cihebot-98
|
ebdaf6eac2
|
fix docker bug
|
7 months ago |
|
cihebot-98
|
09648ce50c
|
add docker
|
7 months ago |
|
cihebot-98
|
748ec4f751
|
Merge branch 'hq_study' of https://git.chiebot.com:10000/HuQiong/mae-cnn into hq_study
|
7 months ago |
|
cihebot-98
|
e362e92d05
|
git remove runs
|
7 months ago |
 captainfffsama
|
d52bf4f275
|
change demo show
|
7 months ago |
|
captainfffsama
|
d0db6fbe15
|
add debug show net train
|
12 months ago |
|
captainfffsama
|
87b0f6f84a
|
fix ipynb bug
|
12 months ago |
|
captainfffsama
|
ca7496fc9d
|
support yolov8 cspnet
|
12 months ago |
|
captainfffsama
|
3ee54730eb
|
support own dataset
|
12 months ago |
keyu-tian
|
a63e386f8e
|
[fix] (#58) fix default hyperparameters
|
1 year ago |
|
keyu-tian
|
00883b885d
|
[fix] (#58) in case the ckpt keys are not aligned
|
1 year ago |
|
Keyu Tian
|
3c384e9b18
|
[upd] readme
|
2 years ago |
|
Keyu Tian
|
2ac9627944
|
[upd] note for customized CNN with SE, global average pooling, or GroupNorm
|
2 years ago |
|
Keyu Tian
|
cea9e21cdd
|
[upd] refine tutorials
|
2 years ago |
|
Keyu Tian
|
74552bc23e
|
[upd] typo
|
2 years ago |
|
Keyu Tian
|
f2ffa9b9cb
|
[upd] (#44, #35) more tutorial for pretraining/selfsup finetuning on customized dataset
|
2 years ago |
|
keyu-tian
|
c4b2527b67
|
[upd] fix typo
|
2 years ago |
|
keyu-tian
|
1468df8a88
|
[upd] (#38, #14) update viz_reconstruction.ipynb for visualizing your own model
|
2 years ago |
|
keyu-tian
|
f02a5a926b
|
[upd] (#38) add `--init_weight` for pretraining
|
2 years ago |
|
keyu-tian
|
a64bdf729f
|
[upd] READMEs, add new recorded video on bilibili
|
2 years ago |
|
keyu-tian
|
99c3df9d85
|
[upd] make it more easier to modify for regtangle input (H != W); remove main.sh and launch.py, simplify the command for exp (use torchrun directly)
|
2 years ago |
|
Keyu Tian
|
d49588bf19
|
[fix] mixup return values
|
2 years ago |
|
keyu-tian
|
dfc16e76f6
|
[fix] Mixup error when the batch size is odd
|
2 years ago |
|
Keyu Tian
|
d884c0ecff
|
[upd] typo in pretrain/README.md
|
2 years ago |
|
Keyu Tian
|
00a114c680
|
[upd] README.md
|
2 years ago |
|
Keyu Tian
|
4f292d1756
|
[upd] ICLR poster link
|
2 years ago |
|
keyu-tian
|
d25bc58dde
|
[upd] vis for resnets
|
2 years ago |
|
Keyu Tian
|
02f7444fd0
|
[upd] README.md
|
2 years ago |
|
keyu-tian
|
85d96dd1ef
|
[fix] DDP error
|
2 years ago |
|
keyu-tian
|
36aa8210b2
|
[upd] add support for 1GPU debug and customized dataset
|
2 years ago |
|
keyu-tian
|
cba9d26a66
|
[upd] upd README, release pretrained weights with SparK's decoder
|
2 years ago |
|
keyu-tian
|
4fc8893577
|
[upd] more detailed tutorial for customizing your CNN model
|
2 years ago |
|
keyu-tian
|
35e5b496ad
|
[upd] more detailed tutorial for customizing your CNN model
|
2 years ago |
|
keyu-tian
|
ec877d300d
|
[upd] update /pretrain/README.md
|
2 years ago |
|
keyu-tian
|
6ffe453fa5
|
[upd] 1. refactor a lot to simplify the pretraining codes; 2. add tutorial for customizing your own CNN model; 3. update some READMEs
|
2 years ago |
|
keyu-tian
|
46f9ad2871
|
[upd] typo
|
2 years ago |
|
keyu-tian
|
5fbd93d3df
|
[upd] add small model for more lightweight visualization
|
2 years ago |
|
keyu-tian
|
48401f6ad9
|
[upd] add small model for more lightweight visualization
|
2 years ago |
|
Keyu Tian
|
bea2b8281b
|
Update README.md
|
2 years ago |
|
keyu-tian
|
936f81421a
|
[upd] README.md
|
2 years ago |
|
keyu-tian
|
cf48522037
|
[upd] README.md; upload the PPT slides
|
2 years ago |
keyu tian
|
22ec3787f5
|
[fix] typo in README.md
|
2 years ago |
|
keyu tian
|
11d4f51bad
|
[upd] README
|
2 years ago |
|
keyu tian
|
4826bc84e4
|
[upd] README
|
2 years ago |
|
keyu tian
|
66b27db5e1
|
[add] add demo: pretrain/viz_reconstruction.ipynb
|
2 years ago |
|
keyu tian
|
8a1412af36
|
Merge remote-tracking branch 'origin/main'
|
2 years ago |
|
keyu tian
|
c4d28a1c6a
|
[upd] add notes for 384 training
|
2 years ago |
|
Keyu Tian
|
f02a412861
|
Update README.md
|
2 years ago |
|
Keyu Tian
|
01fd012d2b
|
Update README.md
|
2 years ago |
|
Keyu Tian
|
b303720c3d
|
Update README.md
|
2 years ago |
