adler-j / learned_primal_dual

Learned Primal-Dual Reconstruction
https://arxiv.org/abs/1707.06474
98 stars 37 forks source link

memory error when runing mayo_learned_primal_dual #10

Open hjahan58 opened 3 years ago

hjahan58 commented 3 years ago

hello. thanks for your attention. I trying to train _mayo_learned_primaldual with some database. I run program on gpu enabled env Quadro RTX 4000 with 8G memory .but i get OOM Error I run program on cpu enabled env with 32G ram. but I get OOM error.

. . . 2021-06-07 17:42:42.878792: I tensorflow/core/common_runtime/bfc_allocator.cc:645] Sum Total of in-use chunks: 25.98GiB 2021-06-07 17:42:42.878807: I tensorflow/core/common_runtime/bfc_allocator.cc:647] Stats: Limit: 33547444224 InUse: 27897504512 MaxInUse: 27897504512 NumAllocs: 536 MaxAllocSize: 6710886400

2021-06-07 17:42:42.878844: W tensorflow/core/common_runtime/bfc_allocator.cc:271] ******__**_**____*** 2021-06-07 17:42:43.028891: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at mkl_relu_op.cc:861 : Resource exhausted: OOM when allocating tensor with shape[1677721600] and type float on /job:localhost/replica:0/task:0/device:CPU:0 by allocator mklcpu Traceback (most recent call last): . . .