76 Commits (867ff7ae97e00a91c346cd38e10b2a5169377c71)

Author SHA1 Message Date
  nihuini 253e505765 improve compatibility with onnx slice opset version 10+ 6 years ago
  nihuini 9b09cc16b5 fix compability with onnx clip opset11, fix #1609 6 years ago
  nihui 14b8b68c2c
onnx lstm (#1613) 6 years ago
  nihui 9929d52885 less duplicated code for crop layer, slice axes starts from 0 6 years ago
  nihuini be3f978770 fix onnx opset11 Pad conversion, improve Resize conversion, fix #1556 6 years ago
  nihui 30c2954160 fuse onnx pixelshuffle 6 years ago
  Anton Kochkov 07170542c9 Fix GCC 9.x warnings (#1462) 6 years ago
  nihuini 84a5d5a502 improve onnx shufflechannel compability with torchvision model 6 years ago
  nihuini 5eae1b8325 always write normalize scale data, do not break optimize tool 6 years ago
  nihuini af3643ed81 normalize eps mode for different training framework conventions 6 years ago
  nihuini c95e7bbb3c onnx fuse flatten 6 years ago
  nihuini 253e54351d onnx fuse normalize, first try 6 years ago
  nihuini e5a834b299 do not fuse weight reshape 6 years ago
  nihuini 66b8a7ee36 fuse onnx unsqueeze-prelu 6 years ago
  nihuini 6ce5da28a0 convert onnx pooling ceil_mode 6 years ago
  Guoxia Wang 4336cc15f7 fix bug that MemoryData as multi operations input (#1357) 6 years ago
  nihuini d5d7886517 resolve onnx Split conversion with empty split attribute 6 years ago
  Guoxia Wang 51bca0b718 fix onnx2ncnn MemoryData dims problem #1330 (#1331) 6 years ago
  nihuini 517d1b2053 improve compability with megvii shufflenet model 6 years ago
  nihuini e8bb88830d convert mxnet squeeze expanddims, convert onnx squeeze unsqueeze 6 years ago
  nihuini a6c60068e6 convert numpy style slice to crop 6 years ago
  bindog 04b4b02324 [WIP] add reduce op support for onnx (#1308) 6 years ago
  bindog 53d54115a8 add split op support for onnx2ncnn (#1299) 6 years ago
  nihuini c3eea4cc0b fuse onnx unsqueeze-batchnorm-squeeze as batchnorm 6 years ago
  nihuini 7d3ebe2d69 %g may output number without dot, stick to %e 6 years ago
  nihuini d67d2531ae convert some well-known noop, printf %g to preserve precision 6 years ago
  nihuini 296e0022df deconvolution output adj and output shape 6 years ago
  nihuini 0e26e3094e autopad SAME_LOWER 6 years ago
  nihuini 18833c9286 convert asymmetric pad parameter 6 years ago
  nihuini f5ca997a66 fuse x * hardsigmoid to hardswish 6 years ago
  nihuini 46bfb4bca9 write hardswish parameter 6 years ago
  nihuini da6a42ed96 fix onnx2ncnn build 6 years ago
  nihuini 012fdd46de convert onnx hardsigmoid hardswish, op chain fusion code reconstructed 6 years ago
  nihuini eced73c186 convert mxnet onnx pooling count_include_pad 6 years ago
  Howave ac60acfad0 Fix for onnx2ncnn Div op (#1187) 6 years ago
  nihuini 1b910efea5 convert slice properly 6 years ago
  nihuini ad9e0a9725 read reshape shape data from raw_data, fuse onnx reshape-transpose-reshape to shufflechannel 6 years ago
  Chang Si d3be711c52 fix onnx2ncnn bug in Slice layer 6 years ago
  wincle fb9309e460 implement reflect padding. 6 years ago
  wincle d1a1a0dbcd fix onnx2ncnn padding bug. fix #1103 6 years ago
  wincle bb6b93568d Tanh Support 6 years ago
  Arnaud Brejeon e30464cdba Fix bug #1093 (#1094) 6 years ago
  daquexian dbe8a32082 Add onnx resize op 7 years ago
  ShuangLiu1992 98bb0bb243 Update onnx2ncnn.cpp (#948) 7 years ago
  Gemfield 506d9f74aa eliminate the warning msg when compile the onnx2ncnn (#850) 7 years ago
  Gemfield d91d7419eb Add upsample linear mode (#846) 7 years ago
  Gemfield 5407deb5ce Apply splitcnn function to graph input (#844) 7 years ago
  Gemfield 582ea371ce Enhance upsample scale value retrieval with raw_data API (#839) 7 years ago
  nihuini 3b404dcedd fuse transpose(weight)+matmul to matmul, fix #620 7 years ago
  nihuini e811778a4a trival out-of-index fix 7 years ago