can not reproduce validation accuracy with pretrained model #23

Open

opened

on Jun 11, 2018

On ImageNet, I only get a top-5 accuracy of 83.86%.
Any idea what's wrong?

Metadata

Assignees

No one assigned

Labels

No labels

No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests