Enums¶
Details¶
- class GstAnalytics.ModelInfoTensorDirection(value)¶
Bases:
GLib.EnumNew in version 1.28.
- UNKNOWN = 0¶
Tensor location is unknown
- INPUT = 1¶
Input tensor
- OUTPUT = 2¶
Output tensor
- class GstAnalytics.SegmentationType(value)¶
Bases:
GLib.EnumEnum value describing supported segmentation type
New in version 1.26.
- SEMANTIC = 0¶
Segmentation where the belonging of each pixel to a class of objects is identified.
- INSTANCE = 1¶
Segmentation where the belonging of each pixel to instance of an object is identified.
- class GstAnalytics.TensorDataType(value)¶
Bases:
GLib.EnumDescribe the type of data contain in the tensor.
New in version 1.26.
- INT4 = 0¶
signed 4 bit integer tensor data
- INT8 = 1¶
signed 8 bit integer tensor data
- FLOAT16 = 10¶
16 bit floating point tensor data
- FLOAT32 = 11¶
32 bit floating point tensor data
- FLOAT64 = 12¶
64 bit floating point tensor data
- BFLOAT16 = 13¶
“brain” 16 bit floating point tensor data
- STRING = 14¶
UTF-8 string
New in version 1.28.
- BOOL = 15¶
A boolean value stored in 1 byte.
New in version 1.28.
- COMPLEX64 = 16¶
A 64-bit complex number stored in 2 32-bit values.
New in version 1.28.
- COMPLEX128 = 17¶
A 128-bit complex number stored in 2 64-bit values.
New in version 1.28.
- FLOAT8E4M3FN = 18¶
A non-IEEE 8-bit floating point format with 4 exponent bits and 3 mantissa bits, with NaN and no infinite values (FN). See this paper for more details
New in version 1.28.
- FLOAT8E4M3FNUZ = 19¶
A non-IEEE 8-bit floating point format with 4 exponent bits and 3 mantissa bits, with NaN, no infinite values (FN) and no negative zero (UZ). See this paper for more details
New in version 1.28.
- INT16 = 2¶
signed 16 bit integer tensor data
- FLOAT8E5M2 = 20¶
A non-IEEE 8-bit floating point format with 5 exponent bits and 2 mantissa bits. See this paper for more details
New in version 1.28.
- FLOAT8E5M2FNUZ = 21¶
A non-IEEE 8-bit floating point format with 5 exponent bits and 2 mantissa bits, with NaN, no infinite values (FN) and no negative zero (UZ). See this paper for more details
New in version 1.28.
- INT32 = 3¶
signed 32 bit integer tensor data
- INT64 = 4¶
signed 64 bit integer tensor data
- UINT4 = 5¶
unsigned 4 bit integer tensor data
- UINT8 = 6¶
unsigned 8 bit integer tensor data
- UINT16 = 7¶
unsigned 16 bit integer tensor data
- UINT32 = 8¶
unsigned 32 bit integer tensor data
- UINT64 = 9¶
unsigned 64 bit integer tensor data