Это видео недоступно.
Сожалеем об этом.
how to run keras and tensorflow on gpu on windows, step by step
HTML-код
- Опубликовано: 6 май 2023
- In this video, I will explain, step by step, how to run your keras or tensorflow model on GPU on Windows.
Tensorflow Installation Page:
www.tensorflow...
Supported NVIDIA GPUs:
developer.nvid...
Compatible Python and Package Versions:
www.tensorflow...
Working with Anaconda:
• python tutorials 1 - g...
One-Time Contribution Through PayPal:
www.paypal.com...
Country Cue 1 by Audionautix is licensed under a Creative Commons Attribution 4.0 license. creativecommon...
Artist: audionautix.com/
U r just God bro after 1 year finally can run on gpu
Sameee
wtf?
Sir i have a gaming pc but in terms of ml its been useless till now. Tried many tutorials but any thing didnt work. You made it very simple and easy to understand. Thank you!!
Glad I could help
I have been trying for days to find a way to use GPU for my neural networks project and your tutorial seems to be the only one that worked. Good job. I give a like
Wonderful. Thanks a lot, brother. It works for me. Now a great relief. I search a lot on RUclips and probably your way of installing is the easiest and most effective. Thanks.
Glad it helped
The only video that worked, thanks a lot!!
Thanks, this worked for me. Heads up though if anyone is following this; if you are using additional libraries like i did they also have to be installed into the new conda environment. I have some image processing etc in the same file so I had to go back and install it all into the new environment. Took me a while to realize how it all worked.
It works so well, thank you so much! After trying many solutions this is the one that finally works.
You're welcome!
I have been trying to use my gpu on tensorflow for over 2 years but every tutorial always presented some sort of unfixable error. I'm grateful that yours was not the case.
Thank you so much Johnny, I really appreciate it. 🤝
Thank you very much brother. This tutorial made me use my laptop gpu effectively for performing ML tasks. It works efficiently. Easy explanation throughout the video. Thanks once again!!
thanks so much after wasting more than a week i found you johnny LORD
brdr I actually don't know how to thank you, it is finallly working
You're magic! Thanks a lot bro, I have lost a lot of time trying to solve this.
Happy to help
Really great videos! Much more better than guys that have thousand of views ... i really hope you'll get much more views cause your videos are really excellent! (i mean it)
Wow, thank you! I really appreciate it.
Great Job! Answered all the key questions!
Thanks Johhny waited 8 month for gpu thanks
Thank you very much for this tutorial!
You are an actual life saver
That's an excelent tutorial, congrats!
Thank you for sharing! It helped me a lot!
thank you so much. finally my gpu is working
Glad it helped
Are you using anaconda ?
Thank you very much for your tutorial !
Thank you sooo much mam it was very helpful and it worked for me
Glad to hear that
Like your videos my friend, just one compaint could you put links in description? Other than that great content
Great feedback! I've added the links in the description. Thanks!
Can you do videos on OOP? and more on machine learning (CNN, RNN, LSTM, reinforcement learning ...) , i'm asking a lot, i know, but your videos goes straight to the point no bla bla , keep up the great work!
Hi, right now I am focusing on a series of video on natural language processing with Harry Potter, which eventually will cover RNN, LSTM, and even transformer. I am thinking about working on another video series on deep learning, which will start from the basics, such as logistic regression, gradient descent...etc. This series will eventually cover CNN and other topics, such as reinforcement learning, autoencoder...etc. About OOP, whenever I have time, I will continue on my series on python tutorial, which will eventually cover python classes.
@@johnnythedatascientist538 Great, i'm waiting for it!!!(i have the basics, i've done all kind of simple models, random forest, knn, svm, also perceptron) i'm looking for CNN, autoencoder, LSTM, and especially reinforcement learning (i would like to mitigate reinforcement learning and random forest as they did for alphago) , i've heard about transformer too, seems great subject ... So much things to learn, 😊
You're a hero!
Very helpfull sir. Thank you so much.. Arigato..
Thank you !!!
thank u so much
Thanks a lot! This installation method seems to work fine in Jupyter Notebook, but when I work in my VS Code (Jupyter Notebook environment) it still doesn't detect detect my GPU. Any idea how to make GPU works in VS Code as well?
how to do this without joppitter and gonda?
Mine is GeFORCE GTX 1660 Ti but it's not there to download. There is GeFORCE GTX 1650 Ti. What to do?
Anaconda needs me to manually download python libraries and literally couldnt launch jupyter when i used this, so yeah was quite troublesome
if I run Jupyter through Anaconda it works but if I install Tensor Flow and again run Jupyter Notebook through the specified environment it shows me an error.Please help
I fixed this.Go to the anaconda prompt after giving
pip install tensorflow=2.10
then in terminal give this line
conda install -c conda-forge chardet
all thing will run
thanks
Thank you so much.
Subscribed !!!
Thanks!
how to connect this with pycharm
i have 1660 super but it is not avalible?
One important point. WSL2 is limited to 64GB RAM. If you have more, you won't use it
Thanks for sharing.
oh i got a question, in the config.physical_devices it prints out gpu 0 is that mean i am using my gpu 0(the intel one) because my gpu 0 heat does not rise so much but my gpu (the nvidia one)rises as hell. Does this mean i am using the nvidia one??
It must be the nvidia one. I don't think intel gpu is cuda supported.
@@johnnythedatascientist538 thanks bro, i didn't get the announce of your replied.
Your vid help me a lot
@@johnnythedatascientist538 but i got a problem, i'm using spyder to train a model(deep learning with keras) but i got out of memory bug a lot, but my gpu only run 2.5/4.0 GB whats going on and how can i fix it??
After completing the process if i try to install tensorflowit pops module not found error what should i do
I am not sure why it is happening to you. Make sure you are importing tensorflow under the same environment that you installed the tensorflow package.
please it tells me that No matching distribution found for tensorflow==2.10
seems like pip can't install this version any more!
what should i do????
Hi, I've just installed tensorflow==2.10 on a brand new conda environment and had no issues. Not sure what's going on on your end. However, I found this stack overflow page that might be able to help you. stackoverflow.com/questions/48390095/no-matching-distribution-found-for-tensorflow
@@johnnythedatascientist538 Hi, what is the python version that you are using to install tensorflow==2.10?
which version should be used for Geforce MX350
Hi, I don't think GeForce MX350 is cuda-enabled. I don't see it in the Nvidia's cuda-gpus page.
for me 3 rd command for cuda and cuddn is not working ,can u help me??i have a gpu rtx 3060
Do you have any error messages? Also make sure there are no spaces before and after the equal signs.
i followed all the steps but still was not able to see gpu in the list of devices
can you please help
what graphic card do you have?
11:35 launching jupyter notebook gives me error
I fixed this.Go to the anaconda prompt after giving
pip install tensorflow=2.10
then in terminal give this line
conda install -c conda-forge chardet
all thing will run
thanks
So no need of VISUAL STUDIO installations ??
I don't think you need Visual Studio.
If I do this will it also affect tensor flow on vs code?
Hi, after installing all the required packages with correct versions, you will just need to select the right kernel/environment in vs code. Of course, you will also need to verify that your GPU is cuda-enabled.
can i use mx 450 training small models?
unfortunately, mx450 is not cuda enabled.
Saved me !
Aconda is so trash several packages not existing
where do i donate, i literally jumped up and down like a child
Hi, I am really glad that my video is helping you. I finally set up a PayPal account to receive donations, but please treat this video as complimentary. You can make donations when you find my future videos are helping you again next time.
rtx 90080 ?
sorry... that was a mistake. It should be GTX980. Good catch!
Hey man, thanks for this amazing video. I have an error just after the step of installing tensorflow for that environment.
The error is
"Unable to create process using 'C:\Users\Abdul Basit\.conda\envs\tensorflowgpu\python.exe "C:\Users\Abdul Basit\.conda\envs\tensorflowgpu\Scripts\pip-script.py" install tensorflow==2.10'"