How is the picture processed in sequence_generate?

[xiaoxiaokuye]

The shape of the picture is (Batch, Channel, H, W)
The data shape that the sequence_generate can receive is (batch, seq_len,...)
I did not find a solution in your code, how did you deal with this problem?
Thank you

[zhiqwang]

Hi @xiaoxiaokuye , I haven't followed this code for a while. I guess this is done in

sightseq/sightseq/models/text_recognition_encoder.py

Line 74 in b1eac1d

x = x.permute(3, 0, 1, 2).view(x.size(3), x.size(0), -1) # seq_len x bsz x embed_dim