Skip to content

Out of memory using random images #2

Description

@Neonsign247

Hi. Thanks for great code.

I modified you code to utilize my own image dataset, and I failing with out of memory.

Traceback (most recent calls WITHOUT Sacred internals):
File "experiments/patch_experiment.py", line 128, in main
learning_loop.train_step(observer, c.inner_loop_steps, c.mnist_classes)
File "/lhi/GTN/GTN_clean/gtn/models/learning_loop.py", line 146, in train_step
self.optimizer_teacher.step()
File "/opt/conda/lib/python3.7/site-packages/torch/optim/adam.py", line 103, in step
denom = (exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(group['eps'])
RuntimeError: CUDA out of memory. Tried to allocate 2.00 GiB (GPU 0; 11.93 GiB total capacity; 9.53 GiB already allocated; 1.85 GiB free; 9.62 GiB reserved in total by PyTorch)

I'm using 64x64 images so that I changed input and output of learner/teacher to 64.
I reduced batch size to 4 / test_batch_size 4

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions