id特征太多，gpu放不下？

2021-12-01 04:09:29.838463: I tensorflow/core/common_runtime/bfc_allocator.cc:929] Stats:
Limit: 38920012432
InUse: 25354372352
MaxInUse: 25354372352
NumAllocs: 336712
MaxAllocSize: 12542967808

2021-12-01 04:09:29.838483: W tensorflow/core/commonruntime/bfcallocator.cc:424] **__**__**
2021-12-01 04:09:29.838589: W tensorflow/core/framework/op_kernel.cc:1651] OP_REQUIRES failed at topk_op.cc:83 : Resource exhausted: OOM when allocating tensor with shape[1024,3062248] and type int32 on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc

解决方法：
https://stackoverflow.com/questions/43288147/how-do-i-use-a-very-large-2m-word-embedding-in-tensorflow

UnicodeEncodeError

UnicodeEncodeError: ‘ascii’ codec can’t encode character ‘\u2122’ in position 67273: ordinal not in range(128)