Closed uezo closed 1 month ago
Add high-performance and dynamic vision control feature. To use vision, implement get_image() and configure ChatGPTProcessor.
get_image()
import io import pyautogui from aiavatar.processors.chatgpt import ChatGPTProcessor # Implement get_image async def get_image(source: str=None) -> bytes: buffered = io.BytesIO() image = pyautogui.screenshot(region=(0, 0, 1280, 720)) image.save(buffered, format="PNG") image.save("image_to_gemini.png") # Save current image for debug return buffered.getvalue() # Configure ChatGPTProcessor chat_processor_gpt = ChatGPTProcessor( api_key=OPENAI_API_KEY, model="gpt-4o", system_message_content="ユーザーからの要求を処理するために画像データが必要な場合、[vision:screenshot]を応答メッセージに含めてください。\n\n例\n[vision:screenshot]承知しました。画像を確認しています。" ) chat_processor_gpt.use_vision = True chat_processor_gpt.get_image = get_image
Add high-performance and dynamic vision control feature. To use vision, implement
get_image()
and configure ChatGPTProcessor.