eb9600ef46
update hpi config
2024-11-08 10:11:55 +00:00
4b3503b6de
support face recgnition ( #3282 )
...
* add mobilefacenet
* exp with pplcnet on face
* support face recognition
* fix face recognition models bugs
* add face recognition README
2024-11-01 19:48:28 +08:00
d178931fca
fix gpu memory growth
2024-10-18 17:05:11 +08:00
542b320008
fix export bug
2024-09-26 01:32:49 +08:00
a9a730e3b3
support uniform output
2024-09-24 23:25:41 +08:00
104b378828
modify ema.pdparams to pdema
2024-09-04 15:20:28 +08:00
fe5700fce6
export with label ( #3166 )
2024-07-25 21:31:21 +08:00
6aa3047157
Respect `is_rec` when eval ( #3187 )
2024-07-10 15:45:37 +08:00
caa6fdf57d
Fix missing label for classification model in eval mode ( #3185 )
2024-07-09 17:02:15 +08:00
1dcab0a7bc
Fix wrong InputSpec setting for AST Dy2St ( #3176 )
2024-07-05 14:05:02 +08:00
e23bdb4df5
merge uapi paddleclas ( #3135 )
2024-05-13 14:27:09 +08:00
07de476c51
refine
2024-04-26 14:58:55 +08:00
0cb74b1741
refine
2024-04-26 14:58:55 +08:00
be33107d89
refine batch_cost stat
2024-04-26 14:58:55 +08:00
d189384ab8
Control print memory info ( #3079 )
...
* support ctrl print memory info
* support ctrl print memory info
* fix print log bug
2024-03-05 11:28:23 +08:00
034236d0f5
【benchmark】fix gpu_mem unit ( #3052 )
...
* add max_memory_reserved for benchmark
* change unit
* fix
2023-12-05 15:04:25 +08:00
656c149d6b
Fix bug when running on XPU ( #3045 )
2023-11-24 15:08:26 +08:00
e4abcc08d0
add max_memory_reserved for benchmark ( #3042 )
2023-11-22 17:31:30 +08:00
ab087065e9
support to specify rank to log when using Fleet API ( #3039 )
...
* support to specify rank to log when using Fleet API
* log max mem reserved
* log_ranks support str type
example: -o Global.log_ranks="0,1"
* log max mem allocated
* support to specify rank to log in static mode
* log max mem reserved and max mem allocated in static mode
2023-11-16 11:32:29 +08:00
aae1e9543f
del load pretrained from url for resnet ( #2997 )
...
* del load pretrained from url for resnet
* del load_dygraph_pretrain_from_url
* del load_dygraph_pretrain_from_url
* modify save_load
2023-10-30 13:44:16 +08:00
98935e0bb7
Add ML-Decoder Support ( #2957 )
...
* 添加ML-Decoder，并使其可以在Arch作用域下控制
* 添加MultiLabelAsymmetricLoss
* 添加MultiLabelMAP评价方式，并避免其在回合内计算，每回合计算一次，节约时间
* 添加COCO2017数据集格式转换脚本
* 添加OneCycleLR学习率调度策略
* 添加ResNet50_ml_decoder
* 添加ResNet_ml_decoder
* 添加ResNet_ml_decoder
* 添加ResNet_ml_decoder preprocess
* 添加ResNet_ml_decoder
* 添加ResNet_ml_decoder
* 直接从model中拉class_num参数
* fix message error
* fix message error
* 给出基于inference model预测的文档
* 融合cutout使其具有随机生成填充值的能力
* 更正变量名称
* 更新configs
* 更新README
* fix bugs
* fix bugs
* update
* update
* update
2023-09-08 16:17:35 +08:00
74a33b7f50
fix gbk ( #2941 )
2023-09-01 17:49:33 +08:00
f3b2b2f4ad
[uapi]Save predict result ( #2926 )
...
* sava predict result
2023-08-29 14:32:07 +08:00
75a5bb17ba
add 2 more custom devices intel_gpu and apple mps
2023-06-29 19:42:38 +08:00
de5c4e1b1c
Change vdl dir
2023-06-26 14:20:38 +08:00
d6137854e2
Accommodate UAPI
2023-06-26 14:20:38 +08:00
8405882f11
debug
2023-05-29 19:52:09 +08:00
2d8346cd3b
fix _init_amp when export
2023-05-29 19:52:09 +08:00
14d06fb6bd
support AMP.use_amp arg
2023-05-25 16:16:02 +08:00
162f013ebe
fix: minimize() dont support parameter_list of type dict
...
there are diffs that step()+update() and minimize().
this will be fixed in https://github.com/PaddlePaddle/Paddle/pull/53773 .
2023-05-25 11:58:05 +08:00
8b218b01ac
refactor amp auto_cast context manager & loss scaler
2023-05-25 11:58:05 +08:00
f884f28853
refactor amp
2023-05-25 11:58:05 +08:00
b3678234fe
fix bug when update_freq > iter_per_epoch
2023-05-17 15:19:13 +08:00
9f621279b8
fix infer output
2023-04-17 20:28:40 +08:00
52f16cc85d
Update engine.py
2023-04-11 19:23:57 +08:00
6e6586f59b
Fixed the incorrect infer outputs
2023-04-11 19:23:57 +08:00
5f2eaa7cb1
bugfix: set_epoch after reume
2023-04-06 15:33:30 +08:00
f37cb543b1
rm op black list in amp
...
the op flatten_contiguous_range and greater_than has supported amp mode since paddle 2.4
2023-03-29 14:57:02 +08:00
a7ba6eabd2
optimizer must be decorated when training with AMPO2
2023-03-28 18:42:26 +08:00
5d06a88a36
Revert "refactor: simplify engine"
...
This reverts commit 376d83d46e
2023-03-14 16:47:13 +08:00
6aabb94d8c
Revert "refactor: add ClassModel to unify model forward interface"
...
This reverts commit 75a20ba557
2023-03-14 16:47:13 +08:00
e7e4f68b5c
Revert "refactor: build_train_func & build_eval_func"
...
This reverts commit 6bed0f5707
2023-03-14 16:47:13 +08:00
f2fc43baeb
Revert "refactor: mv all dataloaders to engine.dataloader_dict"
...
This reverts commit 284e2a6756
2023-03-14 16:47:13 +08:00
a1e840e0da
Revert "refactor: iter_per_epoch -> max_iter"
...
This reverts commit a38e42f644
2023-03-14 16:47:13 +08:00
0efda2c75e
Revert "refactor: simpfy engine.train()"
...
This reverts commit fad5c8e348
2023-03-14 16:47:13 +08:00
5a4ee1aec4
Revert "refactor"
...
This reverts commit 0e28a39da3
2023-03-14 16:47:13 +08:00
f42719afbb
Revert "replace the arg engine with config"
...
This reverts commit f525cea006
2023-03-14 16:47:13 +08:00
7243f1429b
Revert "rm codes for compatibility with old version"
...
This reverts commit 6e77bd6cd5
2023-03-14 16:47:13 +08:00
56e8c5a992
Revert "mv model_saver to __init__()"
...
This reverts commit 0d7e595fc7
2023-03-14 16:47:13 +08:00
f1a7a22a34
Revert "mv some attrs to __init__()"
...
This reverts commit 73e2cde617
2023-03-14 16:47:13 +08:00
zhangyubo0722
学卿
zhouchangda
zhangyubo0722
Nyakku Shigure
Wang Huan
changdazhou
gmm
RuohengMa
Tingquan Gao
zhenming lin
baocheny
Bobholamovic
parap1uie-s
Yang Nie
Tingquan Gao