Skip to content
DeepForgeby svx
All problems
EasyDeep Learningdl-107

AdamW State Bytes

Problem

Compute the AdamW optimizer state size in bytes. AdamW keeps two moment estimates per parameter, so the state is 2 * num_params * bytes_per_value.

The signature is adamw_state_bytes(num_params, bytes_per_value=4). Return an integer.

Hint

The two moments are the first moment and the second moment.

Test cases

case 1
in: [1,4]
expected: 8
case 2
in: [1000000,4]
expected: 8000000
case 3
in: [1000,2]
expected: 4000
case 4
in: [0,4]
expected: 0
Starter codePython
def adamw_state_bytes(num_params, bytes_per_value=4):
    # Your code here
    pass

4 test cases · run your code against them in the editor