-
|
from sahi.predict import predict predict(...)怎么保存预测的标注文件 在code的页面中提到的使用指南文档无法访问,但其它文档是正常的,是我的网络出问题了吗,能不能直接发我一份 |
Beta Was this translation helpful? Give feedback.
Answered by
fcakyon
Aug 31, 2025
Replies: 2 comments 1 reply
-
1.这样的情况不是个例,使用了切片预测后许多图片存在将背景错误标记为目标的情况,而没有使用切块预测时下反而不会出现这种情况,请问应该如何改善?
2.在不使用切块预测时,能够正确的识别图中的ship,而在使用了切块预测之后,目标的相对尺寸更大应该比正常预测要更好,但是我却遗憾地发现其将部分ship预测为car,这是怎么回事呢?有什么方法改善吗?
|
Beta Was this translation helpful? Give feedback.
1 reply
-
|
@YWJ-sudo when you provide Or you can convert each ObjectPrediction object to COCO prediction format by using Line 122 in e7109ee |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
fcakyon
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment


@YWJ-sudo when you provide
coco_dataset_jsonto sahi predict cli or function, it saves results in coco prediction format: cli usage | functional usageOr you can convert each ObjectPrediction object to COCO prediction format by using
object_prediction.to_coco_prediction(image_id=...):sahi/sahi/prediction.py
Line 122 in e7109ee