CODEWITHSUNDEEP
pythonpython-3.xmemoryout-of-memory
Leo hk
Leo hk

Reputation: 41

Same Memory Usage for float 32 data and float 64 data?

I have tried to use tracemalloc to trace the memory usage of my coding and I found that for float 32 data and float 64 data, they have same peak memory usage. Why?

Screenshot of output:

Photo - Result

float64:

import tracemalloc
import numpy as np
tracemalloc.start()
arr = np.zeros((100000000,), dtype=np.float64)
for i in range(100000000):
    arr[i] = i
current, peak = tracemalloc.get_traced_memory()
#current memory is the memory the code is currently using and 
#peak memory is the maximum space the program used while executing.
tracemalloc.stop()
print(f"Current memory usage is {current / 10**6}MB; Peak was {peak / 10**6}MB")

Output:

Current memory usage is 800.14981MB; Peak was 800.160145MB

float32:

import tracemalloc
import numpy as np
tracemalloc.start()
arr = np.zeros((100000000,), dtype=np.float32)
arr = arr.astype(np.float32)
for i in range(100000000):
    arr[i] = i
current, peak = tracemalloc.get_traced_memory()
tracemalloc.stop()
print(f"Current memory usage is {current / 10**6}MB; Peak was {peak / 10**6}MB")

Output:

Current memory usage is 400.149626MB; Peak was 800.149468MB

Upvotes: 1

Views: 1422

Answers (1)

user202729
user202729

Reputation: 3955

You can see that the 2 snippets are not equivalent. There's this additional statement in the second snippet

arr = arr.astype(np.float32)

That statement (even though arr already have type np.float32):

  • copies the array (which takes the same amount of memory as the old array)
  • then assign the new array to arr,
  • then discards the old array.

Therefore the peak memory doubles.

Upvotes: 2

Related Questions