 FlyingQianMM
|
639820c46e
Merge pull request #875 from FlyingQianMM/develop_export
|
vor 4 Jahren |
|
FlyingQianMM
|
47aab15235
add fixed_input_shape to self
|
vor 4 Jahren |
 will-jl944
|
fd2093e6b6
train and quant_aware_train support resume_checkpoint
|
vor 4 Jahren |
|
will-jl944
|
491c5b88b2
not necessary to check shm size on non-Linux systems
|
vor 4 Jahren |
|
FlyingQianMM
|
3512d7d5c5
Merge branch 'develop' of https://github.com/PaddlePaddle/PaddleX into develop_fix
|
vor 4 Jahren |
|
will-jl944
|
ed65b01157
convert tensor to list while saving pruner_inputs
|
vor 4 Jahren |
|
will-jl944
|
bd785f12c0
no need to skip conv2d with small channel number with paddleslim 2.1
|
vor 4 Jahren |
|
will-jl944
|
2275e8ea59
refine multi-card evaluation logistics, elaborate return value of evaluate only on card 0
|
vor 4 Jahren |
|
FlyingQianMM
|
6973e5994d
fix conflicts
|
vor 4 Jahren |
|
FlyingQianMM
|
4afdbdca12
fix input_shape
|
vor 4 Jahren |
|
will-jl944
|
9ec76631c0
add confusion matrix computation to seg
|
vor 4 Jahren |
|
will-jl944
|
62a0862f81
deal with comments
|
vor 4 Jahren |
|
will-jl944
|
2213969dce
remove image_shape from quant_aware_train since no inference model is saved in qat process
|
vor 4 Jahren |
|
will-jl944
|
c365a853ba
qat
|
vor 4 Jahren |
|
will-jl944
|
598a8d446d
qat
|
vor 4 Jahren |
|
will-jl944
|
620898b48a
quantization aware training
|
vor 4 Jahren |
|
will-jl944
|
e1f495f55e
set pretrain_weights to None in qat
|
vor 4 Jahren |
|
will-jl944
|
fce21d27e7
add qat
|
vor 4 Jahren |
|
will-jl944
|
a94d27c639
compatible with static apis
|
vor 4 Jahren |
|
will-jl944
|
62a1b636ec
add ema
|
vor 4 Jahren |
|
will-jl944
|
dc16b046b6
log only if local_rank is 0
|
vor 4 Jahren |
|
will-jl944
|
79d4164c5c
move logging info out of build_dataloader
|
vor 4 Jahren |
|
will-jl944
|
8fa1234d5d
import from PaddleClas submodule instead
|
vor 4 Jahren |
|
FlyingQianMM
|
d4f745add9
add dygrah
|
vor 4 Jahren |
