CODEWITHSUNDEEP
pythontimegoogle-colaboratorycellexecution
vinu
vinu

Reputation: 179

Is there a way to see the execution time and execution status in Google Colab?

I am trying to run the notebook in Google Colab. I am wondering if there is a way for me to know if the cell is run and how long it took to run the cell (in milliseconds).

Upvotes: 16

Views: 38875

Answers (5)

ZygD
ZygD

Reputation: 24478

%timeit is for one-liners. E.g., we can use several in one cell:

%timeit type(5)
%timeit -n 100 -r 3 type(5)

# 73.9 ns ± 19.5 ns per loop (mean ± std. dev. of 7 runs, 10000000 loops each)
# 90.7 ns ± 12.6 ns per loop (mean ± std. dev. of 3 runs, 100 loops each)

%%timeit is for the code in the whole cell:

%%timeit
rng = range(1000000)
sum([x for x in rng])

# 63.4 ms ± 5.82 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)

or (with some options provided)

%%timeit -n 100 -r 3
rng = range(1000000)
sum([x for x in rng])

# 62.6 ms ± 6.06 ms per loop (mean ± std. dev. of 3 runs, 100 loops each)

More options

Upvotes: 0

Bruce N.
Bruce N.

Reputation: 21

You can accumulate the time taken to run. Something simple but useful.

Example:

import time

class Timer:

  def __init__(self):
    self.start = time.perf_counter()

  def calcMsUntil(self):
    return (time.perf_counter() - self.start) * 1000

# Cell [A]
print('\nCell [A] - start counter')
timer = Timer()

# Cell [B]
print('\nCell [B] - something happens (1 SECOND)')
time.sleep(1)

# Cell [C]
print('\nCell [C]')
print("time until [C]:", timer.calcMsUntil())

# Cell [D]
print('\nCell [D] - something happens (2 SECONDS)')
time.sleep(2)

# Cell [E]
print('\nCell [E]')
print("time until [E]:", timer.calcMsUntil())

Result:

Cell [A] - start counter

Cell [B] - something happens (1 SECOND)

Cell [C]
time until [C]: 1001.7266389995712

Cell [D] - something happens (2 SECONDS)

Cell [E]
time until [E]: 3005.6352470001

Upvotes: 0

Harsha
Harsha

Reputation: 583

You can do this to get the running time of each cell (similar to the ExecuteTime extension for Jupyter Notebook):

!pip install ipython-autotime
%load_ext autotime

This is how it'll look after running the above code: enter image description here

Upvotes: 31

MD Mushfirat Mohaimin
MD Mushfirat Mohaimin

Reputation: 2066

You can check the execution time of absolutely anything by calculating the time taken

import time
t1 = time.perf_counter()

### Your code goes here ###

t2 = time.perf_counter()
print('time taken to run:',t2-t1)

Upvotes: 3

Saatvik Ramani
Saatvik Ramani

Reputation: 477

When you hover over the code block, you can see a sort of play button with a circular progression ring around while it's still running.

To track times, you can add %%timeit at the start of the execution block like

%%timeit
for i in range(1000):
    print(i)
# above is a placeholder for your code

and it will give you the time it took to run the code block in milliseconds. Alternatively you can do %timeit l = [i for i in range(10)] #placeholder for a single line of code to get the execution time for the single line.

Upvotes: 15

Related Questions