|
|
|
@ -12,7 +12,6 @@ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
See the License for the specific language governing permissions and
|
|
|
|
See the License for the specific language governing permissions and
|
|
|
|
limitations under the License. */
|
|
|
|
limitations under the License. */
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
#pragma once
|
|
|
|
#pragma once
|
|
|
|
|
|
|
|
|
|
|
|
#include <stddef.h>
|
|
|
|
#include <stddef.h>
|
|
|
|
@ -112,7 +111,8 @@ public:
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
* Create A Matrix with height,width, which is filled by zero.
|
|
|
|
* Create A Matrix with height,width, which is filled by zero.
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
static Matrix* createZero(size_t height, size_t width,
|
|
|
|
static Matrix* createZero(size_t height,
|
|
|
|
|
|
|
|
size_t width,
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
@ -124,8 +124,11 @@ public:
|
|
|
|
*
|
|
|
|
*
|
|
|
|
* @note the default sparse type is SPARSE_CSR.
|
|
|
|
* @note the default sparse type is SPARSE_CSR.
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
static Matrix* createSparse(size_t height, size_t width, size_t nnz,
|
|
|
|
static Matrix* createSparse(size_t height,
|
|
|
|
bool isNonVal = true, bool trans = false,
|
|
|
|
size_t width,
|
|
|
|
|
|
|
|
size_t nnz,
|
|
|
|
|
|
|
|
bool isNonVal = true,
|
|
|
|
|
|
|
|
bool trans = false,
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
@ -134,13 +137,17 @@ public:
|
|
|
|
* @param data list of float should be passed in python.
|
|
|
|
* @param data list of float should be passed in python.
|
|
|
|
* @note the value will be copy into a new matrix.
|
|
|
|
* @note the value will be copy into a new matrix.
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
static Matrix* createDense(const std::vector<float>& data, size_t height,
|
|
|
|
static Matrix* createDense(const std::vector<float>& data,
|
|
|
|
size_t width, bool useGpu = isUsingGpu());
|
|
|
|
size_t height,
|
|
|
|
|
|
|
|
size_t width,
|
|
|
|
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
|
|
|
|
|
|
|
|
static Matrix* createDenseFromNumpy(float* data, int dim1, int dim2,
|
|
|
|
static Matrix* createDenseFromNumpy(
|
|
|
|
|
|
|
|
float* data,
|
|
|
|
|
|
|
|
int dim1,
|
|
|
|
|
|
|
|
int dim2,
|
|
|
|
bool copy = true,
|
|
|
|
bool copy = true,
|
|
|
|
bool useGpu = isUsingGpu())
|
|
|
|
bool useGpu = isUsingGpu()) throw(UnsupportError);
|
|
|
|
throw (UnsupportError);
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
* Create Cpu Dense Matrix from numpy matrix, dtype=float32
|
|
|
|
* Create Cpu Dense Matrix from numpy matrix, dtype=float32
|
|
|
|
@ -151,7 +158,9 @@ public:
|
|
|
|
* @param copy true if copy into a new matrix, false will create
|
|
|
|
* @param copy true if copy into a new matrix, false will create
|
|
|
|
* matrix inplace.
|
|
|
|
* matrix inplace.
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
static Matrix* createCpuDenseFromNumpy(float* data, int dim1, int dim2,
|
|
|
|
static Matrix* createCpuDenseFromNumpy(float* data,
|
|
|
|
|
|
|
|
int dim1,
|
|
|
|
|
|
|
|
int dim2,
|
|
|
|
bool copy = false);
|
|
|
|
bool copy = false);
|
|
|
|
|
|
|
|
|
|
|
|
/// Create Gpu Dense Matrix from numpy matrix, dtype=float32
|
|
|
|
/// Create Gpu Dense Matrix from numpy matrix, dtype=float32
|
|
|
|
@ -171,11 +180,13 @@ public:
|
|
|
|
* numpy_mat = m.toNumpyMat()
|
|
|
|
* numpy_mat = m.toNumpyMat()
|
|
|
|
* @endcode
|
|
|
|
* @endcode
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
void toNumpyMatInplace(float** view_data, int* dim1,
|
|
|
|
void toNumpyMatInplace(float** view_data,
|
|
|
|
|
|
|
|
int* dim1,
|
|
|
|
int* dim2) throw(UnsupportError);
|
|
|
|
int* dim2) throw(UnsupportError);
|
|
|
|
|
|
|
|
|
|
|
|
/// Copy To numpy mat.
|
|
|
|
/// Copy To numpy mat.
|
|
|
|
void copyToNumpyMat(float** view_m_data, int* dim1,
|
|
|
|
void copyToNumpyMat(float** view_m_data,
|
|
|
|
|
|
|
|
int* dim1,
|
|
|
|
int* dim2) throw(UnsupportError);
|
|
|
|
int* dim2) throw(UnsupportError);
|
|
|
|
|
|
|
|
|
|
|
|
/// Copy From Numpy Mat
|
|
|
|
/// Copy From Numpy Mat
|
|
|
|
@ -248,15 +259,18 @@ public:
|
|
|
|
static Vector* create(const std::vector<float>& data,
|
|
|
|
static Vector* create(const std::vector<float>& data,
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
|
|
|
|
|
|
|
|
static Vector* createVectorFromNumpy(float* data, int dim, bool copy = true,
|
|
|
|
static Vector* createVectorFromNumpy(
|
|
|
|
bool useGpu = isUsingGpu())
|
|
|
|
float* data,
|
|
|
|
throw (UnsupportError);
|
|
|
|
int dim,
|
|
|
|
|
|
|
|
bool copy = true,
|
|
|
|
|
|
|
|
bool useGpu = isUsingGpu()) throw(UnsupportError);
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
* Create Cpu Vector from numpy array, which dtype=float32
|
|
|
|
* Create Cpu Vector from numpy array, which dtype=float32
|
|
|
|
*
|
|
|
|
*
|
|
|
|
* If copy is false, it will create vector inplace.
|
|
|
|
* If copy is false, it will create vector inplace.
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
static Vector* createCpuVectorFromNumpy(float* data, int dim,
|
|
|
|
static Vector* createCpuVectorFromNumpy(float* data,
|
|
|
|
|
|
|
|
int dim,
|
|
|
|
bool copy = false);
|
|
|
|
bool copy = false);
|
|
|
|
|
|
|
|
|
|
|
|
/// Create Gpu Vector from numpy array, which dtype=float32
|
|
|
|
/// Create Gpu Vector from numpy array, which dtype=float32
|
|
|
|
@ -312,16 +326,19 @@ public:
|
|
|
|
static IVector* create(const std::vector<int>& data,
|
|
|
|
static IVector* create(const std::vector<int>& data,
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
bool useGpu = isUsingGpu());
|
|
|
|
|
|
|
|
|
|
|
|
static IVector* createVectorFromNumpy(int* data, int dim, bool copy = true,
|
|
|
|
static IVector* createVectorFromNumpy(
|
|
|
|
bool useGpu = isUsingGpu())
|
|
|
|
int* data,
|
|
|
|
throw (UnsupportError);
|
|
|
|
int dim,
|
|
|
|
|
|
|
|
bool copy = true,
|
|
|
|
|
|
|
|
bool useGpu = isUsingGpu()) throw(UnsupportError);
|
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
* Create Cpu IVector from numpy array, which dtype=int32
|
|
|
|
* Create Cpu IVector from numpy array, which dtype=int32
|
|
|
|
*
|
|
|
|
*
|
|
|
|
* If copy is false, it will create vector inplace
|
|
|
|
* If copy is false, it will create vector inplace
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
static IVector* createCpuVectorFromNumpy(int* data, int dim,
|
|
|
|
static IVector* createCpuVectorFromNumpy(int* data,
|
|
|
|
|
|
|
|
int dim,
|
|
|
|
bool copy = false);
|
|
|
|
bool copy = false);
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
* Create Gpu IVector from numpy array, which dtype=int32
|
|
|
|
* Create Gpu IVector from numpy array, which dtype=int32
|
|
|
|
@ -605,7 +622,8 @@ class ParameterTraverseCallback {
|
|
|
|
public:
|
|
|
|
public:
|
|
|
|
~ParameterTraverseCallback();
|
|
|
|
~ParameterTraverseCallback();
|
|
|
|
|
|
|
|
|
|
|
|
void apply(const std::vector<Vector*>& vecs, const ParameterConfig& config,
|
|
|
|
void apply(const std::vector<Vector*>& vecs,
|
|
|
|
|
|
|
|
const ParameterConfig& config,
|
|
|
|
size_t sparseId);
|
|
|
|
size_t sparseId);
|
|
|
|
|
|
|
|
|
|
|
|
private:
|
|
|
|
private:
|
|
|
|
@ -638,7 +656,8 @@ public:
|
|
|
|
|
|
|
|
|
|
|
|
void finishBatch();
|
|
|
|
void finishBatch();
|
|
|
|
|
|
|
|
|
|
|
|
void update(const std::vector<Vector*>& vecs, const ParameterConfig& conf,
|
|
|
|
void update(const std::vector<Vector*>& vecs,
|
|
|
|
|
|
|
|
const ParameterConfig& conf,
|
|
|
|
size_t sparseId = NO_SPARSE_ID);
|
|
|
|
size_t sparseId = NO_SPARSE_ID);
|
|
|
|
|
|
|
|
|
|
|
|
std::vector<int> getParameterTypes() const;
|
|
|
|
std::vector<int> getParameterTypes() const;
|
|
|
|
@ -678,7 +697,8 @@ public:
|
|
|
|
* model config by TrainerConfig
|
|
|
|
* model config by TrainerConfig
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
static GradientMachine* createByModelConfig(
|
|
|
|
static GradientMachine* createByModelConfig(
|
|
|
|
ModelConfig* conf, GradientMatchineCreateMode mode = CREATE_MODE_NORMAL,
|
|
|
|
ModelConfig* conf,
|
|
|
|
|
|
|
|
GradientMatchineCreateMode mode = CREATE_MODE_NORMAL,
|
|
|
|
const std::vector<int>& parameterTypes = defaultParamTypes);
|
|
|
|
const std::vector<int>& parameterTypes = defaultParamTypes);
|
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
@ -701,7 +721,8 @@ public:
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
* Combine forward/backward
|
|
|
|
* Combine forward/backward
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
void forwardBackward(const Arguments& inArgs, Arguments* outArgs,
|
|
|
|
void forwardBackward(const Arguments& inArgs,
|
|
|
|
|
|
|
|
Arguments* outArgs,
|
|
|
|
PassType passType,
|
|
|
|
PassType passType,
|
|
|
|
const UpdateCallback& callback = UpdateCallback());
|
|
|
|
const UpdateCallback& callback = UpdateCallback());
|
|
|
|
|
|
|
|
|
|
|
|
@ -722,14 +743,17 @@ public:
|
|
|
|
*/
|
|
|
|
*/
|
|
|
|
SequenceGenerator* asSequenceGenerator(
|
|
|
|
SequenceGenerator* asSequenceGenerator(
|
|
|
|
const std::vector<std::string>& dict = std::vector<std::string>(),
|
|
|
|
const std::vector<std::string>& dict = std::vector<std::string>(),
|
|
|
|
size_t begin_id = 0UL, size_t end_id = 0UL, size_t max_length = 100UL,
|
|
|
|
size_t begin_id = 0UL,
|
|
|
|
|
|
|
|
size_t end_id = 0UL,
|
|
|
|
|
|
|
|
size_t max_length = 100UL,
|
|
|
|
size_t beam_size = -1UL);
|
|
|
|
size_t beam_size = -1UL);
|
|
|
|
|
|
|
|
|
|
|
|
private:
|
|
|
|
private:
|
|
|
|
GradientMachinePrivate* m;
|
|
|
|
GradientMachinePrivate* m;
|
|
|
|
|
|
|
|
|
|
|
|
static GradientMachine* createFromPaddleModelPtr(
|
|
|
|
static GradientMachine* createFromPaddleModelPtr(
|
|
|
|
const void* confPtr, GradientMatchineCreateMode mode,
|
|
|
|
const void* confPtr,
|
|
|
|
|
|
|
|
GradientMatchineCreateMode mode,
|
|
|
|
const std::vector<int>& types);
|
|
|
|
const std::vector<int>& types);
|
|
|
|
|
|
|
|
|
|
|
|
// Not to use c++ 11 init-list, so we use static var as function default arg.
|
|
|
|
// Not to use c++ 11 init-list, so we use static var as function default arg.
|
|
|
|
@ -751,8 +775,8 @@ public:
|
|
|
|
/// Create A Trainer By TrainerConfig. using paddle command line.
|
|
|
|
/// Create A Trainer By TrainerConfig. using paddle command line.
|
|
|
|
static Trainer* createByCommandLine() throw(IOError);
|
|
|
|
static Trainer* createByCommandLine() throw(IOError);
|
|
|
|
|
|
|
|
|
|
|
|
static Trainer* create(TrainerConfig* optConfig, GradientMachine* gm)
|
|
|
|
static Trainer* create(TrainerConfig* optConfig,
|
|
|
|
throw(IOError);
|
|
|
|
GradientMachine* gm) throw(IOError);
|
|
|
|
|
|
|
|
|
|
|
|
/// Start training
|
|
|
|
/// Start training
|
|
|
|
void startTrain();
|
|
|
|
void startTrain();
|
|
|
|
|
dangqingqing
9 years ago