17d6f1c2f9
add option for graph kernel and mixed precision
4 years ago
7131707359
!9366 fix bert export bug
...
From: @yuzhenhua666
Reviewed-by: @c_34
Signed-off-by: @c_34
4 years ago
0d1f209d69
fix bert export bug
4 years ago
0c7ba7a7fa
!9354 mode_mindir_suffix
...
From: @bai-yangfan
Reviewed-by: @c_34,@guoqi1024
Signed-off-by: @c_34
4 years ago
c46c4dffe4
mindir_suffix
4 years ago
b1189cd118
compatible name
4 years ago
a744ef9113
add squad
4 years ago
bde36af04c
Modify the default activation function for bert
4 years ago
0aa63f21c5
!9039 register FastGelu for activation
...
From: @shibeiji
Reviewed-by:
Signed-off-by:
4 years ago
a1db640211
fix bert bug and add export file for yolov3 and add test file for maskrcnn
4 years ago
cd850bd210
register activation operator fast_gelu
4 years ago
d0f05a7670
add export script for vgg16,bert. fix faster-rcnn and inception bug
4 years ago
4851a67bb5
add layer of clipbyglobalnorm.
4 years ago
acb44b91b9
edit readme
4 years ago
f3860f5111
Fix typo of hccl time out options
4 years ago
1464bf3dd8
add hccl_time_out options in bert distribute launcher
4 years ago
263a591473
edit readme
4 years ago
9cee0d2143
Add num_epochs to non-sink training
4 years ago
bd5a8b42ac
!6724 redundant codes clean
...
Merge pull request !6724 from TuDouNi/master
4 years ago
a712489a28
!6643 Modifying the hardware information in Readme.
...
Merge pull request !6643 from linqingke/modelzoo
4 years ago
4e85071055
redundant codes clean
4 years ago
d6110a3cd5
!6656 move batch_size from net_cfg to outer cfg
...
Merge pull request !6656 from yoonlee666/hub
4 years ago
25b2c1944f
tinybert hub
4 years ago
01c9e8b373
add tokenization and score file
4 years ago
3f7a5ed394
README A+X to A+K
4 years ago
10add158f8
!6592 bugfix bert config
...
Merge pull request !6592 from yoonlee666/master
4 years ago
92d0a29911
bugfix bert config
4 years ago
ac627e046b
update links for README
...
Signed-off-by: Ting Wang <kathy.wangting@huawei.com>
4 years ago
31ad1654a1
!6267 delete redundant codes in model zoo
...
Merge pull request !6267 from zhaoting/clean_warnings
4 years ago
59d9f8bf6a
remove importing _selected_ops from model_zoo
4 years ago
cc4859f0b2
!6358 fix loss print in bert, change epoch number to integer
...
Merge pull request !6358 from chenhaozhe/fix-loss-print
4 years ago
d421f49e60
delete redundant codes
4 years ago
db0c0caf10
!6393 bugfix delete batch_size
...
Merge pull request !6393 from yoonlee666/master
4 years ago
1a8dffe4c6
change epoch to integer while print loss
4 years ago
c95ed54fe1
!5239 reduce cyclomatic complexity in model zoo
...
Merge pull request !5239 from zhaoting/master
4 years ago
d90ac3c6d1
bugfix
4 years ago
3671244ff8
!6233 move batch_size from bert_cfg_cfg to cfg
...
Merge pull request !6233 from yoonlee666/master
4 years ago
069a1cfbd7
!6256 modify ckpt path
...
Merge pull request !6256 from TuDouNi/master
4 years ago
528072f45f
move batch_size from bert_cfg_cfg to cfg
4 years ago
c8200af525
modify the ckpt path
4 years ago
a4a65ffe06
reduce cyclomatic complexity
4 years ago
dc78af7365
fix bug in bert
4 years ago
105422fbb9
!6188 fix loss print of bert, add attention about bind cores in README
...
Merge pull request !6188 from chenhaozhe/fix-bert-loss-print
4 years ago
8b0793eb84
!6137 modify the model_zoo ckpt path
...
Merge pull request !6137 from TuDouNi/master
4 years ago
7578f23632
optim TrainOneStepCell
4 years ago
e1f4c066b3
fix loss print in bert and corresponding downstream task
4 years ago
ec947ebf3d
modify the ckpt path
4 years ago
4e587420c0
fix hccl environ in get_distribute_pretrain_cmd.py
4 years ago
91c65a734a
fix some doc error
4 years ago
326feab42f
!6067 Generate cmd file for distributed_pretrain in bert scripts
...
Merge pull request !6067 from chenhaozhe/fix-bert-codex
4 years ago
d8c09067ce
!5801 [MD][Feature] minddata iterator output ms_tensor
...
Merge pull request !5801 from xiefangqi/md_modify_output_tensor
4 years ago
28cb0da168
remove os.system in launch scripts
4 years ago
9b3c33e157
minddata iterator output ms_tensor
4 years ago
be62fd7fa6
!5768 Save the GPU backend multi card output in different folders.
...
Merge pull request !5768 from linqingke/fasterrcnn
4 years ago
92e99ff224
change map calls
4 years ago
845ff342ed
Save the GPU backend multi card output in different folders.
4 years ago
994cfcb8a0
!5914 edit bert hub config file and tinybert script bugfix
...
Merge pull request !5914 from yoonlee666/master
5 years ago
cd07acc8a4
!5903 add set_grad() for pynative mode in model_zoo networks
...
Merge pull request !5903 from lvchangquan/master
5 years ago
16c2c17436
edit bert hub config
5 years ago
3bac9d3713
switch input columns and operation
...
change ImagefolderDV2 name
change ds.transforms.vision to ds.vision
change batch api to match map api more closely
compose op changes
test_pylint
remove compose op from vision, move to transform module, refactor map and batch to use column_order
5 years ago
7abe692fb5
add set_grad() for pynative mode in model_zoo network
5 years ago
0a2f472578
add hub config for bert base and nezha
5 years ago
f2d3fd34ce
rectification_allreduce_fusion_api
5 years ago
d4cfe55c04
rename mirror_mean to gradients_mean
5 years ago
4d9d8c3e74
Modelzoo interface change.
5 years ago
dfd85caa1b
delete enable_fused_layernorm
5 years ago
44a9c25251
!5632 Add clip_by_global_nrom in bert
...
Merge pull request !5632 from chenhaozhe/add-global-norm-to-bert
5 years ago
ac95836257
add global norm in bert
5 years ago
d5e02cf474
!5584 delete the redundant argument while initializing the class of GradOperation
...
Merge pull request !5584 from shibeiji/master
5 years ago
d57960ed4c
delete the redundant argument while initializing class of GradOperation
5 years ago
ff4e02a09d
bert script bugfix
5 years ago
3f1e05881d
!5474 bert script bugfix
...
Merge pull request !5474 from yoonlee666/bugfix
5 years ago
954f53da9e
enhancement
5 years ago
e6a4d932b4
!5350 [AutoParallel]Rectification distributed init
...
Merge pull request !5350 from lichen/rectification_init
5 years ago
d3e55b543e
rectification init
5 years ago
c247f5cdbf
!5325 add schema file introduce for BERT and TinyBERT
...
Merge pull request !5325 from wanghua/master
5 years ago
f347d1c9e7
add schema for BERT and TinyBERT
5 years ago
07117e4dd4
mv ParallelMode to context
5 years ago
1a54785fe2
remove name arg from gradoperation
5 years ago
40fc11e9a4
Mimic higher batch size by accumulating gradients N times before weight update
5 years ago
a40cc12fae
modify BERT and TinyBERT README.md
5 years ago
7cec00b491
!4577 Optimize the print of bert
...
Merge pull request !4577 from chenhaozhe/optimize-print-of-bert
5 years ago
cc9bd8a3e5
!4182 add_tf_and_ms_checkpoint_transfer_tools
...
Merge pull request !4182 from Guomenghao319/add_tf_and_ms_checkpoint_transfer_tools
5 years ago
fa10a4e483
optimize print of bert scripts'
5 years ago
32846d791b
bugfix bert script
5 years ago
1f7d72dd2a
!4227 fix pylint warning in model_zoo
...
Merge pull request !4227 from panbingao/pylintfix
5 years ago
3f60bf6dfe
fix pylint warnings in model_zoo
5 years ago
a5ac2427a7
bugfix bert script
5 years ago
52f2d581cb
add_tf_and_ms_checkpoint_transfer_tools
5 years ago
b535b6427b
!4122 fix bug in bert script
...
Merge pull request !4122 from yoonlee666/master_bug
5 years ago
e18c7292d6
bug fix
5 years ago
776d094c5b
quant export fix up for atc tools
5 years ago
7113f1f22c
!3857 modify bert and tinybert scripts and README
...
Merge pull request !3857 from wanghua/master
5 years ago
89fa2d3708
modify bert and tinybert readme
5 years ago
59530248a3
!3795 remove old MINDSPORE_HCCL_CONFIG_PATH in model zoo 2
...
Merge pull request !3795 from panbingao/master
5 years ago
6e21618fe1
!3798 add shuffle switch for finetune dataset
...
Merge pull request !3798 from yoonlee666/master-finetune
5 years ago
1165b27f41
!3693 Change the column order and add drop_reminder option to make the generated dataset compatible with BertCLS
...
Merge pull request !3693 from dessyang/master
5 years ago
9bdece71d4
add switch for data shuffle
5 years ago
98b76b9020
remove old MINDSPORE_HCCL_CONFIG_PATH in model zoo 2
5 years ago
4307c1fa61
change the column order and add drop_reminder option to make this script compatible with BertCLS model
5 years ago
tronzhang
mindspore-ci-bot
yuzhenhua
bai-yangfan
caozhou
yoonlee666
shibeiji
jzg
chenhaozhe
hesham
root
linqingke
yoonlee666
Ting Wang
dayschan
zhaoting
wangnan39@huawei.com
xiefangqi
nhussain
lvchangquan
lichenever
yao_yf
wanghua
panyifeng
panbingao
GuoMengHao
Wei Luning
dessyang