marie-ai
marie-ai copied to clipboard
ERROR marie@32 Error in PSM_SPARSE_STEP : grid_sampler_2d_cpu not implemented for Half
Need to handle this better. ref : 215000470
ERROR marie@32 Error in PSM_SPARSE_STEP : grid_sampler_2d_cpu not implemented for Half
INFO marie@31 Batch time [2, 101]: 1.0219602584838867 [10/03/24 10:19:31]
UserWarning: torch.utils.checkpoint: the use_reentrant parameter should be passed explicitly. In version 2.4 we will raise an exception if use_reentrant is not passed. use_reentrant=False is recommended, but if you need to preserve the current default behavior, you can pass use_reentrant=True. Refer to docs for more details on the differences between the two variants. [checkpoint.py:464]
UserWarning: None of the inputs have requires_grad=True. Gradients will be None [checkpoint.py:91]
| WARNING: ran out of memory
WARNI… marie@32 OOM detected, clearing cache and retrying [10/03/24 10:19:31]
ERROR marie@32 Error in PSM_SPARSE_STEP : CUDA out of memory. Tried to allocate 1.34 GiB. GPU [10/03/24 10:19:31]
| WARNING: ran out of memory
WARNI… marie@32 OOM detected, clearing cache and retrying [10/03/24 10:19:32]
ERROR marie@32 Error in PSM_SPARSE_STEP : CUDA out of memory. Tried to allocate 1.34 GiB. GPU [10/03/24 10:19:32]
INFO marie@32 Refinement step 1 : No change in image
WARNI… marie@32 Empty bounding boxes, possibly a blank page [10/03/24 10:19:32]
INFO marie@31 Batch time [3, 42]: 0.4753885269165039 [10/03/24 10:19:32]
INFO marie@31 ICR Time elapsed: 3.549083709716797
ERROR marie@32 Error in PSM_SPARSE_STEP : grid_sampler_2d_cpu not implemented for Half
Payload: