experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Add quantized_dimension field to quantization params.

Browse Source 
PiperOrigin-RevId: 229109989
This commit is contained in:
Suharsh Sivakumar 2019-01-13 18:02:02 -08:00 committed by TensorFlower Gardener
parent 898738bce4
commit ac4d368293
2 changed files with 36 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
tensorflow/lite/schema/schema.fbs
View File

@ -64,9 +64,20 @@ table QuantizationParameters {
scale:[float]; // For dequantizing the tensor's values.
zero_point:[long];
// If this is not none, the quantization parameters above are ignored and the
// value of the QuantizationDetails union below should be used.
// If this is not none, the other quantization parameters (i.e. min, max,
// scale, zero_point fields above) are ignored and the value of the
// QuantizationDetails union should be used.
details:QuantizationDetails;
// Specifies the dimension of the Tensor's shape that the scales and
// zero_points correspond to. For example, a tensor t, with dims=[4, 3, 2, 1]
// with quantization params:
// scale=[1.0, 2.0, 3.0], zero_point=[1, 2, 3], quantization_dimension=1
// will be quantized across the second dimension of t.
// t[:, 0, :, :] will have scale[0]=1.0, zero_point[0]=1
// t[:, 1, :, :] will have scale[1]=2.0, zero_point[0]=2
// t[:, 2, :, :] will have scale[2]=3.0, zero_point[0]=3
quantized_dimension:int;
}
table Tensor {

29
tensorflow/lite/schema/schema_generated.h
View File

@ -2335,7 +2335,9 @@ struct QuantizationParametersT : public flatbuffers::NativeTable {
std::vector<float> scale;
std::vector<int64_t> zero_point;
QuantizationDetailsUnion details;
QuantizationParametersT() {
int32_t quantized_dimension;
QuantizationParametersT()
: quantized_dimension(0) {
}
};
@ -2347,7 +2349,8 @@ struct QuantizationParameters FLATBUFFERS_FINAL_CLASS : private flatbuffers::Tab
VT_SCALE = 8,
VT_ZERO_POINT = 10,
VT_DETAILS_TYPE = 12,
VT_DETAILS = 14
VT_DETAILS = 14,
VT_QUANTIZED_DIMENSION = 16
};
const flatbuffers::Vector<float> *min() const {
return GetPointer<const flatbuffers::Vector<float> *>(VT_MIN);
@ -2371,6 +2374,9 @@ struct QuantizationParameters FLATBUFFERS_FINAL_CLASS : private flatbuffers::Tab
const CustomQuantization *details_as_CustomQuantization() const {
return details_type() == QuantizationDetails_CustomQuantization ? static_cast<const CustomQuantization *>(details()) : nullptr;
}
int32_t quantized_dimension() const {
return GetField<int32_t>(VT_QUANTIZED_DIMENSION, 0);
}
bool Verify(flatbuffers::Verifier &verifier) const {
return VerifyTableStart(verifier) &&
VerifyOffset(verifier, VT_MIN) &&
@ -2384,6 +2390,7 @@ struct QuantizationParameters FLATBUFFERS_FINAL_CLASS : private flatbuffers::Tab
VerifyField<uint8_t>(verifier, VT_DETAILS_TYPE) &&
VerifyOffset(verifier, VT_DETAILS) &&
VerifyQuantizationDetails(verifier, details(), details_type()) &&
VerifyField<int32_t>(verifier, VT_QUANTIZED_DIMENSION) &&
verifier.EndTable();
}
QuantizationParametersT *UnPack(const flatbuffers::resolver_function_t *_resolver = nullptr) const;
@ -2416,6 +2423,9 @@ struct QuantizationParametersBuilder {
void add_details(flatbuffers::Offset<void> details) {
fbb_.AddOffset(QuantizationParameters::VT_DETAILS, details);
}
void add_quantized_dimension(int32_t quantized_dimension) {
fbb_.AddElement<int32_t>(QuantizationParameters::VT_QUANTIZED_DIMENSION, quantized_dimension, 0);
}
explicit QuantizationParametersBuilder(flatbuffers::FlatBufferBuilder &_fbb)
: fbb_(_fbb) {
start_ = fbb_.StartTable();
@ -2435,8 +2445,10 @@ inline flatbuffers::Offset<QuantizationParameters> CreateQuantizationParameters(
flatbuffers::Offset<flatbuffers::Vector<float>> scale = 0,
flatbuffers::Offset<flatbuffers::Vector<int64_t>> zero_point = 0,
QuantizationDetails details_type = QuantizationDetails_NONE,
flatbuffers::Offset<void> details = 0) {
flatbuffers::Offset<void> details = 0,
int32_t quantized_dimension = 0) {
QuantizationParametersBuilder builder_(_fbb);
builder_.add_quantized_dimension(quantized_dimension);
builder_.add_details(details);
builder_.add_zero_point(zero_point);
builder_.add_scale(scale);
@ -2453,7 +2465,8 @@ inline flatbuffers::Offset<QuantizationParameters> CreateQuantizationParametersD
const std::vector<float> *scale = nullptr,
const std::vector<int64_t> *zero_point = nullptr,
QuantizationDetails details_type = QuantizationDetails_NONE,
flatbuffers::Offset<void> details = 0) {
flatbuffers::Offset<void> details = 0,
int32_t quantized_dimension = 0) {
return tflite::CreateQuantizationParameters(
_fbb,
min ? _fbb.CreateVector<float>(*min) : 0,
@ -2461,7 +2474,8 @@ inline flatbuffers::Offset<QuantizationParameters> CreateQuantizationParametersD
scale ? _fbb.CreateVector<float>(*scale) : 0,
zero_point ? _fbb.CreateVector<int64_t>(*zero_point) : 0,
details_type,
details);
details,
quantized_dimension);
}
flatbuffers::Offset<QuantizationParameters> CreateQuantizationParameters(flatbuffers::FlatBufferBuilder &_fbb, const QuantizationParametersT *_o, const flatbuffers::rehasher_function_t *_rehasher = nullptr);
@ -8263,6 +8277,7 @@ inline void QuantizationParameters::UnPackTo(QuantizationParametersT *_o, const
{ auto _e = zero_point(); if (_e) { _o->zero_point.resize(_e->size()); for (flatbuffers::uoffset_t _i = 0; _i < _e->size(); _i++) { _o->zero_point[_i] = _e->Get(_i); } } };
{ auto _e = details_type(); _o->details.type = _e; };
{ auto _e = details(); if (_e) _o->details.value = QuantizationDetailsUnion::UnPack(_e, details_type(), _resolver); };
{ auto _e = quantized_dimension(); _o->quantized_dimension = _e; };
}
inline flatbuffers::Offset<QuantizationParameters> QuantizationParameters::Pack(flatbuffers::FlatBufferBuilder &_fbb, const QuantizationParametersT* _o, const flatbuffers::rehasher_function_t *_rehasher) {
@ -8279,6 +8294,7 @@ inline flatbuffers::Offset<QuantizationParameters> CreateQuantizationParameters(
auto _zero_point = _o->zero_point.size() ? _fbb.CreateVector(_o->zero_point) : 0;
auto _details_type = _o->details.type;
auto _details = _o->details.Pack(_fbb);
auto _quantized_dimension = _o->quantized_dimension;
return tflite::CreateQuantizationParameters(
_fbb,
_min,
@ -8286,7 +8302,8 @@ inline flatbuffers::Offset<QuantizationParameters> CreateQuantizationParameters(
_scale,
_zero_point,
_details_type,
_details);
_details,
_quantized_dimension);
}
inline TensorT *Tensor::UnPack(const flatbuffers::resolver_function_t *_resolver) const {