Unverified50% confidenceFactExact time
The Dice coefficient measures the degree of overlap between the predicted mask and the ground-truth annotation, ranging from 0 to 1, with the formula being twice the intersection divided by the sum of the total number of pixels (or voxels) of the two; it is sensitive to overlap in large-area regions, but tends to give overly optimistic scores on small targets or boundary regions.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDice损失是基于整个预测区域与真实区域的重叠度计算的区域级别度量73% similarUnverifiedThe paper derives generalization error upper bounds separately for the preliminary estimator and the constrained estimator; these bounds are expressed in terms of complexity measures such as the covering number or VC dimension of the family of measurable sets, and do not depend on the specific structure of the response space.71% similarUnverifiedIf a saturated dataset is still included in the aggregate total score, its effect is equivalent to mixing a constant into the average score, which not only contributes no information but also dilutes the signal weight of other datasets that still have discriminating power.71% similarUnverifiedMean rather than median should be used to estimate on-device throughput, because the median filters out slow but real use cases, making capacity estimates overly optimistic.69% similarUnverifiedPR曲线在样本不平衡场景下更有价值，根本原因在于精确率的计算不涉及真负例（TN）的数量69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1023870API
curl https://kongchang.com/api/v1/knowledge/claims/1023870MCP
get_claim(id=1023870)