This website requires JavaScript.
Explore
Help
Sign In
周翔宇
xzhou
0 Followers
·
0 Following
Joined on
2026-08-27
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues.
Learn more about blocking a user.
User to block:
Optional note:
The note is not visible to the blocked user.
Cancel
Block
Repositories
1
Projects
Packages
Public Activity
Starred Repositories
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-07 20:36:57 +08:00
c9e7e4104c
fix(configs): update dataset root path in film.yaml
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-07 19:30:19 +08:00
92c88b785c
fix(config): --set 拦下不存在的键,并支持续训时重开学习率曲线
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-07 19:11:33 +08:00
506fde0290
fix(cluster): 配置层级写错不再静默忽略,并始终打印生效的超参
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-07 19:06:40 +08:00
105cb38831
fix(trainer): 检测训练续训以配置为准,并恢复学习率调度进度
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-07 14:26:02 +08:00
01cfff9b02
fix(configs): update optimizer settings and add projection parameters in film_simclr_resnet50.yaml
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-07 11:39:01 +08:00
0f5bb4a5c0
feat(cluster): 加 --train-only,自监督训练与聚类可分两步跑
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-07 11:19:44 +08:00
5d395b7c25
chore: 数据迁到 data/raw/ 后同步所有路径引用
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-07 11:07:56 +08:00
5f31ed0a92
feat(configs): 新增 resnet50 的自监督聚类配置
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-06 13:17:56 +08:00
f8a52c9382
feat(scripts): 按聚类结果导出数据集
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-06 13:05:55 +08:00
7f13ec5d18
fix(configs): update training epochs and optimizer settings in film_simclr_resnet18.yaml
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-05 11:10:07 +08:00
004d7a1ce7
fix(simclr): 换优化器种类时不再加载不兼容的动量缓冲
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 21:46:43 +08:00
b7505a743d
fix(simclr): 续训以配置为准,不再被存档里的超参静默覆盖
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 21:41:48 +08:00
703ba7076b
fix(configs): update num_clusters and training parameters in film_simclr_resnet18.yaml
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 17:25:47 +08:00
865cdd7210
fix(simclr): 续训时改变总轮数不再让学习率卡在 0
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 17:19:24 +08:00
6b11128357
fix(configs): update film_simclr_resnet18.yaml with new clustering parameters and optimizer settings
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 15:51:31 +08:00
6f00074199
feat(cluster): 新增 resnet18 自监督聚类配置,并修复 --set 的 YAML 标量解析
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 15:42:17 +08:00
47fb0f1125
fix(configs): 修复 resnet18 训练配置无法构建,并补上对应的评估配置
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 15:28:51 +08:00
e581d9b3d7
fix(backbones): 修复 shufflenet 无法构建,并让换骨架不再踩配置合并的坑
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 15:09:38 +08:00
e4bd794025
feat(cluster): 输出簇分布图,直观展示簇间距离
xzhou
pushed to
main
at
xzhou/ModelTrainEval
2026-09-04 14:54:18 +08:00
4e6c8aa2ea
fix: update work directory in film.yaml and add pearlescent.yaml configuration
Previous
4
Next