Intel® Machine Learning Scaling Library
2018
A library providing an efficient implementation of communication patterns used in deep learning.
|
Public external interface to Intel® Machine Learning Scaling Library (Intel® MLSL) More...
#include <cstddef>
Go to the source code of this file.
Classes | |
struct | MLSL::QuantParams |
A struct to hold quantization parameters. More... | |
class | MLSL::CommBlockInfo |
A class to hold block information for activations packing/unpacking. More... | |
class | MLSL::Activation |
A wrapper class for operation input and output activations. More... | |
class | MLSL::ParameterSet |
A wrapper class for operation parameters. More... | |
class | MLSL::Distribution |
A class to hold information about the parallelism scheme being used. More... | |
class | MLSL::OperationRegInfo |
A class to hold Operation registration information. More... | |
class | MLSL::Operation |
A class to hold information about learnable parameters (parameter sets) and activations corresponding to a certain operation of the computational graph. More... | |
class | MLSL::Statistics |
A class to measure and store performance statistics of communication among processes that perform computation in the computational graph. More... | |
class | MLSL::Session |
A class to represent a collection of Operation objects with the same global mini-batch size. More... | |
class | MLSL::Environment |
A singleton object that holds global Intel MLSL functions. More... | |
Namespaces | |
MLSL | |
The namespace containing all Intel MLSL classes and types. | |
Macros | |
#define | MLSL_MAJOR_VERSION 1 |
#define | MLSL_MINOR_VERSION 0 |
#define | MLSL_VERSION(major, minor) ((major << 16) | (minor)) |
#define | MLSL_MAJOR(version) (version >> 16) |
#define | MLSL_MINOR(version) (version & 0xFFFF) |
#define | MLSL_VERSION_GE(v1, v2) |
#define | MLSL_VERSION_LT(v1, v2) |
#define | NO_EXPLICIT_CREATION(ClassName) |
Typedefs | |
typedef int | MLSL::CommReq |
Enumerations | |
enum | MLSL::DataType { MLSL::DT_FLOAT = 0, MLSL::DT_DOUBLE = 1, MLSL::DT_BYTE = 2 } |
enum | MLSL::PhaseType { MLSL::PT_TRAIN = 0, MLSL::PT_TEST = 1 } |
enum | MLSL::GroupType { MLSL::GT_DATA = 0, MLSL::GT_MODEL = 1, MLSL::GT_GLOBAL = 2 } |
Groups of processes supported by Intel MLSL. More... | |
enum | MLSL::ReductionType { MLSL::RT_SUM = 0, MLSL::RT_MIN = 1, MLSL::RT_MAX = 2 } |
enum | MLSL::OpType { MLSL::OT_CC = 0, MLSL::OT_BIAS = 1, MLSL::OT_ACT = 2, MLSL::OT_POOL = 3, MLSL::OT_SPLIT = 4, MLSL::OT_CONCAT = 5, MLSL::OT_BCAST = 6, MLSL::OT_REDUCE = 7, MLSL::OT_DATA = 8, MLSL::OT_EVAL = 9 } |
Compute operation types. More... | |
enum | MLSL::CompressionType { MLSL::CT_NONE = 0, MLSL::CT_QUANTIZATION = 1 } |
Public external interface to Intel® Machine Learning Scaling Library (Intel® MLSL)
#define MLSL_MAJOR | ( | version | ) | (version >> 16) |
A macro to retrieve the major API version from the full API version.
version | full API version |
#define MLSL_MAJOR_VERSION 1 |
Intel MLSL API major version.
#define MLSL_MINOR | ( | version | ) | (version & 0xFFFF) |
A macro to retrieve the minor version from the full version.
version | full API version |
#define MLSL_MINOR_VERSION 0 |
Intel MLSL API minor version.
#define MLSL_VERSION | ( | major, | |
minor | |||
) | ((major << 16) | (minor)) |
A macro to create a full API version from the major and minor API versions.
major | major API version |
minor | minor API version |
#define MLSL_VERSION_GE | ( | v1, | |
v2 | |||
) |
A macro to check the condition (v1 >= v2).
v1 | the first full API version to compare |
v2 | the second full API version to compare |
#define MLSL_VERSION_LT | ( | v1, | |
v2 | |||
) |
A macro to check the condition (v1 < v2).
v1 | the first full API version to compare |
v2 | the second full API version to compare |
#define NO_EXPLICIT_CREATION | ( | ClassName | ) |
A macro to prevent explicit creation of a class instance by user.