Add op read_file and decode_jpeg #32564
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
qingqing01
reviewed
Apr 28, 2021
| "of the JPEG image. It is a tensor with rank 1."); | ||
| AddOutput("Out", "The output tensor of DecodeJpeg op"); | ||
| AddComment(R"DOC( | ||
| This operator decode a JPEG image into a 3 dimensional RGB Tensor. |
There was a problem hiding this comment.
下面的注释是支持灰度图的吧，这里需要正确的comments
| OP_INOUT_CHECK(ctx->HasOutput("Out"), "Output", "Out", "DecodeJpeg"); | ||
|
|
||
| auto out_dims = std::vector<int>(1, -1); | ||
| ctx->SetOutputDim("Out", framework::make_ddim(out_dims)); |
There was a problem hiding this comment.
out_dims的rank是3吧，这里的shape不正确， 依据mode， 是 {1, -1, -1} 或 {3，-1， -1}
|
|
||
| static nvjpegHandle_t nvjpeg_handle = nullptr; | ||
|
|
||
| void init_nvjpegImage(nvjpegImage_t* img) { |
| int height = heights[0]; | ||
|
|
||
| nvjpegOutputFormat_t outputFormat; | ||
| int outputComponents; |
There was a problem hiding this comment.
outputFormat -> output_format
outputComponents -> output_components
| "The provided mode is not supported for JPEG files on GPU")); | ||
| } | ||
|
|
||
| nvjpegImage_t outImage; |
|
|
||
| nvjpegStatus_t decode_status = platform::dynload::nvjpegDecode( | ||
| nvjpeg_handle, nvjpeg_state, x_data, x->numel(), outputFormat, | ||
| &outImage, stream); |
There was a problem hiding this comment.
Maybe need another cuda stream
| out->Resize(framework::make_ddim(out_shape)); | ||
|
|
||
| uint8_t* data = out->mutable_data<T>(ctx.GetPlace()); | ||
| std::memcpy(data, reinterpret_cast<uint8_t*>(image_data.data()), file_size); |
| platform::errors::InvalidArgument( | ||
| "Output(Out) of ReadFileOp is null.")); | ||
|
|
||
| auto out_dims = std::vector<int>(1, -1); |
There was a problem hiding this comment.
rank是1的话，这里的out_dims是 {-1}, （1，-1)代表的rank是2
| @@ -0,0 +1,67 @@ | |||
| # Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. | |||
python/paddle/vision/ops.py
Outdated
|
|
||
| def decode_jpeg(x, mode='unchanged', name=None): | ||
| """ | ||
| Decodes a JPEG image into a 3 dimensional RGB Tensor. |
There was a problem hiding this comment.
Decodes a JPEG image into a 3 dimensional RGB Tensor or 1 dimensional Gray Tensor
lyuwenyu
reviewed
Apr 29, 2021
| int sz = widths[0] * heights[0]; | ||
|
|
||
| auto* out = ctx.Output<framework::LoDTensor>("Out"); | ||
| std::vector<int64_t> out_shape = {output_components, height, width}; |
There was a problem hiding this comment.
function make_dims only accept type int64_t
| input.seekg(0, std::ios::beg); | ||
|
|
||
| auto* out = ctx.Output<framework::LoDTensor>("Out"); | ||
| std::vector<int64_t> out_shape = {file_size}; |
qingqing01
approved these changes
Apr 29, 2021
|
|
||
| nvjpegStatus_t decode_status = platform::dynload::nvjpegDecode( | ||
| nvjpeg_handle, nvjpeg_state, x_data, x->numel(), output_format, | ||
| &out_image, nvjpeg_stream); |
There was a problem hiding this comment.
需要调用nvjpegJpegStateDestroy释放nvjpeg_state
| namespace operators { | ||
|
|
||
| static cudaStream_t nvjpeg_stream = nullptr; | ||
| static nvjpegHandle_t nvjpeg_handle = nullptr; |
LielinJiang
added a commit
to LielinJiang/Paddle
that referenced
this pull request
Apr 29, 2021
* add op read_file and decode_jpeg
XiaoguangHu01
pushed a commit
that referenced
this pull request
Apr 30, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
New features
PR changes
APIs
Describe
Add op read_file and decode_jpeg
file_path -> gpu tensor的时间比较（单位：s/1000次）
