[BUG] memory allocator does not consider non-output leaf tensors

microsoft / nnfusion

A flexible and efficient deep neural network (DNN) compiler that generates high-performance executable from a DNN model description.

MIT License

957 stars 161 forks source link

[BUG] memory allocator does not consider non-output leaf tensors #277

Open xysmlx opened 3 years ago

xysmlx commented 3 years ago

🐛 Bug

If an operator has multiple outputs and some of them cannot achieve by the reversed-dfs from model's output tensors, the memory allocators will not allocate memory for these leaf tensors, leading to failure in executing this operator.

nnfbot commented 3 years ago

Thanks for the report @xysmlx! I will look into it ASAP! (I'm a bot).