TensorDebugMode

enum akhir publik TensorDebugMode

 Available modes for extracting debugging information from a Tensor.
 TODO(cais): Document the detailed column names and semantics in a separate
 markdown file once the implementation settles.
Protobuf enum tensorflow.TensorDebugMode

Konstanta

ke dalam CONCISE_HEALTH_VALUE
 A concise health summary for float-type tensors.
ke dalam CURT_HEALTH_VALUE
 A minimalist health summary for float-type tensors.
ke dalam FULL_HEALTH_VALUE
 A detailed health summary.
ke dalam FULL_NUMERICS_VALUE
 Full numeric summary.
ke dalam FULL_TENSOR_VALUE
 Full tensor value.
ke dalam NO_TENSOR_VALUE
 Only records what tensors are computed, eagerly or in graphs.
ke dalam REDUCE_INF_NAN_THREE_SLOTS_VALUE
 Reduce the elements of a tensor to a rank-1 tensor of shape [3], in which
 - the 1st element is -inf if any element of the tensor is -inf,
   or zero otherwise.
ke dalam BENTUK_VALUE
 Provides full runtime shape information, up to a maximum rank, beyond
 which the dimension sizes are truncated.
ke dalam UNSPECIFIED_VALUE UNSPECIFIED = 0;

Metode Warisan

Nilai Enum

TensorDebugMode final statis publik CONCISE_HEALTH

 A concise health summary for float-type tensors.
 Contains more information that CURT_HEALTH.
 Infinity and NaN are treated differently.
 Applicable only to float and integer dtypes.
CONCISE_HEALTH = 3;

TensorDebugMode final statis publik CURT_HEALTH

 A minimalist health summary for float-type tensors.
 Contains information only about the presence/absence of pathological
 values including Infinity and NaN.
 Applicable only to float dtypes.
CURT_HEALTH = 2;

TensorDebugMode final statis publik FULL_HEALTH

 A detailed health summary.
 Contains further detailed information than `CONCISE_HEALTH`.
 Information about device, dtype and shape are included.
 Counts for various types of values (Infinity, NaN, negative, zero,
 positive) are included.
 Applicable to float, integer and boolean dtypes.
FULL_HEALTH = 4;

TensorDebugMode final statis publik FULL_NUMERICS

 Full numeric summary.
 Including device, dtype, shape, counts of various types of values
 (Infinity, NaN, negative, zero, positive), and summary statistics
 (minimum, maximum, mean and variance).
 Applicable to float, integer and boolean dtypes.
FULL_NUMERICS = 6;

TensorDebugMode final statis publik FULL_TENSOR

 Full tensor value.
FULL_TENSOR = 7;

TensorDebugMode NO_TENSOR final statis publik

 Only records what tensors are computed, eagerly or in graphs.
 No information regarding the value of the tensor is available.
NO_TENSOR = 1;

TensorDebugMode final statis publik REDUCE_INF_NAN_THREE_SLOTS

 Reduce the elements of a tensor to a rank-1 tensor of shape [3], in which
 - the 1st element is -inf if any element of the tensor is -inf,
   or zero otherwise.
 - the 2nd element is +inf if any element of the tensor is +inf,
   or zero otherwise.
 - the 3rd element is nan if any element of the tensor is nan, or zero
   otherwise.
REDUCE_INF_NAN_THREE_SLOTS = 8;

BENTUK TensorDebugMode akhir statis publik

 Provides full runtime shape information, up to a maximum rank, beyond
 which the dimension sizes are truncated.
SHAPE = 5;

TensorDebugMode final statis publik TIDAK DIAKUI

TensorDebugMode final statis publik TIDAK DITENTUKAN

UNSPECIFIED = 0;

Konstanta

int akhir statis publik CONCISE_HEALTH_VALUE

 A concise health summary for float-type tensors.
 Contains more information that CURT_HEALTH.
 Infinity and NaN are treated differently.
 Applicable only to float and integer dtypes.
CONCISE_HEALTH = 3;

Nilai Konstan: 3

int akhir statis publik CURT_HEALTH_VALUE

 A minimalist health summary for float-type tensors.
 Contains information only about the presence/absence of pathological
 values including Infinity and NaN.
 Applicable only to float dtypes.
CURT_HEALTH = 2;

Nilai Konstan: 2

int akhir statis publik FULL_HEALTH_VALUE

 A detailed health summary.
 Contains further detailed information than `CONCISE_HEALTH`.
 Information about device, dtype and shape are included.
 Counts for various types of values (Infinity, NaN, negative, zero,
 positive) are included.
 Applicable to float, integer and boolean dtypes.
FULL_HEALTH = 4;

Nilai Konstan: 4

int akhir statis publik FULL_NUMERICS_VALUE

 Full numeric summary.
 Including device, dtype, shape, counts of various types of values
 (Infinity, NaN, negative, zero, positive), and summary statistics
 (minimum, maximum, mean and variance).
 Applicable to float, integer and boolean dtypes.
FULL_NUMERICS = 6;

Nilai Konstan: 6

int akhir statis publik FULL_TENSOR_VALUE

 Full tensor value.
FULL_TENSOR = 7;

Nilai Konstan: 7

int akhir statis publik NO_TENSOR_VALUE

 Only records what tensors are computed, eagerly or in graphs.
 No information regarding the value of the tensor is available.
NO_TENSOR = 1;

Nilai Konstan: 1

int akhir statis publik REDUCE_INF_NAN_THREE_SLOTS_VALUE

 Reduce the elements of a tensor to a rank-1 tensor of shape [3], in which
 - the 1st element is -inf if any element of the tensor is -inf,
   or zero otherwise.
 - the 2nd element is +inf if any element of the tensor is +inf,
   or zero otherwise.
 - the 3rd element is nan if any element of the tensor is nan, or zero
   otherwise.
REDUCE_INF_NAN_THREE_SLOTS = 8;

Nilai Konstan: 8

int akhir statis publik SHAPE_VALUE

 Provides full runtime shape information, up to a maximum rank, beyond
 which the dimension sizes are truncated.
SHAPE = 5;

Nilai Konstan: 5

int akhir statis publik UNSPECIFIED_VALUE

UNSPECIFIED = 0;

Nilai Konstan: 0