I use utils.checkpoint with DistributedDataParallel at the same time to save the CUDA memory.
When I use a single GPU to run my code, it works.
But if I use multi GPUs to run my code, it will report an error:
RuntimeError: RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one. This error indicates that your module has parameters that were not used in producing loss. You can enable unused parameter detection by passing the keyword argument
find_unused_parameters=True
totorch.nn.parallel.DistributedDataParallel
, and by
making sure allforward
function outputs participate in calculating loss.
If you already have done the above, then the distributed data parallel module wasn’t able to locate the output tensors in the return value of your module’sforward
function. Please include the loss function and the structure of the return value offorward
of your module when reporting this issue (e.g. list, dict, iterable).
Parameter indices which did not receive grad for rank 2: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 …
In addition, you can set the environment variable TORCH_DISTRIBUTED_DEBUG to either INFO or DETAIL to print out information about which particular parameters did not receive gradient on this rank as part of this error
But I checked my code on single GPU, all params have their grad. I do not know what cause this error. It looks like all my params have no grad.