A question about `Fused Embedding` demo
#2
by
mcm0123
- opened
From the Usage section of the model card:
# Text and image embedding
text_embeddings = model.get_text_embeddings(texts)
image_embeddings = model.get_image_embeddings(images)
print('Text and image embeddings', (text_embeddings @ image_embeddings.T).tolist())
# Fused Embedding
text_with_image_embeddings = model.get_fused_embeddings(texts=texts, images=images, instruction=t2i_prompt)
print('Text and image embeddings', (text_embeddings @ image_embeddings.T).tolist())
print('Text and image embeddings', (text_embeddings @ image_embeddings.T).tolist()) occured in both Text and image embedding and Fused Embedding part, I wonder if the fused embedding part should have a different result?