Closed Karami-m closed 1 year ago
The problem was solved by adding the following to the run.py file:
import tensorflow as tf
gpus = tf.config.experimental.list_physical_devices('GPU')
for gpu in gpus:
tf.config.experimental.set_memory_growth(gpu, True)
We used NVIDIA 3080 or sometimes the NVIDIA A10 for training. Usually changing the train_batch_size to a smaller value helps with the out-of-memory error problem. I am glad you found a solution that worked for you, hence your error might be related to a different version of TensorFlow.
Hi,
Thanks for making the code available for such an interesting work.
I have tried to train the Relation Prediction model on GPUs with 32 GB of memory but it lead to CUDA out of memory error. I have also tried to train with vgg16(pretrain=True) but still ran into the same problem. So, I wonder what kind of GPU you used for your experiments and how you manage the memory in training.