X2Paddle Versions Save

Deep learning model converter for PaddlePaddle. (『飞桨』深度学习模型转换工具)

v1.4.1

1 year ago

What's Changed

[Bug]Fixed ONNXDecoder by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/939
Add ONNX2Paddle for visualdl by @rainyfly in https://github.com/PaddlePaddle/X2Paddle/pull/954
[Bug]Support YOLOv8-cls by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/952

New Contributors

@rainyfly made their first contribution in https://github.com/PaddlePaddle/X2Paddle/pull/954

v1.4.0

1 year ago

What's Changed

Add input_shape_dict and rm raw_input by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/899
Fixed ToPILImage && rm SymbolicShapeInference by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/903
Support GFPGAN by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/906
Add FastDeploy YOLOv5/v6/v7 demo in readme by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/911
Updated torch.nn.utils.clip_grad_value_.md by @PommesPeter in https://github.com/PaddlePaddle/X2Paddle/pull/918
[Bug] Fixed ONNX Gemm bug by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/917
[Bug] Fixed typo error by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/920
Support fmod=1 in Mod OP by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/921
Fixed NonZero bug by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/922
add optim by @AndPuQing in https://github.com/PaddlePaddle/X2Paddle/pull/927
Support YOLOv8 for dynamic shape by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/931

New Contributors

@PommesPeter made their first contribution in https://github.com/PaddlePaddle/X2Paddle/pull/918
@AndPuQing made their first contribution in https://github.com/PaddlePaddle/X2Paddle/pull/927

v1.3.9

1 year ago

What's Changed

Support 7-15 ONNX opset version by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/838
Modify opset7~15 inheritance relationship by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/843
Add autoscan test by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/836
Add autoscan test for Elementwise ops by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/854
Add_test_abs by @qqj1130247885 in https://github.com/PaddlePaddle/X2Paddle/pull/859
Add logical and compare ops by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/855
Fixed torch pad by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/863
Solve the overload problem by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/865
Add AveragePool tests by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/872
Add Reduce_ops and HardSigmoid by @qqj1130247885 in https://github.com/PaddlePaddle/X2Paddle/pull/867
Add isinf by @qqj1130247885 in https://github.com/PaddlePaddle/X2Paddle/pull/868
Add IsNaN by @qqj1130247885 in https://github.com/PaddlePaddle/X2Paddle/pull/869
Add Sum by @qqj1130247885 in https://github.com/PaddlePaddle/X2Paddle/pull/873
Fixed Readme by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/879
Fixed reduce tests bug by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/880
Set enable_code_optim default to False by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/889

v1.3.8

1 year ago

What's Changed

Add HuggingFace Model Conversion Tutorial by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/823
Add aten::upsample_trilinear3d by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/817
Fixed misspell by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/821
Add sysmpy in requirment by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/824
Fixed Misspell by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/829
Fixed rename bug by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/820
Fix ONNX resize op by @qqj1130247885 in https://github.com/PaddlePaddle/X2Paddle/pull/830
Fix bug in docs directory about code combination of save_image by @ZoruaSama in https://github.com/PaddlePaddle/X2Paddle/pull/833
Remove scale for add by @wjj19950828 in https://github.com/PaddlePaddle/X2Paddle/pull/835

New Contributors

@qqj1130247885 made their first contribution in https://github.com/PaddlePaddle/X2Paddle/pull/830
@ZoruaSama made their first contribution in https://github.com/PaddlePaddle/X2Paddle/pull/833

Full Changelog: https://github.com/PaddlePaddle/X2Paddle/compare/v1.3.7...v1.3.8

v1.3.7

1 year ago

新增功能

全面支持 HuggingFace 模型，包括Bert、Albert等热点模型以及对应不同任务类型，共55种 TorchScript 以及40种 ONNX 模型
针对 Bert结构 ONNX模型，新增 Gelu pass，预测速度提升20%以上 https://github.com/PaddlePaddle/X2Paddle/pull/809
针对 Bert结构 ONNX模型，新增 div2scale pass，使其命中PaddleInference multihead_matmul_fuse_pass，性能提升30%以上 https://github.com/PaddlePaddle/X2Paddle/pull/813

新增模型支持

PyTorch->Paddle

ONNX->Paddle

新增op

新增PyTorch op（7个）：aten::rsqrt、aten::replication_pad1d、aten::full、 aten::group_norm、aten::argmax、aten::copy、aten::linear https://github.com/PaddlePaddle/X2Paddle/pull/783 https://github.com/PaddlePaddle/X2Paddle/pull/781 https://github.com/PaddlePaddle/X2Paddle/pull/786 https://github.com/PaddlePaddle/X2Paddle/pull/791 https://github.com/PaddlePaddle/X2Paddle/pull/789 https://github.com/PaddlePaddle/X2Paddle/pull/769
新增ONNX op（2个）：CumSum 、LessOrEqual https://github.com/PaddlePaddle/X2Paddle/pull/793 https://github.com/PaddlePaddle/X2Paddle/pull/794

修复op

修复PyTorch op（2个）：aten::masked_fill、aten::index https://github.com/PaddlePaddle/X2Paddle/pull/786 https://github.com/PaddlePaddle/X2Paddle/pull/787
修复ONNX op（12个）：ConvTranspose 、Unsqueeze 、MatMul 、Gather 、Slice 、Squeeze 、Flatten、Constant 、ConstantOfShape 、Where 、Topk、NonZero https://github.com/PaddlePaddle/X2Paddle/pull/792 https://github.com/PaddlePaddle/X2Paddle/pull/794 https://github.com/PaddlePaddle/X2Paddle/pull/796 https://github.com/PaddlePaddle/X2Paddle/pull/797 https://github.com/PaddlePaddle/X2Paddle/pull/805 https://github.com/PaddlePaddle/X2Paddle/pull/808

修复功能

修复LN fuser pass https://github.com/PaddlePaddle/X2Paddle/pull/777
修复Linear fuser pass https://github.com/PaddlePaddle/X2Paddle/pull/800

其他

在转换完成之后添加问卷链接 https://github.com/PaddlePaddle/X2Paddle/pull/801
清除paddle.fluid接口 https://github.com/PaddlePaddle/X2Paddle/pull/806

v1.3.6

2 years ago

新增功能

新增转换记录统计功能 https://github.com/PaddlePaddle/X2Paddle/pull/766
针对 ONNX Transformer 模型，新增 LayerNorm pass，优化预测速度 https://github.com/PaddlePaddle/X2Paddle/pull/761
HuggingFace 参数迁移功能 https://github.com/PaddlePaddle/X2Paddle/pull/748

新增模型支持

PyTorch->Paddle

图像修复算法：Saicinpainting_LaMa

ONNX->Paddle

SOTA backbone：SwinTransformer
关键点检测算法：Blazepose

Caffe->Paddle

视频分割算法：SegFlow

TensorFlow->Paddle

图像生成算法：Pix2Pix

新增op

新增PyTorch op（5个）：aten::complex、aten::real、aten::imag、aten::fft_rfftn、aten::fft_irfftn https://github.com/PaddlePaddle/X2Paddle/pull/729
新增ONNX op（5个）：Sin、Cos、Neg、SpaceToDepth、GatherElement https://github.com/PaddlePaddle/X2Paddle/pull/723
新增PyTorch-Paddle api（2个）：gather、scatter https://github.com/PaddlePaddle/X2Paddle/pull/726 https://github.com/PaddlePaddle/X2Paddle/pull/730

修复op

修复PyTorch op（1个）：aten::_convolution https://github.com/PaddlePaddle/X2Paddle/pull/725
修复ONNX op（7个）：pad、resize、Upsample 、Gather 、Slice 、convtranspose 、Split https://github.com/PaddlePaddle/X2Paddle/pull/716 https://github.com/PaddlePaddle/X2Paddle/pull/720 https://github.com/PaddlePaddle/X2Paddle/pull/722 https://github.com/PaddlePaddle/X2Paddle/pull/723 https://github.com/PaddlePaddle/X2Paddle/pull/749 https://github.com/PaddlePaddle/X2Paddle/pull/751
修复Caffe op（3个）：scale、crop、Pow https://github.com/PaddlePaddle/X2Paddle/pull/721 https://github.com/PaddlePaddle/X2Paddle/pull/757
修复TensorFlow（3个）：FusedBatchNorm、SpaceToBatchND、BatchToSpaceND https://github.com/PaddlePaddle/X2Paddle/pull/738 https://github.com/PaddlePaddle/X2Paddle/pull/741

修复功能

优化转换过程中的 log 提示 https://github.com/PaddlePaddle/X2Paddle/pull/764
更新外部框架一键转换为 Paddle Lite 模型文档，新增lite_valid_places选择类型 https://github.com/PaddlePaddle/X2Paddle/pull/758
PyTorch2Paddle 动态shape转换指引 https://github.com/PaddlePaddle/X2Paddle/pull/732

Contributors

Thanks to external developers for their contributions to open source projects！@geoyee @HUSTLX

v1.3.5

2 years ago

新增功能

添加enable_code_optim开关，解决部分场景下代码优化可能导致转换出错的问题 https://github.com/PaddlePaddle/X2Paddle/pull/696

新增模型支持

PyTorch->Paddle

首次支持声音克隆模型：MockingBird
轻量级检测算法：YOLOX

新增op

新增PyTorch op（5个）：aten::repeat_interleave、aten::maxpool1d、aten::frobenius_norm、aten::silu、aten::format https://github.com/PaddlePaddle/X2Paddle/pull/691 https://github.com/PaddlePaddle/X2Paddle/pull/696
新增ONNX op（1个）：GatherND https://github.com/PaddlePaddle/X2Paddle/pull/645

修复op

修复PyTorch op（5个）：aten::div、aten::index_select、aten::select、aten::upsample_nearest2d、aten::_convolution https://github.com/PaddlePaddle/X2Paddle/pull/691 https://github.com/PaddlePaddle/X2Paddle/pull/689
修复ONNX op（3个）：Split、Conv、Pad https://github.com/PaddlePaddle/X2Paddle/pull/645 https://github.com/PaddlePaddle/X2Paddle/pull/707 https://github.com/PaddlePaddle/X2Paddle/pull/687
修复PyTorch训练代码转换（2个）：ToPILImage、Split https://github.com/PaddlePaddle/X2Paddle/pull/690 https://github.com/PaddlePaddle/X2Paddle/pull/694

修复功能

Windows环境生成代码问题 https://github.com/PaddlePaddle/X2Paddle/pull/706
修复ONNX和PyTorch版本问题 https://github.com/PaddlePaddle/X2Paddle/pull/708
去除代码优化部分to_tensor操作，优化预测速度 https://github.com/PaddlePaddle/X2Paddle/pull/696

Contributors

Thanks to external developers for their contributions to open source projects！@littletomatodonkey @zhoucz97 @firedent @dawn1206

v1.3.4

2 years ago

修复功能

修复X2Paddle调用ResNet以及Vgg预训练模型bug(https://github.com/PaddlePaddle/X2Paddle/pull/684)

v1.3.3

2 years ago

新增功能

集成Paddle-Lite opt工具，提供一键转换API(https://github.com/PaddlePaddle/X2Paddle/pull/681)

新增op

新增PyTorch op转换支持（3个）：aten::bitwise_not、aten::bitwise_xor、aten::bitwise_and(https://github.com/PaddlePaddle/X2Paddle/pull/677)
新增ONNX op转换支持（5个）：Max、Min、GreaterOrEqual、GatherND、And(https://github.com/PaddlePaddle/X2Paddle/pull/682)

修复op

修复ONNX op转换支持（4个）：ReduceSum、Unsqueeze(https://github.com/PaddlePaddle/X2Paddle/pull/682) Clip、PRelu(https://github.com/PaddlePaddle/X2Paddle/pull/680)
修复Caffe op转换支持（1个）：pooling(https://github.com/PaddlePaddle/X2Paddle/pull/681)

v1.3.2

2 years ago

新增功能

新增PyTorch op映射（1个）：aten::sum

修复功能

修复PyTorch op映射（2个）：aten::instance_norm、aten::split
修复Caffe op映射（3个）：directly_map, Eltwise and crop
修复ONNX版本判断bug