whu-dft
whu-dft
I'm also trying to train ImageNet 1k from scratch using my own optimizer and hyper-parameters. I got better result of 55%@top1 which is still far from the SOTA result reported...
Hi, thanks for your attention. I read a few papers these days, and found that the metric that Graphormer used is the std. MAE. The std. MAE originates from the...
@zhengsx Hi, I feel sorry to tell you that I have made a mistake. The way Graphormer calculates energy MAE loss is quit right. It is me who misunderstanded the...
> What does atom stand for? Isn't it the node? What are the factors that set its size? Very confusing Yes, Graphormer uses nodes to represent atoms. A vector with...
> I think you just need to replace the embedding with your feature, then it should works. And this would requires some studies on the code implementation of Graphormer. (Note...
1. Install the Graphormer ```bash git clone [email protected]:microsoft/Graphormer.git cd Graphormer bash install.sh ``` 2. Find the way to start the training of Graphormer ```bash cd example/oc20 cat oc20.sh # try...
> Sorry, I haven't try any tests yet and I also don't know any benchmarks for non-molecular problems.
@hipudding Hello, I have recently utilized the draft code [here](https://github.com/hipudding/pytorch-lightning/tree/npu_support) for training with PyTorch Lightning on NPU. It has proven to be quite useful, and I would like to extend...