Update multi_modal_infer.py (meta-llama#696)

albertodepaola · web-flow · commit fe7ac56f872f · 2024-09-30T16:08:42.000-03:00
diff --git a/recipes/quickstart/inference/local_inference/multi_modal_infer.py b/recipes/quickstart/inference/local_inference/multi_modal_infer.py
@@ -38,7 +38,7 @@ def generate_text_from_image(model, processor, image, prompt_text: str, temperat
         {"role": "user", "content": [{"type": "image"}, {"type": "text", "text": prompt_text}]}
     ]
     prompt = processor.apply_chat_template(conversation, add_generation_prompt=True, tokenize=False)
-    inputs = processor(prompt, image, return_tensors="pt").to(model.device)
+    inputs = processor(image, prompt, return_tensors="pt").to(model.device)
     output = model.generate(**inputs, temperature=temperature, top_p=top_p, max_new_tokens=512)
     return processor.decode(output[0])[len(prompt):]
 

Original file line number	Diff line number	Diff line change
`@@ -38,7 +38,7 @@ def generate_text_from_image(model, processor, image, prompt_text: str, temperat`
`38`	`38`	`{"role": "user", "content": [{"type": "image"}, {"type": "text", "text": prompt_text}]}`
`39`	`39`	`]`
`40`	`40`	`prompt = processor.apply_chat_template(conversation, add_generation_prompt=True, tokenize=False)`
`41`		`- inputs = processor(prompt, image, return_tensors="pt").to(model.device)`
	`41`	`+ inputs = processor(image, prompt, return_tensors="pt").to(model.device)`
`42`	`42`	`output = model.generate(**inputs, temperature=temperature, top_p=top_p, max_new_tokens=512)`
`43`	`43`	`return processor.decode(output[0])[len(prompt):]`
`44`	`44`