install charset_normalizer
pip uninstall requests
pip install requests
Worked post that
It works , thank you very much
Please watch other videos as well
Hi and thank you. Bu how to install it on Windows WSL?
The only solutions I have seen require a virtual environment like anaconda or setenv.
I prefer not to use those virtual environments.
Bro is it necessary to install different CUDA and cuDNN for tensorflow and torch ?
thnak you sir
the commands you wrote in description says "tensorflow=2.10" which wont work , it should be
Getting error in installing tensorflow what to do
Important Message: This method shown in the video will work, if and only if you already set the environment variables properly. Otherwise, without setting the environment variable, if anyone just follows the video, it will not work. I tested and failed. Please watch another video to see details, on how to set Env. Var. After setting env. var. I followed the video and it works. Thanks.
I followed same thing but GPU is not active. Although NVIDIA GPU is present in my PC.
CUDA toolkit version 11.2 is only available for windows 10 so is it ok to install in Windows 11?
Thank You So Much ❤
Thank you so much, bro. 😁
python -m pip install "tensorflow
try installing tensorflow 2.10.0
its becuase your python version is not the correct one did you do what he did?
Thank you very much! 🙂
Your welcome
thanks a lot boss
Thanks for the video but now launching jupyter notebook it says application notebook may have produced error.
Please help this out. Replying with the error text file
Regards
My anaconda is installed in C drive... But whenever I use Jupiter notebook I open it in D drive by using a command. Does this virtual environment is accessible in D drive ?
Thanks alot
Thank you so much bro :)
can't we install Cuda tensorflow without an Anaconda env?
i have hp ci7 laptop with geforce rtx2050 graphic card, i followed the steps but it still not available. help please?
thanks, helpfully,
can i install without conda and custom for cuda and cudnn version?
How to use these for the latest python 3.11 version
but i want to use it in vs code how can i do that?
Bro my conda env always shows "failed at initial attempt retrying with flexible solve" What can i do?
Will it install the package despite of that message?
thank you soooo much bro
You're welcome
thank you very much
Thank you so much sir...
hermano muchas gracias
lo intenté con otros videos y de otras formas y no lo lograba hacer
Can you help me
I have nvdia 940mx
I have install the toolkit manually 11.2
and cudnn=8.0.1
and paste the files
I have environment with 3.9
tensorflow 2.5.0
but as soon as I list the command for GPU its does not get any
bro mine doest have a anaconda prompt . How do i download it ?
hey finally it worked
Cool
thank you
you are best
Thanks
thanks
thanks . I can't use 3080 gpus. any body can help me?