You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2017/11/15 04:16:27 UTC
[GitHub] piiswrong closed pull request #8657: [WIP][vision]fix
piiswrong closed pull request #8657: [WIP][vision]fix
URL: https://github.com/apache/incubator-mxnet/pull/8657
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):
diff --git a/src/operator/image/image_random-inl.h b/src/operator/image/image_random-inl.h
index 1281a7141a..fa1d6dc529 100644
--- a/src/operator/image/image_random-inl.h
+++ b/src/operator/image/image_random-inl.h
@@ -31,6 +31,7 @@
#include <opencv2/core/mat.hpp>
#include "../mxnet_op.h"
#include "image_common.h"
+#include "operator/operator_common.h"
namespace mxnet {
@@ -41,32 +42,32 @@ enum ImageRandomResource { kRandom };
template<typename xpu>
static void RandomFlip(const nnvm::NodeAttrs &attrs,
- const OpContext &ctx,
- const std::vector<TBlob> &inputs,
- const std::vector<OpReqType> &req,
- const std::vector<TBlob> &outputs) {
+ const OpContext &ctx,
+ const std::vector<TBlob> &inputs,
+ const std::vector<OpReqType> &req,
+ const std::vector<TBlob> &outputs) {
}
template<typename xpu>
static void ToTensor(const nnvm::NodeAttrs &attrs,
- const OpContext &ctx,
- const std::vector<TBlob> &inputs,
- const std::vector<OpReqType> &req,
- const std::vector<TBlob> &outputs) {
+ const OpContext &ctx,
+ const std::vector<TBlob> &inputs,
+ const std::vector<OpReqType> &req,
+ const std::vector<TBlob> &outputs) {
}
template<typename xpu>
static void Normalize(const nnvm::NodeAttrs &attrs,
- const OpContext &ctx,
- const std::vector<TBlob> &inputs,
- const std::vector<OpReqType> &req,
- const std::vector<TBlob> &outputs) {
+ const OpContext &ctx,
+ const std::vector<TBlob> &inputs,
+ const std::vector<OpReqType> &req,
+ const std::vector<TBlob> &outputs) {
}
struct RandomBrightnessParam : public dmlc::Parameter<RandomBrightnessParam> {
float max_brightness;
DMLC_DECLARE_PARAMETER(RandomBrightnessParam) {
DMLC_DECLARE_FIELD(max_brightness)
- .set_default(0.0)
- .describe("Max Brightness.");
+ .set_default(0.0)
+ .describe("Max Brightness.");
}
};
@@ -88,7 +89,7 @@ static void RandomBrightness(const nnvm::NodeAttrs &attrs,
const RandomBrightnessParam ¶m = nnvm::get<RandomBrightnessParam>(attrs.parsed);
float alpha_b = 1.0 + std::uniform_real_distribution<float>(-param.max_brightness, param.max_brightness)(prnd->GetRndEngine());
MSHADOW_TYPE_SWITCH(outputs[0].type_flag_, DType, {
- MXNET_ASSIGN_REQ_SWITCH(req[0], Req,{
+ MXNET_ASSIGN_REQ_SWITCH(req[0], Req, {
mxnet_op::Kernel<mxnet_op::op_with_req<mshadow::op::mul, Req>, xpu>::Launch(
s, inputs[0].Size(), outputs[0].dptr<DType>(), inputs[0].dptr<DType>(), DType(alpha_b));
});
@@ -100,8 +101,8 @@ struct RandomContrastParam : public dmlc::Parameter<RandomContrastParam> {
float max_contrast;
DMLC_DECLARE_PARAMETER(RandomContrastParam) {
DMLC_DECLARE_FIELD(max_contrast)
- .set_default(0.0)
- .describe("Max Contrast.");
+ .set_default(0.0)
+ .describe("Max Contrast.");
}
};
@@ -139,7 +140,7 @@ static void RandomContrast(const nnvm::NodeAttrs &attrs,
static const float coeffs0[] = { R2YF, G2YF, B2YF };
MSHADOW_TYPE_SWITCH(outputs[0].type_flag_, DType, {
- auto input_3d = input.FlatTo3D<xpu, DType>(s);
+ auto input_3d = input.get<xpu, 3, DType>(s);
DType sum = (DType)0.0;
for (int c = 0; c < channel; ++c) {
for (int h = 0; h < hight; ++h) {
@@ -160,36 +161,85 @@ static void RandomContrast(const nnvm::NodeAttrs &attrs,
}
+struct RandomSaturationParam : public dmlc::Parameter<RandomSaturationParam> {
+ float max_saturation;
+ DMLC_DECLARE_PARAMETER(RandomSaturationParam) {
+ DMLC_DECLARE_FIELD(max_saturation)
+ .set_default(0.0)
+ .describe("Max Saturation.");
+ }
+};
+
template<typename xpu>
static void RandomSaturation(const nnvm::NodeAttrs &attrs,
- const OpContext &ctx,
- const std::vector<TBlob> &inputs,
- const std::vector<OpReqType> &req,
- const std::vector<TBlob> &outputs) {
+ const OpContext &ctx,
+ const std::vector<TBlob> &inputs,
+ const std::vector<OpReqType> &req,
+ const std::vector<TBlob> &outputs) {
+ using namespace mshadow;
+ auto input = inputs[0];
+ auto output = outputs[0];
+ int channel = input.shape_[0];
+ int hight = input.shape_[1];
+ int weight = input.shape_[2];
+ Stream<xpu> *s = ctx.get_stream<xpu>();
+ Random<xpu> *prnd = ctx.requested[kRandom].get_random<xpu, real_t>(s);
+ const RandomSaturationParam ¶m = nnvm::get<RandomSaturationParam>(attrs.parsed);
+ float alpha_s = 1.0 + std::uniform_real_distribution<float>(-param.max_saturation, param.max_saturation)(prnd->GetRndEngine());
+ float alpha_o = 1 - alpha_s;
+ const float R2YF = 0.299f;
+ const float G2YF = 0.587f;
+ const float B2YF = 0.114f;
+ static const float coeffs0[] = { R2YF, G2YF, B2YF };
+
+
+ MSHADOW_TYPE_SWITCH(outputs[0].type_flag_, DType, {
+ MXNET_ASSIGN_REQ_SWITCH(req[0], Req, {
+ auto input_3d = input.get<xpu, 3, DType>(s);
+ auto output_3d = output.get<xpu, 3, DType>(s);
+ switch (channel) {
+ case 1:
+ Assign(output_3d, Req, input_3d)
+ break;
+ case 3:
+ for (int h = 0; h < hight; ++h) {
+ for (int w = 0; w < weight; ++w) {
+ float gray = input_3d[0][h][w] * R2YF + input_3d[1][h][w] * G2YF + input_3d[2][h][w] * B2YF;
+ Assign(output_3d[0][h][w], Req, DType(gray * alpha_s + input_3d[0][h][w] * alpha_o))
+ }
+ }
+ break;
+ default:
+ LOG(FATAL) << "not support channel" << channel;
+
+ }
+ });
+ });
+
}
template<typename xpu>
static void RandomHue(const nnvm::NodeAttrs &attrs,
- const OpContext &ctx,
- const std::vector<TBlob> &inputs,
- const std::vector<OpReqType> &req,
- const std::vector<TBlob> &outputs) {
+ const OpContext &ctx,
+ const std::vector<TBlob> &inputs,
+ const std::vector<OpReqType> &req,
+ const std::vector<TBlob> &outputs) {
}
template<typename xpu>
static void RandomColorJitter(const nnvm::NodeAttrs &attrs,
- const OpContext &ctx,
- const std::vector<TBlob> &inputs,
- const std::vector<OpReqType> &req,
- const std::vector<TBlob> &outputs) {
+ const OpContext &ctx,
+ const std::vector<TBlob> &inputs,
+ const std::vector<OpReqType> &req,
+ const std::vector<TBlob> &outputs) {
}
template<typename xpu>
static void RandomLighting(const nnvm::NodeAttrs &attrs,
- const OpContext &ctx,
- const std::vector<TBlob> &inputs,
- const std::vector<OpReqType> &req,
- const std::vector<TBlob> &outputs) {
+ const OpContext &ctx,
+ const std::vector<TBlob> &inputs,
+ const std::vector<OpReqType> &req,
+ const std::vector<TBlob> &outputs) {
}
diff --git a/src/operator/image/image_random.cc b/src/operator/image/image_random.cc
index ade43d8ab2..32648bbfc0 100644
--- a/src/operator/image/image_random.cc
+++ b/src/operator/image/image_random.cc
@@ -49,5 +49,37 @@ NNVM_REGISTER_OP(_image_random_brightness)
.add_argument("data", "NDArray-or-Symbol", "The input.")
.add_arguments(RandomBrightnessParam::__FIELDS__());
+DMLC_REGISTER_PARAMETER(RandomContrastParam);
+NNVM_REGISTER_OP(_image_random_contrast)
+.describe(R"code()code" ADD_FILELINE)
+.set_num_inputs(1)
+.set_num_outputs(1)
+.set_attr_parser(ParamParser<RandomContrastParam>)
+.set_attr<FResourceRequest>("FResourceRequest", [](const NodeAttrs& attrs) {
+ return std::vector<ResourceRequest>{ResourceRequest::kRandom};
+})
+.set_attr<nnvm::FInferShape>("FInferShape", ElemwiseShape<1, 1>)
+.set_attr<nnvm::FInferType>("FInferType", ElemwiseType<1, 1>)
+.set_attr<FCompute>("FCompute<cpu>", RandomContrast<cpu>)
+.set_attr<nnvm::FGradient>("FGradient", ElemwiseGradUseNone{ "_copy" })
+.add_argument("data", "NDArray-or-Symbol", "The input.")
+.add_arguments(RandomContrastParam::__FIELDS__());
+
+DMLC_REGISTER_PARAMETER(RandomSaturationParam);
+NNVM_REGISTER_OP(_image_random_saturation)
+.describe(R"code()code" ADD_FILELINE)
+.set_num_inputs(1)
+.set_num_outputs(1)
+.set_attr_parser(ParamParser<RandomSaturationParam>)
+.set_attr<FResourceRequest>("FResourceRequest", [](const NodeAttrs& attrs) {
+ return std::vector<ResourceRequest>{ResourceRequest::kRandom};
+})
+.set_attr<nnvm::FInferShape>("FInferShape", ElemwiseShape<1, 1>)
+.set_attr<nnvm::FInferType>("FInferType", ElemwiseType<1, 1>)
+.set_attr<FCompute>("FCompute<cpu>", RandomSaturation<cpu>)
+.set_attr<nnvm::FGradient>("FGradient", ElemwiseGradUseNone{ "_copy" })
+.add_argument("data", "NDArray-or-Symbol", "The input.")
+.add_arguments(RandomSaturationParam::__FIELDS__());
+
}
}
diff --git a/src/operator/mxnet_op.h b/src/operator/mxnet_op.h
index e5c3b51410..584ca0f742 100644
--- a/src/operator/mxnet_op.h
+++ b/src/operator/mxnet_op.h
@@ -313,6 +313,12 @@ struct op_with_req {
KERNEL_ASSIGN(out[i], req, OP::Map(in[i], value));
}
+ /*! \brief input is tensor and two scalar value */
+ template<typename DType>
+ MSHADOW_XINLINE static void Map(int i, DType *out, const DType *in, const DType value_1, const DType value_2) {
+ KERNEL_ASSIGN(out[i], req, OP::Map(in[i], value_1, value_2));
+ }
+
/*! \brief No inputs (ie fill to constant value) */
template<typename DType>
MSHADOW_XINLINE static void Map(int i, DType *out) {
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services