Dimension error for geometric and appearance features in Relation Encoding

[Jian-Xi]

Thanks for sharing the codes. It's solid organized and compact programmed.

I'd like to have two questions about the RelationTransformerModel.py based on my running results.

1. Code at llne 454, after this compare operation, we got a array with boolean indexes which can not be added in the following line, i changed it
   seq_mask = (seq.data > 0) to seq_mask = (seq.data > 0).type(torch.int8)
   It's quiet thereafter.

2. In the function box_attention at line 236,
   # multiplying log of geometric weights by feature weights w_mn = torch.log(torch.clamp(w_g, min = 1e-6)) + w_a

the dimensions of these two geometric and appearance features are not matched. Thus I got error as follows:

RuntimeError: The size of tensor a (54) must match the size of tensor b (50) at non-singleton dimension 3

I've tried to figure it out what's going on there quite for a while but got no idea as so far. I'm not sure whether it depends on my environments (I think not) or it's just a typo in coding.

• torch 0.4.1
• torchvision 0.2.1
• 4 x Tesla V100-SXM2 Driver Version: 410.104 CUDA Version: 10.0

Any input will be appreciated.
Jian

[Jian-Xi]

The model structure is not designed for multiple devices. It can be launched just on single GPU. All the relation box modules use the same inputs( same length) to calculate the geo plus appearance features. If the batch data got chunked and scattered on different devices. it leads to dimension problem.