CUDA out of memory batch size=2, (V100 32G) about score_sde_pytorch HOT 2 OPEN

Face the same issue here. Adding the following code solves the problem for me.

gpus = tf.config.list_physical_devices('GPU')
if gpus:
  try:
    # Currently, memory growth needs to be the same across GPUs
    for gpu in gpus:
      tf.config.experimental.set_memory_growth(gpu, True)
    logical_gpus = tf.config.list_logical_devices('GPU')
    print(len(gpus), "Physical GPUs,", len(logical_gpus), "Logical GPUs")
  except RuntimeError as e:
    # Memory growth must be set before GPUs have been initialized
    print(e)

from score_sde_pytorch.

Face the same issue here. Adding the following code solves the problem for me.

gpus = tf.config.list_physical_devices('GPU')
if gpus:
  try:
    # Currently, memory growth needs to be the same across GPUs
    for gpu in gpus:
      tf.config.experimental.set_memory_growth(gpu, True)
    logical_gpus = tf.config.list_logical_devices('GPU')
    print(len(gpus), "Physical GPUs,", len(logical_gpus), "Logical GPUs")
  except RuntimeError as e:
    # Memory growth must be set before GPUs have been initialized
    print(e)

Thanks, it worked for me. FYI, I'm using 4 RTX 3090 (24 G) with batch size 28 (7 samples per GPU), and 128 x 128 images.

from score_sde_pytorch.