注意力的维度

[Eight3H]

你好，作者，如果我现在不仅有h,w ，还有一个d ，作者有没有考虑过这方面，请您联系我www.756233138@foxmail.com

[Eight3H]

只针对 Agentattention那快的代码，注意力偏置

[tian-qing001]

Hi @Eight3H, please provide more details regarding your question.

[Eight3H]

self.ah_bias = nn.Parameter(torch.zeros(1, num_heads, agent_num, window_size[0], 1)) self.aw_bias = nn.Parameter(torch.zeros(1, num_heads, agent_num, 1, window_size[1])) self.ha_bias = nn.Parameter(torch.zeros(1, num_heads, window_size[0], 1, agent_num)) self.wa_bias = nn.Parameter(torch.zeros(1, num_heads, 1, window_size[1], agent_num)) 你的注意力偏置是这样的，现在我不止有h和w,还有d也就是处理的是3d的 该怎么处理，有考虑过吗

[tian-qing001]

It's essential to provide a clear and detailed description of your problem, avoiding assumptions about my understanding of the specific area and task you are addressing. If you are denoting depth by 'd', consider expanding the dimensions of agent bias and treating it similarly to height and width. If 'd' represents dimension, treat it in a manner analogous to num_heads.

[Eight3H]

是深度