How do I get my script in python to use the GPU on google colab?

Question

I know how to activate the GPU in the runtime type, but I'm used to doing machine learning with sklearn or XGBoost which automatically make use of the GPU. Now I've made my own machine learning algorithm but I don't know how to force it do the computations on the GPU. I need the extra RAM from the GPU runtime type, but I don't know how to benefit from the speed of the GPU...

@jit(target ="cuda")
popsize = 1000
  
File "<ipython-input-82-7cb543a75250>", line 2
    popsize = 1000
          ^
SyntaxError: invalid syntax

Just note that scikit-learn does not have any kind of GPU support, only XGBoost has it. — Dr. Snoopy
– Dr. Snoopy, Commented Jun 23, 2020 at 13:09
@Dr.Snoopy is right. The program must be written and Executed with TensorFlow or any other Framework that supports GPU usage. — Ishan Dixit
– Ishan Dixit, Commented Jun 23, 2020 at 13:10

Ishan Dixit · Accepted Answer · 2020-06-23 13:02:24Z

1

As you can see here Numba and Jit are ways to put your scripts on GPU like follows:

from numba import jit, cuda 
import numpy as np 
# to measure exec time 
from timeit import default_timer as timer 

# normal function to run on cpu 
def func(a):                                 
    for i in range(10000000): 
        a[i]+= 1    

# function optimized to run on gpu 
@jit(target ="cuda")                         
def func2(a): 
    for i in range(10000000): 
        a[i]+= 1
if __name__=="__main__": 
    n = 10000000                            
    a = np.ones(n, dtype = np.float64) 
    b = np.ones(n, dtype = np.float32) 
    
    start = timer() 
    func(a) 
    print("without GPU:", timer()-start)     
    
    start = timer() 
    func2(a) 
    print("with GPU:", timer()-start)

There is one more reference link you can utilize

answered Jun 23, 2020 at 13:02

Ishan Dixit

3991 gold badge3 silver badges11 bronze badges

Sign up to request clarification or add additional context in comments.

7 Comments

Ishan Dixit Over a year ago

Also, Installing Tensorflow(GPU version) will ensure the use of GPU.

xxanissrxx Over a year ago

Does it matter where I insert the @jit(target = "cuda")? Or is just any code after that line is sent to GPU, and anything before it is sent to CPU?

Ishan Dixit Over a year ago

Yes. As it is a Python Decorator.

xxanissrxx Over a year ago

it doesn, it says invalid syntax on the line after the @jit(target = "cuda")

Ishan Dixit Over a year ago

Could you post the entire traceback.?

|

Minakshi Boruah · Accepted Answer · 2023-10-30 14:52:02Z

0

Now its target_background as target is deprecated.

answered Oct 30, 2023 at 14:52

Minakshi Boruah

53 bronze badges

1 Comment

Péter Szilvási Over a year ago

This does not provide an answer to the question. Once you have sufficient reputation you will be able to comment on any post; instead, provide answers that don't require clarification from the asker. - From Review

Collectives™ on Stack Overflow

How do I get my script in python to use the GPU on google colab?

2 Answers 2

7 Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

7 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related