Need some clarity in the code

[archmaester]

In line 41 and 42 of fusion.py file Instead of this fusion_visual = tf.reduce_mean(visual_vector * visual_f, 1) fusion_text = tf.reduce_mean(text_vector * text_f, 1) Shouldn't it be fusion_visual = tf.reduce_mean(text_vector * visual_f, 1) fusion_text = tf.reduce_mean(visual_vector * text_f, 1)

Since we want to generate sentence aware visual embeddings ....

[AmingWu]

Yes, it should be fusion_visual = tf.reduce_mean(text_vector * visual_f, 1) and fusion_text = tf.reduce_mean(visual_vector * text_f, 1). Please see the line 53 and 54 of fusion.py.

Thank you very much.