Equiformer feature input as float #22
-
|
Hi I have the feature vector from Bert which is the float. Can I adjust the code in model for accept input float? |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 1 reply
-
|
could you guide me if the function will effect if I input float? |
Beta Was this translation helpful? Give feedback.
-
|
Sorry, I forgot to notice at "the adjacency matrix" topic. I try to debug from int to the float several time. until I come back in main page where you mention the shape of (batch, node, feat) and coincident that I think the paper mention vividly about the embedding use one-hot encode to input. Thank you for simplify the code for us. 🤗 |
Beta Was this translation helpful? Give feedback.
Sorry, I forgot to notice at "the adjacency matrix" topic. I try to debug from int to the float several time. until I come back in main page where you mention the shape of (batch, node, feat) and coincident that I think the paper mention vividly about the embedding use one-hot encode to input. Thank you for simplify the code for us. 🤗