test tti example

.gitattributes

@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+examples/root.jpg filter=lfs diff=lfs merge=lfs -text

app.py

@@ -124,9 +124,10 @@ def single_inference_tti(head_txt, head_id, tail_txt, tail_id, question_img, que
         inputs['attention_mask'][i, :idx[2], idx[2]:] = 0
     
     # image
-    pixel_values = processor(images=[question_img], return_tensors='pt')['pixel_values'].squeeze()
-    inputs['pixel_values'] = pixel_values.unsqueeze(0)
-    
+    pixel_values = processor(images=question_img, return_tensors='pt')['pixel_values'].unsqueeze(1)
+    pixel_values = torch.cat((pixel_values, torch.zeros_like(pixel_values)), dim=1)
+    inputs['pixel_values'] = pixel_values
+
     input_ids = inputs['input_ids']
 
     model_output = mkgformer.model(**inputs, return_dict=True)
@@ -224,7 +225,7 @@ def single_tab_tti():
         inputs=[head_text, head_ent, tail_text, tail_ent, question_image, question_ent], 
         outputs=[output_text])
         
-    examples=[['qinghai_lake', 'Q201294', 'inland_lake', 'Q31805992', 'examples/qinghai_lake.png', 'Q18812548']]
+    examples=[['scrap', 'Q3217573', 'watch', 'Q178794', 'examples/root.jpg', 'Q111029']]
     ex = gr.Examples(
         examples=examples, 
         fn=single_inference_iit,