Skip to content
This repository has been archived by the owner on Jan 18, 2023. It is now read-only.

Error while training on colab TPU #23

Open
waleedrazakhan92 opened this issue Jul 27, 2022 · 0 comments
Open

Error while training on colab TPU #23

waleedrazakhan92 opened this issue Jul 27, 2022 · 0 comments

Comments

@waleedrazakhan92
Copy link

waleedrazakhan92 commented Jul 27, 2022

While trying to train I keep getting this error
File "/content/stylegan2-tpu/dnnlib/tflib/autosummary.py", line 116, in image_grid (grid_shape, int(input_tensor.shape[0]))) ValueError: Grid shape (3, 3) incompatible with minibatch size 8.

First I tried with a larger batch size but kept getting File "/tensorflow-1.15.2/python3.7/tensorflow_core/python/framework/ops.py", line 1748, in __init__ self._traceback = tf_stack.extract_stack() this error. So I googled around and got to know that its an overflow error. So i set the BATCH_PER to 1 and tried training again but now I'm getting the incompatible minibatch size error mentioned above.
Please let me know if I'm doing anything wrong. Or the grid size to save is somehow hardcoded and not working with the new BATCH_PER size. I'm using image size to be 256x256

@waleedrazakhan92 waleedrazakhan92 changed the title Error while training on colab t Error while training on colab TPU Jul 27, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant