Zhen Wang
Have you cheacked your training data? Your training captions may contain extra spaces which leads to this.
I find that not all of your training caption end with the '.', since the end token for beamsearch is '.', thus the model may not know when to end...
> 貌似SCAN里面并没有这个数据 你好，请问你后来提取到边界框坐标的数据了嘛
We did not notice the exact memory requirement, since our server have 504g memory and it did not crash when constructing coco_ee. You may try to switch to a server...