 will-jl944
|
e19567b9ca
add picodet export
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
ad5c61f28a
disable mkl-dnn for maskrcnn
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
1bbb8fecf9
Merge branch 'develop' of https://github.com/PaddlePaddle/PaddleX into hubsrv_jf
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
eef4b3e2d9
fix timer bug
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
838809a031
add hubserving
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
ba23453e3b
add batch_predict, which is alias of predict
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
34b2ca4873
disable trt acceleration for rcnn and segmentation models
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
a2804bf07c
optimize seg python deploy postprocess
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
743ea5fef1
glog does not need to be ebabled explictly
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
5d3a191a08
set use_mkl to true by default
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
1a24bb2b37
Merge remote-tracking branch 'origin/develop' into deploy_jf
|
%!s(int64=4) %!d(string=hai) anos |
 FlyingQianMM
|
6dba2ab54e
Merge pull request #1067 from will-jl944/develop_jf
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
05f6c194d0
refine cpp seg deploy postprocess
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
3758a44928
refine cpp seg deploy
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
eb2e2952cc
rcnn models support prediction w/ bs greater than 1
|
%!s(int64=4) %!d(string=hai) anos |
|
FlyingQianMM
|
0baa55fd29
add python deployment docs
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
b3bd512d18
merge develop
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
7f055dc4eb
add warmup for python deploy
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
077bc33bbf
add warmup for python deploy
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
0bd41d9bb5
add warmup for python deploy
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
cdf7496000
deal with comments
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
a1ca94c3a8
add fake batch prediction for rcnn models
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
8df2fcd3ec
support loading exported infer mdoel
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
5daf6c287d
add python deploy
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
18415bcabc
add softmax and argmax while exporting models
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
683a44be2b
add python deploy
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
d05f2ad919
models can be initilized without building net
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
76cb346e5f
solve conflict
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
8af180a94b
support loading exported infer mdoel
|
%!s(int64=4) %!d(string=hai) anos |
|
will-jl944
|
cac8f037d0
add python deploy
|
%!s(int64=4) %!d(string=hai) anos |
