Flynn

Results 4 issues of Flynn

Environment: VMware Ubuntu14.04 When the computer connects to the bebop by wifi and roslaunch bebop_driver bebop_node.launch, The following error occurred: ...... [ INFO] [1501751964.009304462]: [ARCONTROLLER_Network] 17:19:24:009 | ARCONTROLLER_Network_GetAvailableSocketPort:128 - d2c_port...

`X1_full = torch.repeat_interleave(X1.unsqueeze(0), len(indicies), dim=0)[ RuntimeError: CUDA out of memory. Tried to allocate 3.89 GiB (GPU 0; 11.93 GiB total capacity; 3.49 GiB already allocated; 3.65 GiB free; 7.67 GiB...

When I run `mig_algebraic_depth_rewriting_splitters` for [c5315.v](https://docs.google.com/document/d/1bDPvN9zxg9DpDWQIWRfXqw-9iv2tIwBeRbPcsg4KDUM/edit?usp=sharing) with the parameter `selective`, I found that the `run_selective ` is endless. ``` ps_alg_rewrite.strategy = mockturtle::mig_algebraic_depth_rewriting_params::selective; mig_algebraic_depth_rewriting_splitters(mig, ps_alg_rewrite); ``` And while in mockturtle::buffer_insertion::optimize can...

bug
wontfix

## ❓ Questions & Help 你好!请问我自定义的图数据集里面,节点特征为categorical,然后我使用了one-hot编码,使用experiment(model="gin", dataset=dataset)训练时,显示如下错误: `cogdl/operators/spmm/spmm.cpp:39: at::Tensor csr_spmm(at::Tensor, at::Tensor, at::Tensor, at::Tensor): Assertion `B.dtype() == torch::kFloat32 || B.dtype() == torch::kFloat16' failed.` 请问有什么解决方法吗?