ёiSSKJr SSKJr SSKJr SSKJrJr SSK J r SSK J r SSK Jr \(aSS KJr S S /rSSS jjr"SS \5rg )) annotations) TYPE_CHECKING)_C_ops) check_typecheck_variable_and_dtype)in_dynamic_or_pir_mode) LayerHelper)Layer)Tensorviterbi_decodeViterbiDecoderNc[5(a[R"XX#5$[USSS/S5 [USSS/S5 [USSS5 [ US[ S5 [ S 0[5D6nS U0nURUR5nURS5nURSUUUS .XxS .US 9 Xx4$)a Decode the highest scoring sequence of tags computed by transitions and potentials and get the viterbi path. Args: potentials (Tensor): The input tensor of unary emission. This is a 3-D tensor with shape of [batch_size, sequence_length, num_tags]. The data type is float32 or float64. transition_params (Tensor): The input tensor of transition matrix. This is a 2-D tensor with shape of [num_tags, num_tags]. The data type is float32 or float64. lengths (Tensor): The input tensor of length of each sequence. This is a 1-D tensor with shape of [batch_size]. The data type is int64. include_bos_eos_tag (`bool`, optional): If set to True, the last row and the last column of transitions will be considered as start tag, the second to last row and the second to last column of transitions will be considered as stop tag. Defaults to ``True``. name (str|None, optional): The default value is None. Normally there is no need for user to set this property. For more information, please refer to :ref:`api_guide_Name`. Returns: scores(Tensor): The output tensor containing the score for the Viterbi sequence. The shape is [batch_size] and the data type is float32 or float64. paths(Tensor): The output tensor containing the highest scoring tag indices. The shape is [batch_size, sequence_length] and the data type is int64. Examples: .. code-block:: python >>> import paddle >>> paddle.seed(2023) >>> batch_size, seq_len, num_tags = 2, 4, 3 >>> emission = paddle.rand((batch_size, seq_len, num_tags), dtype='float32') >>> length = paddle.randint(1, seq_len + 1, [batch_size]) >>> tags = paddle.randint(0, num_tags, [batch_size, seq_len]) >>> transition = paddle.rand((num_tags, num_tags), dtype='float32') >>> scores, path = paddle.text.viterbi_decode(emission, transition, length, False) >>> print(scores) Tensor(shape=[2], dtype=float32, place=Place(cpu), stop_gradient=True, [2.57385254, 2.04533720]) >>> print(path) Tensor(shape=[2, 2], dtype=int64, place=Place(cpu), stop_gradient=True, [[0, 0], [1, 1]]) inputfloat32float64r transitionslengthint64 include_taginclude_bos_eos_tag)Input TransitionLength)ScoresPath)typeinputsoutputsattrs)r ) r rr rrboolr locals"create_variable_for_type_inferencedtype append_op) potentialstransition_paramslengthsrnamehelperr scorespaths Z/var/www/html/banglarbhumi/venv/lib/python3.13/site-packages/paddle/text/viterbi_decode.pyr r s\$$ 7  Gi35E I  Wh9IJ"M49IJ  6VX 6F "$7 8E  6 6z7G7G HF  4 4W =D  + "0  <cp^\rSrSr%SrS\S'S\S'S\S'S S U4S jjjrSS jrS rU=r $)rna Decode the highest scoring sequence of tags computed by transitions and potentials and get the viterbi path. Args: transitions (`Tensor`): The transition matrix. Its dtype is float32 and has a shape of `[num_tags, num_tags]`. include_bos_eos_tag (`bool`, optional): If set to True, the last row and the last column of transitions will be considered as start tag, the second to last row and the second to last column of transitions will be considered as stop tag. Defaults to ``True``. name (str|None, optional): The default value is None. Normally there is no need for user to set this property. For more information, please refer to :ref:`api_guide_Name`. Shape: potentials (Tensor): The input tensor of unary emission. This is a 3-D tensor with shape of [batch_size, sequence_length, num_tags]. The data type is float32 or float64. lengths (Tensor): The input tensor of length of each sequence. This is a 1-D tensor with shape of [batch_size]. The data type is int64. Returns: scores(Tensor): The output tensor containing the score for the Viterbi sequence. The shape is [batch_size] and the data type is float32 or float64. paths(Tensor): The output tensor containing the highest scoring tag indices. The shape is [batch_size, sequence_length] and the data type is int64. Examples: .. code-block:: python >>> import paddle >>> paddle.seed(2023) >>> batch_size, seq_len, num_tags = 2, 4, 3 >>> emission = paddle.rand((batch_size, seq_len, num_tags), dtype='float32') >>> length = paddle.randint(1, seq_len + 1, [batch_size]) >>> tags = paddle.randint(0, num_tags, [batch_size, seq_len]) >>> transition = paddle.rand((num_tags, num_tags), dtype='float32') >>> decoder = paddle.text.ViterbiDecoder(transition, include_bos_eos_tag=False) >>> scores, path = decoder(emission, length) >>> print(scores) Tensor(shape=[2], dtype=float32, place=Place(cpu), stop_gradient=True, [2.57385254, 2.04533720]) >>> print(path) Tensor(shape=[2, 2], dtype=int64, place=Place(cpu), stop_gradient=True, [[0, 0], [1, 1]]) r rr!r str | Noner)cF>[TU]5 XlX lX0lgN)super__init__rrr))selfrrr) __class__s r-r5ViterbiDecoder.__init__s! &#6  r.c\[UURUURUR5$r3)r rrr))r6r&r(s r-forwardViterbiDecoder.forwards.       $ $ II   r.)rr)rTN)rr rr!r)r1returnNone)r&r r(r r=r ) __name__ __module__ __qualname____firstlineno____doc____annotations__r5r:__static_attributes__ __classcell__)r7s@r-rrns_)V  %)  "       r.r<) r&r r'r r(r rr!r)r1r=r ) __future__rtypingrpaddlerbase.data_feederrrbase.frameworkr base.layer_helperr nnr r __all__r rr.r-rPs# C3+ - .!% LLLL L  L  L^B UB r.