099310efe2
Update tipc
2023-02-28 15:01:21 +08:00
6fad15edae
Add tipc for MetaBIN
2023-02-28 15:01:21 +08:00
0034314376
support to train swin with dynamic2static
2023-02-14 10:46:57 +08:00
283cb12817
timeout 5m
2023-02-14 10:46:57 +08:00
ffca17ac6a
replace the dataset with full ILSVRC2012
2023-02-14 10:46:57 +08:00
e355f2e720
static training swin is not yet supported
2023-02-14 10:46:57 +08:00
2fc143f338
support to_static
2023-02-14 10:46:57 +08:00
ada7a3becf
ResNet50_amp_O2_ultra.yaml is not used by default.
2023-02-14 10:46:57 +08:00
2d66aeeb77
support specify model_type in config
...
such as norm_train and to_static_train
2023-02-14 10:46:57 +08:00
7a88058a2f
modify
2023-01-20 14:08:06 +08:00
add3ea238c
modify tipc
2023-01-20 14:08:06 +08:00
74de24f621
add swinv2
2023-01-20 14:08:06 +08:00
65e7a41a1d
add docs
2023-01-20 14:08:06 +08:00
62c9d0f45d
remove train benchmark for dsnet
2023-01-05 16:05:08 +08:00
d336944566
rm unset CUDA_VISIBLE_DEVICES before training
2022-12-29 11:44:27 +08:00
8d57bb31cf
modified DSNet config and TIPC txt
2022-12-22 14:55:46 +08:00
e069dedd7e
modified DSNet model names
2022-12-22 14:55:46 +08:00
28e094e097
add docs, config, model for DSNet
2022-12-22 14:55:46 +08:00
b61da414e5
add uniformer lst
2022-12-02 19:45:59 +08:00
d8f28bf320
[dy2static] fix bugs in MobileNetV1
2022-11-24 11:03:22 +08:00
ec2029b821
Add NextViT code and docs
2022-11-23 14:59:55 +08:00
509202a339
1. add dystatic config file for 6 additional models.
...
2. fix some bugs in test_train_dy2static_python.sh
2022-11-23 00:15:34 +08:00
96ebdcbd53
fix
2022-11-23 00:15:34 +08:00
34215e3873
change name of MobileNetV1
2022-11-23 00:15:34 +08:00
0e5cbd2b6a
[dy2static-tipc] add txt config for dy2static test
2022-11-22 11:17:38 +08:00
5e814882e3
change python3 -> python, otherwise < python3.6 is used.
2022-11-15 16:31:48 +08:00
efa7d3035f
make loss deterministic while running on gpu
2022-11-15 10:58:46 +08:00
5ad2817470
add statuscheck and follow the tipc_standard
2022-11-14 15:25:58 +08:00
c9f8b87ef4
fix by code review
2022-11-14 15:25:58 +08:00
0f6d71f021
[TIPC-Dy2St] add correctness tipc script
2022-11-14 15:25:58 +08:00
b033661bcf
tipc: VAN_tiny -> VAN_B0
2022-10-20 18:49:44 +08:00
cde853bfa7
Polish benchmark_train.sh for dynamicTostatic
2022-10-19 14:55:32 +08:00
d7014129f1
tipc: set DataLoader.Train.sampler.shuffle=False
...
benchmark needs the field to set Global.print_batch_step=1. Check file "./test_tipc/benchmark_train.sh#L76" for details.
2022-10-17 11:24:57 +08:00
4438acfda8
Merge pull request #2361 from TingquanGao/dev/tipc_fix_log_path
...
turn off the mkldnn and trt in ConvNeXt infer
2022-09-29 14:07:24 +08:00
9f64ff6c1c
tipc: turn off mkldnn and trt
...
it will time out when turn on trt or mkldnn
2022-09-29 05:22:44 +00:00
be434dc66c
tipc: fix the inference model name
2022-09-29 11:16:22 +08:00
b55791f696
tipc: styling code
2022-09-28 13:33:19 +00:00
091cb607f6
fix tipc log
2022-09-23 18:53:35 +08:00
8897aba3d0
fix tipc ptq log
2022-09-23 18:53:35 +08:00
69b53bc95f
tipc: modify the statistical range of log
...
the ips is unstable in the first 50 iter when training lite modle.
2022-09-21 13:58:00 +08:00
983d512a45
tipc: fix the keyword of ShiTu model
...
In benchmark_train mode, the config name is benchmark.txt. So the keyword of ShiTu should be model_name instead of the config name.
2022-09-19 14:30:54 +08:00
42758f5463
fix benchmark train's config for GeneralRecognitionV2, and polish TIPC docs
2022-09-14 08:54:22 +00:00
dafcb5cddf
tipc: fix log path for GeneralRecognition
2022-09-14 10:44:29 +08:00
b1c2350a26
[TIPC] add more scripts change, test=develop
2022-09-13 15:03:14 +08:00
ec21470a35
Merge pull request #2258 from TingquanGao/dev/tipc_fix_log_path
...
tipc: fix log path
2022-09-10 00:06:54 +08:00
7e239952a8
tipc: set num_workers to 4
...
The benchmark requires threads num to be set 4 when DALI is used for best performance.
And the param num_workers is not worked before commit 80c10bb995
2022-09-09 10:10:18 +00:00
0d9ddc0812
[TIPC] add scripts for npu and xpu, test=develop
2022-09-09 02:39:48 +08:00
0cea550455
tipc: fix log path
2022-09-08 10:57:03 +00:00
2700beea82
tipc: fix log path
2022-09-08 08:15:33 +00:00
fb669a05b8
tipc: fix log path
2022-09-06 13:25:09 +00:00
b3d26a987d
tipc: fix log path
2022-09-05 15:57:49 +00:00
44b55fb67c
tipc: paddle2onnx --enable_dev_version=False
2022-09-05 19:04:20 +08:00
4ee9dd49e0
fix small bug in prepare.sh
2022-09-04 14:22:44 +00:00
c449ba3420
add fleet chain and update inference_rec.yaml and corresponding output for feature_extraction.md
2022-09-04 14:20:07 +00:00
4620cfb9bc
fix typo in config files
2022-09-04 13:52:58 +00:00
94b48ac568
add cpp serving & cpp infer chain, update TIPC docs
2022-09-04 13:42:53 +00:00
38e78b0c81
add paddle2onnx_infer chain
2022-09-01 03:12:11 +00:00
a8278e84c4
add pact chain
2022-09-01 03:01:51 +00:00
6adee2720f
add ptq infer
2022-09-01 02:52:24 +00:00
d2ffc5e408
add lite_train_lite_infer chain
2022-08-31 17:28:48 +00:00
2f5c0c71da
fix: replace ResNet50.yaml with ResNet50_amp_O2_ultra.yaml when AMPO2
...
It is needed that use dali, NHWC and 4 channels when training ResNet50 with AMPO2
2022-08-26 10:59:16 +08:00
ea66adbec1
fix: need install paddleclas whl
2022-08-26 10:59:16 +08:00
5810695672
tipc: set num_wokers=12
2022-08-05 14:34:14 +08:00
b77a431451
tipc: set num workers=12 iff fp16
2022-08-05 14:34:14 +08:00
81b42f627b
tipc: rm useless bs
2022-08-05 14:34:14 +08:00
430fbd5df9
tipc: support fp32/fp16 training benchmark of PP-Shitu
2022-08-05 14:34:14 +08:00
a7f124042b
mv
2022-08-05 14:34:14 +08:00
483c0383e5
tipc: set bs=256 to avoid oom
2022-08-05 14:34:14 +08:00
b9de09e547
tipc: set num_wokers=12
2022-08-05 14:34:14 +08:00
7990fe5767
tipc: set num_wokers=8
2022-08-05 14:34:14 +08:00
3d1fb324a3
tipc: support AMP O2
2022-08-05 14:34:14 +08:00
595a6c5bcf
format code style
2022-08-05 14:34:14 +08:00
ac1c206d4a
tipc: support benchmark with fp32/fp16 training
2022-08-05 14:34:14 +08:00
b1f0d6d083
Merge pull request #1908 from flytocc/PeleeNet_PR
...
add PeleeNet
2022-07-14 17:50:54 +08:00
c5b5b34306
Merge pull request #2151 from RainFrost1/develop
...
修复benchmark中bs过大时导致log过少的问题
2022-07-14 14:21:40 +08:00
46532c446b
fix train list bug for benchmark
2022-07-14 03:35:50 +00:00
1fccb35be7
Merge branch 'develop' into PeleeNet_PR
2022-07-14 10:47:06 +08:00
1bbe4b5ed1
Merge pull request #1885 from flytocc/ConvNeXt
...
ConvNext & EMA & gradient accumulation
2022-07-14 10:45:25 +08:00
3dfc42bed0
set eval_during_train=False and save_interval=2 to reduce checkpoint storage
2022-07-13 19:36:32 +08:00
e66447b3a8
skip python inference when test for mainbody_det
2022-07-13 18:07:50 +08:00
56e8313fe4
update mainbody_det
2022-07-13 17:55:10 +08:00
4cf52bcc16
refine docs
2022-07-11 13:41:03 +08:00
fc0abc8765
refine docs
2022-07-11 13:39:20 +08:00
1ad1268958
refine docs
2022-07-11 13:35:56 +08:00
b7c771640f
add 12 model multi machine multi card training config in TIPC
2022-07-11 13:26:38 +08:00
436f64ad84
Merge branch 'develop' into ConvNeXt
2022-07-09 21:50:31 +08:00
74cc8b808c
add gpu0
2022-07-06 20:53:50 +08:00
ad7ad5c1ad
reduce GoogleNet's lr in normal and amp TIPC config to avoid inf output
2022-07-06 20:48:11 +08:00
1da2e2d36a
Merge branch 'PaddlePaddle:develop' into fix_amp_train
2022-07-06 15:44:13 +08:00
5f1c265a69
add train_amp doc
2022-07-06 15:41:57 +08:00
d4e286c3f3
Merge pull request #2132 from RainFrost1/develop
...
add ResNet50 log number in benchmark
2022-07-06 15:40:31 +08:00
f538a55bf2
add ResNet50 log number in benchmark
2022-07-06 07:06:07 +00:00
7494441082
remove existed symbolic link before ln -s
2022-07-06 14:26:09 +08:00
fa65357777
fix amp O2 train infer config
2022-07-05 22:49:20 +08:00
4f79d00c62
rename config -> configs
2022-07-04 15:26:28 +08:00
10cf8736cc
fix paddle2onnx for GeneralRecognition model
2022-07-04 15:10:46 +08:00
ad707bde4a
slim fleet_train config and fix paddle2onnx for GeneralRecognition model
2022-07-04 15:03:36 +08:00
2e56bec7d6
slim fleet_train config and fix paddle2onnx for GeneralRecognition model
2022-07-04 15:01:48 +08:00
3148f15dc0
remove ResNet59_vd and MobileNetV3_large_x1_0 configs' redundant pathes in config file, fix test_inference_cpp.sh
2022-07-02 13:30:59 +08:00
7b961b5950
remove MobileNetV3_large_x1_0 config redundant pathes in config file
2022-07-01 23:09:26 +08:00
fab572881f
reduce number of eval images to only one
2022-07-01 13:43:47 +08:00
68f636d735
simplify train_infer_python.txt configs
2022-07-01 13:23:00 +08:00
4b0dd292e3
debug
2022-06-28 06:54:50 +00:00
de1e9cab74
update pact chain
2022-06-28 04:06:54 +00:00
27dd1bc73c
merge develop and solve conflicts
2022-06-27 15:48:36 +08:00
5cc6c50ce6
Merge pull request #2104 from HydrogenSulfate/add_ampO2_train
...
Add amp O2 train
2022-06-27 15:45:30 +08:00
3ec0edc311
Merge pull request #2098 from HydrogenSulfate/add_main_KL
...
Add main model's KL chain
2022-06-27 15:45:12 +08:00
d59d68f2ed
debug
2022-06-24 15:49:15 +00:00
7990a12e22
Merge branch 'develop' into add_pact_chain
2022-06-24 22:24:45 +08:00
1e281f8c48
add PACT inference chain
2022-06-24 22:24:27 +08:00
e866a217bf
update configs
2022-06-24 18:37:01 +08:00
8cb2ec8d9c
Merge branch 'PaddlePaddle:develop' into add_ampO2_train
2022-06-24 16:11:14 +08:00
e7d9ba5874
update config and script
2022-06-24 04:25:56 +00:00
cf1e7747de
suppporting for custom PADDLEInfer download link
2022-06-23 20:06:01 +08:00
6ecaaba9fb
delete useless script
2022-06-23 07:51:12 +00:00
e6cea06326
format test_train_inference_python.sh
2022-06-23 07:48:17 +00:00
baab34788f
debug
2022-06-23 07:36:34 +00:00
4400aa2747
debug
2022-06-23 06:50:46 +00:00
480e40eb73
add main model KL transformation chain
2022-06-23 06:46:12 +00:00
2b758dec3f
add pact chain configs and refine related scripts
2022-06-22 17:45:58 +00:00
e77ba41e99
fix GPUID setting and LOG_PATH
2022-06-22 18:21:30 +08:00
9f1ec3a537
fix GPUID settting and correct LOG_PATH
2022-06-22 16:09:50 +08:00
43410aa852
update config, add amp eval for retrieval
2022-06-21 23:03:56 +08:00
f55172394b
fix
2022-06-21 12:41:54 +00:00
223872ae10
fix General_rec KL config
2022-06-21 12:05:50 +00:00
b2606a60fc
fix General_rec KL config
2022-06-21 12:02:53 +00:00
c0e100a3e0
merge develop
2022-06-21 11:16:45 +00:00
dfd0a0d076
add 5 more KL models
2022-06-21 05:57:06 +00:00
35d7b1f5f2
restore faiss installtion
2022-06-20 20:22:15 +08:00
31a736f7e7
move faiss installtion to prepare.sh
2022-06-17 20:02:24 +08:00
b56bb7fd41
fix prepare.sh
2022-06-17 16:37:34 +08:00
280174a647
debug
2022-06-16 19:55:12 +08:00
767433c681
debug
2022-06-16 16:39:33 +08:00
1143a4f217
fix
2022-06-16 15:45:38 +08:00
2702802427
Merge branch 'PaddlePaddle:develop' into add_kl_quant
2022-06-16 15:36:56 +08:00
ab8ff4d3d1
debug
2022-06-16 15:34:14 +08:00
1c3212876c
Merge pull request #2071 from RainFrost1/dynamic_benchmark
...
add r50 dynamic fp16 train for benchmark
2022-06-16 13:59:47 +08:00
dcd17d9437
Merge pull request #2067 from HydrogenSulfate/polish_prepare
...
polish prepare.sh and docs
2022-06-16 13:54:43 +08:00
bdcda20c30
Merge branch 'polish_prepare' into add_kl_quant
2022-06-16 13:33:23 +08:00
725c0da267
add kl_quant chain and polish prepare.sh
2022-06-16 12:21:54 +08:00
2c64a6469e
add kl_quant chain and polish prepare.sh
2022-06-16 11:42:30 +08:00
5827b2c9d1
add r50 dynamic fp16 train for benchmark
2022-06-16 03:12:00 +00:00
49f010a1c6
add r50 dynamic fp16 train for benchmark
2022-06-15 14:33:17 +00:00
eea897d2c3
polish test_serving* docs
2022-06-15 20:26:25 +08:00
8641aed670
polish prepare.sh and docs
2022-06-15 16:05:33 +08:00
ed69e5bc13
change bash to source in prepare.sh, make 'SERVING_BIN' effective in global context
2022-06-15 11:32:26 +08:00
3569d12656
remove gpu infer path in train_fleet_config, remove redundant config 'MobileNetV3_large_x1_0_lite_arm_cpu_cpp.txt' in PPLCNet
2022-06-14 16:14:42 +08:00
05c393d938
Merge pull request #2045 from HydrogenSulfate/add_cpp_serving
...
Add cpp serving chain
2022-06-14 11:41:56 +08:00
5010bee691
Merge pull request #1988 from FeixLiu/update_dataloader
...
[bug fixer] Update dataloader to handle index error
2022-06-14 11:05:29 +08:00
ed523bebab
split test_serving_infer.sh into test_serving_infer_cpp.sh and test_serving_infer_python.sh
2022-06-14 10:51:17 +08:00
tianyi1997
gaotingquan
zhangyubo0722
cuicheng01
LittleMoon
xiongkun
0x45f
cuicheng01
zhengya01
HydrogenSulfate
qili93
root
Walter
Yang Nie