public static enum OnnxMl.Version extends java.lang.Enum<OnnxMl.Version> implements com.google.protobuf.ProtocolMessageEnum
Versioning ONNX versioning is specified in docs/IR.md and elaborated on in docs/Versioning.md To be compatible with both proto2 and proto3, we will use a version number that is not defined by the default value but an explicit enum number.Protobuf enum
onnx.Version| Enum Constant and Description |
|---|
_START_VERSION
proto3 requires the first enum value to be zero.
|
IR_VERSION
IR VERSION 7 published on <TBD>
- Add support to allow function body graph to rely on multiple external opreator sets.
|
IR_VERSION_2017_10_10
The version field is always serialized and we will use it to store the
version that the graph is generated from.
|
IR_VERSION_2017_10_30
IR_VERSION 2 published on Oct 30, 2017
- Added type discriminator to AttributeProto to support proto3 users
|
IR_VERSION_2017_11_3
IR VERSION 3 published on Nov 3, 2017
- For operator versioning:
- Added new message OperatorSetIdProto
- Added opset_import in ModelProto
- For vendor extensions, added domain in NodeProto
|
IR_VERSION_2019_1_22
IR VERSION 4 published on Jan 22, 2019
- Relax constraint that initializers should be a subset of graph inputs
- Add type BFLOAT16
|
IR_VERSION_2019_3_18
IR VERSION 5 published on March 18, 2019
- Add message TensorAnnotation.
|
IR_VERSION_2019_9_19
IR VERSION 6 published on Sep 19, 2019
- Add support for sparse tensor constants stored in model.
|
| Modifier and Type | Field and Description |
|---|---|
static int |
_START_VERSION_VALUE
proto3 requires the first enum value to be zero.
|
static int |
IR_VERSION_2017_10_10_VALUE
The version field is always serialized and we will use it to store the
version that the graph is generated from.
|
static int |
IR_VERSION_2017_10_30_VALUE
IR_VERSION 2 published on Oct 30, 2017
- Added type discriminator to AttributeProto to support proto3 users
|
static int |
IR_VERSION_2017_11_3_VALUE
IR VERSION 3 published on Nov 3, 2017
- For operator versioning:
- Added new message OperatorSetIdProto
- Added opset_import in ModelProto
- For vendor extensions, added domain in NodeProto
|
static int |
IR_VERSION_2019_1_22_VALUE
IR VERSION 4 published on Jan 22, 2019
- Relax constraint that initializers should be a subset of graph inputs
- Add type BFLOAT16
|
static int |
IR_VERSION_2019_3_18_VALUE
IR VERSION 5 published on March 18, 2019
- Add message TensorAnnotation.
|
static int |
IR_VERSION_2019_9_19_VALUE
IR VERSION 6 published on Sep 19, 2019
- Add support for sparse tensor constants stored in model.
|
static int |
IR_VERSION_VALUE
IR VERSION 7 published on <TBD>
- Add support to allow function body graph to rely on multiple external opreator sets.
|
| Modifier and Type | Method and Description |
|---|---|
static OnnxMl.Version |
forNumber(int value) |
static com.google.protobuf.Descriptors.EnumDescriptor |
getDescriptor() |
com.google.protobuf.Descriptors.EnumDescriptor |
getDescriptorForType() |
int |
getNumber() |
com.google.protobuf.Descriptors.EnumValueDescriptor |
getValueDescriptor() |
static com.google.protobuf.Internal.EnumLiteMap<OnnxMl.Version> |
internalGetValueMap() |
static OnnxMl.Version |
valueOf(com.google.protobuf.Descriptors.EnumValueDescriptor desc) |
static OnnxMl.Version |
valueOf(int value)
Deprecated.
Use
forNumber(int) instead. |
static OnnxMl.Version |
valueOf(java.lang.String name)
Returns the enum constant of this type with the specified name.
|
static OnnxMl.Version[] |
values()
Returns an array containing the constants of this enum type, in
the order they are declared.
|
public static final OnnxMl.Version _START_VERSION
proto3 requires the first enum value to be zero. We add this just to appease the compiler.
_START_VERSION = 0;public static final OnnxMl.Version IR_VERSION_2017_10_10
The version field is always serialized and we will use it to store the version that the graph is generated from. This helps us set up version control. For the IR, we are using simple numbers starting with 0x00000001, which was the version we published on Oct 10, 2017.
IR_VERSION_2017_10_10 = 1;public static final OnnxMl.Version IR_VERSION_2017_10_30
IR_VERSION 2 published on Oct 30, 2017 - Added type discriminator to AttributeProto to support proto3 users
IR_VERSION_2017_10_30 = 2;public static final OnnxMl.Version IR_VERSION_2017_11_3
IR VERSION 3 published on Nov 3, 2017
- For operator versioning:
- Added new message OperatorSetIdProto
- Added opset_import in ModelProto
- For vendor extensions, added domain in NodeProto
IR_VERSION_2017_11_3 = 3;public static final OnnxMl.Version IR_VERSION_2019_1_22
IR VERSION 4 published on Jan 22, 2019 - Relax constraint that initializers should be a subset of graph inputs - Add type BFLOAT16
IR_VERSION_2019_1_22 = 4;public static final OnnxMl.Version IR_VERSION_2019_3_18
IR VERSION 5 published on March 18, 2019 - Add message TensorAnnotation. - Add quantization annotation in GraphProto to map tensor with its scale and zero point quantization parameters.
IR_VERSION_2019_3_18 = 5;public static final OnnxMl.Version IR_VERSION_2019_9_19
IR VERSION 6 published on Sep 19, 2019 - Add support for sparse tensor constants stored in model. - Add message SparseTensorProto - Add sparse initializers
IR_VERSION_2019_9_19 = 6;public static final OnnxMl.Version IR_VERSION
IR VERSION 7 published on <TBD> - Add support to allow function body graph to rely on multiple external opreator sets. - Add a list to promote inference graph's initializers to global and mutable variables. Global variables are visible in all graphs of the stored models. - Add message TrainingInfoProto to store initialization method and training algorithm. The execution of TrainingInfoProto can modify the values of mutable variables. - Implicitly add inference graph into each TrainingInfoProto's algorithm.
IR_VERSION = 7;public static final int _START_VERSION_VALUE
proto3 requires the first enum value to be zero. We add this just to appease the compiler.
_START_VERSION = 0;public static final int IR_VERSION_2017_10_10_VALUE
The version field is always serialized and we will use it to store the version that the graph is generated from. This helps us set up version control. For the IR, we are using simple numbers starting with 0x00000001, which was the version we published on Oct 10, 2017.
IR_VERSION_2017_10_10 = 1;public static final int IR_VERSION_2017_10_30_VALUE
IR_VERSION 2 published on Oct 30, 2017 - Added type discriminator to AttributeProto to support proto3 users
IR_VERSION_2017_10_30 = 2;public static final int IR_VERSION_2017_11_3_VALUE
IR VERSION 3 published on Nov 3, 2017
- For operator versioning:
- Added new message OperatorSetIdProto
- Added opset_import in ModelProto
- For vendor extensions, added domain in NodeProto
IR_VERSION_2017_11_3 = 3;public static final int IR_VERSION_2019_1_22_VALUE
IR VERSION 4 published on Jan 22, 2019 - Relax constraint that initializers should be a subset of graph inputs - Add type BFLOAT16
IR_VERSION_2019_1_22 = 4;public static final int IR_VERSION_2019_3_18_VALUE
IR VERSION 5 published on March 18, 2019 - Add message TensorAnnotation. - Add quantization annotation in GraphProto to map tensor with its scale and zero point quantization parameters.
IR_VERSION_2019_3_18 = 5;public static final int IR_VERSION_2019_9_19_VALUE
IR VERSION 6 published on Sep 19, 2019 - Add support for sparse tensor constants stored in model. - Add message SparseTensorProto - Add sparse initializers
IR_VERSION_2019_9_19 = 6;public static final int IR_VERSION_VALUE
IR VERSION 7 published on <TBD> - Add support to allow function body graph to rely on multiple external opreator sets. - Add a list to promote inference graph's initializers to global and mutable variables. Global variables are visible in all graphs of the stored models. - Add message TrainingInfoProto to store initialization method and training algorithm. The execution of TrainingInfoProto can modify the values of mutable variables. - Implicitly add inference graph into each TrainingInfoProto's algorithm.
IR_VERSION = 7;public static OnnxMl.Version[] values()
for (OnnxMl.Version c : OnnxMl.Version.values()) System.out.println(c);
public static OnnxMl.Version valueOf(java.lang.String name)
name - the name of the enum constant to be returned.java.lang.IllegalArgumentException - if this enum type has no constant with the specified namejava.lang.NullPointerException - if the argument is nullpublic final int getNumber()
getNumber in interface com.google.protobuf.Internal.EnumLitegetNumber in interface com.google.protobuf.ProtocolMessageEnum@Deprecated public static OnnxMl.Version valueOf(int value)
forNumber(int) instead.value - The numeric wire value of the corresponding enum entry.public static OnnxMl.Version forNumber(int value)
value - The numeric wire value of the corresponding enum entry.public static com.google.protobuf.Internal.EnumLiteMap<OnnxMl.Version> internalGetValueMap()
public final com.google.protobuf.Descriptors.EnumValueDescriptor getValueDescriptor()
getValueDescriptor in interface com.google.protobuf.ProtocolMessageEnumpublic final com.google.protobuf.Descriptors.EnumDescriptor getDescriptorForType()
getDescriptorForType in interface com.google.protobuf.ProtocolMessageEnumpublic static final com.google.protobuf.Descriptors.EnumDescriptor getDescriptor()
public static OnnxMl.Version valueOf(com.google.protobuf.Descriptors.EnumValueDescriptor desc)