AVhg%T dZddlZddlmZddlmZddlmZddlm Z ddl m Z ddlmZdd l mZdd l mZdd l mZdd lmZdd lmZddlmZddlmZm Z m!Z!ddl"m#Z#ejHejJedgedd8de#e!ejLfde#e!ejNffdZ(edejRe(Z*e(jVjXZ-de#e!ejLfde#e!ejNffdZ.ej^dgdZ0ejHejJedgedd9de#e e!ejNfde#e e!ejNfde#e e!ejbfde#e e!ejbfde#e!ejbfd e#e!ejbfd!e#e e!ejNfd"e#e e!ejbfd#e#e e!ejbfd$e#e!ejbfd%e2d&e3d'e3d(e4d)e4d*e5f d+Z6ed,ejRe6Z7e6jVjXZ8de#e e!ejNfde#e e!ejNfde#e e!ejbfde#e e!ejbfde#e!ejbfd e#e!ejbfd!e#e e!ejNfd"e#e e!ejbfd#e#e e!ejbfd$e#e!ejbfd%e2d&e3d'e3d(e4d)e4d*e5f d-Z9ej^d.gdZ:d9de#e e!ejNfde#e e!ejNfde#e e!ejbfde#e e!ejbfde#e!ejbfd e#e!ejbfd!e#e e!ejNfd"e#e e!ejbfd#e#e e!ejbfd$e#e!ejbfd%e2d&e3d'e3d(e4d)e4d/e5f d0Z;ed1ejRe;Zed6ejRe>Z?e>jVjXZ@d4e#e e!ejbfd&e3d'e3fd7ZAy):zUPython wrappers around TensorFlow ops. This file is MACHINE GENERATED! Do not edit. N) pywrap_tfe)context)core)execute)dtypes)annotation_types)op_def_registry)ops)op_def_library)deprecated_endpoints)dispatch) tf_export)TypeVarListAny) Annotatedztrain.sdca_fprint)v1inputreturnc tjxstj}|j}|jr t j |d||}|St||fd}|tur|S t/j0d||\}}}}|dd}t3j4r&d} |j6} t3j8d| | ||\}|S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t||fd}|tur|St|||S#t j$rYtt f$rHt#j$t&dt)||}|t"j*j,ur|cYSwxYw#tt f$rHt#j$t&dt)||}|t"j*j,ur|cYSwxYw)zComputes fingerprints of the input strings. Args: input: A `Tensor` of type `string`. vector of strings to compute fingerprints on. name: A name for the operation (optional). Returns: A `Tensor` of type `int64`. SdcaFprintN)namectx)rr)_contextr_thread_local_datais_eagerrTFE_Py_FastPathExecute_core_NotOkStatusException_opsraise_from_not_ok_status_FallbackException_dispatcher_for_sdca_fprintNotImplementedsdca_fprint_eager_fallback_SymbolicException TypeError ValueError _dispatchr sdca_fprintdict OpDispatcher NOT_SUPPORTED_op_def_library_apply_op_helper_executemust_record_gradientinputsrecord_gradient) rr_ctxtld_resulte__op_outputs_attrs _inputs_flats R/home/dcms/DCMS/lib/python3.12/site-packages/tensorflow/python/ops/gen_sdca_ops.pyr+r+s     0h..0$ #\\ 11 lD%)g n,* Gn$ n  )::E.Aq#x QK' ""$ F::L lFG5 (' .W  & &- ##At,,  # #    + $.$ g  & ' d&&  # #  z " "" TD9 g  ..<< <  Z    r4e$7 Gi,,::: n  sPC3F4D(DDD E8 EF1AF1/F14AH  H zraw_ops.SdcaFprintctj|tj}|g}d}t j dd||||}t j rt jd||||\}|S)Ns SdcaFprintr3attrsrrr)r!convert_to_tensor_dtypesstringr1rr2r4)rrrr=r<r7s r>r&r&`sp   7%, &   ]Al#)s ?' ""$ lFG5 (' . SdcaOptimizer)out_example_state_dataout_delta_sparse_weightsout_delta_dense_weightsztrain.sdca_optimizersparse_example_indicessparse_feature_indicessparse_feature_valuesdense_featuresexample_weightsexample_labelssparse_indicessparse_weights dense_weightsexample_state_data loss_typel1l2num_loss_partitionsnum_inner_iterations adaptativec tjxstj}|j}|jrE t j |d||||||||||| d| d|d| d| d| d|}t j|}|St|||||||||| | | | | |||fd}|tur|St3|t4t6fst#d|zt9|}t3|t4t6fst#d|zt9||k7rt%dt9||fzt3|t4t6fst#d|zt9||k7rt%dt9||fzt3|t4t6fst#d|zt9||k7rt%dt9||fzt3|t4t6fst#d|zt9|}t3|t4t6fst#d|zt9|}t3|t4t6fst#d|zt9||k7rt%d t9||fzt;j<| d} t;j>| d} t;j>| d} t;j@| d} t;j@|d}|d!}t;jB|d} tEjF d'id |d |d |d|d|d|d|d|d|d| d| d| d| d| d|d|d|\}}}}|dd}t;jHrd|jKdd|jMdd"|jOd"d#|jOd#d$|jOd$d|jKdd|jKdd|jOdd|jOdf}|jP}t;jRd||||dd%|d%d%|zgz|d%|zdz}|dd&|d&dgz}t j|}|S#tj$r }tj||Yd}~nd}~wtj$rYnwxYw t|||||||||| | | | | |||fd}|tur|St|||||||||| | || | | ||| S#tj $rYFt"t$f$rwt'j(t*d t-d id |d |d |d|d|d|d|d|d|d| d| d| d| d| d|d|d|}|t&j.j0ur|cYSwxYw#t"t$f$rwt'j(t*d t-d id |d |d |d|d|d|d|d|d|d| d| d| d| d| d|d|d|}|t&j.j0ur|cYSwxYw)(aDistributed version of Stochastic Dual Coordinate Ascent (SDCA) optimizer for linear models with L1 + L2 regularization. As global optimization objective is strongly-convex, the optimizer optimizes the dual objective at each step. The optimizer applies each update one example at a time. Examples are sampled uniformly, and the optimizer is learning rate free and enjoys linear convergence rate. [Proximal Stochastic Dual Coordinate Ascent](http://arxiv.org/pdf/1211.2717v1.pdf).
Shai Shalev-Shwartz, Tong Zhang. 2012 $$Loss Objective = \sum f_{i} (wx_{i}) + (l2 / 2) * |w|^2 + l1 * |w|$$ [Adding vs. Averaging in Distributed Primal-Dual Optimization](http://arxiv.org/abs/1502.03508).
Chenxin Ma, Virginia Smith, Martin Jaggi, Michael I. Jordan, Peter Richtarik, Martin Takac. 2015 [Stochastic Dual Coordinate Ascent with Adaptive Probabilities](https://arxiv.org/abs/1502.08053).
Dominik Csiba, Zheng Qu, Peter Richtarik. 2015 Args: sparse_example_indices: A list of `Tensor` objects with type `int64`. a list of vectors which contain example indices. sparse_feature_indices: A list with the same length as `sparse_example_indices` of `Tensor` objects with type `int64`. a list of vectors which contain feature indices. sparse_feature_values: A list of `Tensor` objects with type `float32`. a list of vectors which contains feature value associated with each feature group. dense_features: A list of `Tensor` objects with type `float32`. a list of matrices which contains the dense feature values. example_weights: A `Tensor` of type `float32`. a vector which contains the weight associated with each example. example_labels: A `Tensor` of type `float32`. a vector which contains the label/target associated with each example. sparse_indices: A list with the same length as `sparse_example_indices` of `Tensor` objects with type `int64`. a list of vectors where each value is the indices which has corresponding weights in sparse_weights. This field maybe omitted for the dense approach. sparse_weights: A list with the same length as `sparse_example_indices` of `Tensor` objects with type `float32`. a list of vectors where each value is the weight associated with a sparse feature group. dense_weights: A list with the same length as `dense_features` of `Tensor` objects with type `float32`. a list of vectors where the values are the weights associated with a dense feature group. example_state_data: A `Tensor` of type `float32`. a list of vectors containing the example state data. loss_type: A `string` from: `"logistic_loss", "squared_loss", "hinge_loss", "smooth_hinge_loss", "poisson_loss"`. Type of the primal loss. Currently SdcaSolver supports logistic, squared and hinge losses. l1: A `float`. Symmetric l1 regularization strength. l2: A `float`. Symmetric l2 regularization strength. num_loss_partitions: An `int` that is `>= 1`. Number of partitions of the global loss function. num_inner_iterations: An `int` that is `>= 1`. Number of iterations per mini-batch. adaptative: An optional `bool`. Defaults to `True`. Whether to use Adaptive SDCA for the inner loop. name: A name for the operation (optional). Returns: A tuple of `Tensor` objects (out_example_state_data, out_delta_sparse_weights, out_delta_dense_weights). out_example_state_data: A `Tensor` of type `float32`. out_delta_sparse_weights: A list with the same length as `sparse_example_indices` of `Tensor` objects with type `float32`. out_delta_dense_weights: A list with the same length as `dense_features` of `Tensor` objects with type `float32`. rGrUrZrVrWrXrYN)rUrZrVrWrXrYrrrrKrLrMrNrOrPrQrRrSrTrSExpected list for 'sparse_example_indices' argument to 'sdca_optimizer' Op, not %r.SExpected list for 'sparse_feature_indices' argument to 'sdca_optimizer' Op, not %r.List argument 'sparse_feature_indices' to 'sdca_optimizer' Op with length %d must match length %d of argument 'sparse_example_indices'.KExpected list for 'sparse_indices' argument to 'sdca_optimizer' Op, not %r.List argument 'sparse_indices' to 'sdca_optimizer' Op with length %d must match length %d of argument 'sparse_example_indices'.KExpected list for 'sparse_weights' argument to 'sdca_optimizer' Op, not %r.List argument 'sparse_weights' to 'sdca_optimizer' Op with length %d must match length %d of argument 'sparse_example_indices'.RExpected list for 'sparse_feature_values' argument to 'sdca_optimizer' Op, not %r.KExpected list for 'dense_features' argument to 'sdca_optimizer' Op, not %r.JExpected list for 'dense_weights' argument to 'sdca_optimizer' Op, not %r.vList argument 'dense_weights' to 'sdca_optimizer' Op with length %d must match length %d of argument 'dense_features'.Tnum_sparse_featuresnum_sparse_features_with_valuesnum_dense_featuresr@)rG)*rrrrrr_SdcaOptimizerOutput_makerr r!r"r#_dispatcher_for_sdca_optimizerr%sdca_optimizer_eager_fallbackr'r(r)r*r sdca_optimizerr,r-r. isinstancelisttuplelenr1make_str make_floatmake_int make_boolr/r0r2get_attr_get_attr_bool _get_attr_intr3r4)rKrLrMrNrOrPrQrRrSrTrUrVrWrXrYrZrr5r6r7r8_attr_num_sparse_features%_attr_num_sparse_features_with_values_attr_num_dense_featuresr9r:r;r<r=s r>roroqsR    0h..0$ #\\  11 ot%; 5~); j$D"6K35I Kg%**73g nX- !7~ Ir2/Bj$  137 8G n$ n *T5M :  ')? @ AA""89 *T5M :  ')? @ AA  $==  E # $&?@ A BB NT5M 2  ')7 8 99 55  E ^ 78 9 :: NT5M 2  ')7 8 99 55  E ^ 78 9 :: )D%= 9  ')> ? @@+..C*D' NT5M 2  ')7 8 99!0 MD%= 1  ')6 7 88 33  = ] 56 7 88 ;7)2t$"2t$" ))*=?TU!**+?AWXJ!!*l;*# )::;0F;0F;0E;)7 ; *9 ; )7 ;)7;)7;(5;-?;$-;24;9;;.A;/C;%/;6:;Aq#xF QK' ""$3<< 4l  .0E 56/ AB"C$5$56J$KCLL&cll4.@#S%6%67L%M$ 67 9F::L vw8 BQK71Q)B%BCD DwqSlOlOmGn n' BQK712;- '' & &w /' .K  & &- ##At,,  # #   % . !#9  ..- iR1D  D  359 :g  & * "8  ..-  j B,?3$D JJ  # #  z " "" B !G@ !GEG !G:M !G;O !G1; !GBF !G g  ..<< < ' R Z    "d E:P E:P E9N E3A E4C E 3A E 3A E3A E2? E7I E.7 E<> ECE E8K E9M E/9 E@D E G i,,::: n ' sRAQ -A U9 RQ::RR&S>SU61BU64U69BW?=W?zraw_ops.SdcaOptimizercD t|ttfstd|zt |}t|ttfstd|zt ||k7rt dt ||fzt|ttfstd|zt ||k7rt dt ||fzt|ttfstd|zt ||k7rt dt ||fzt|ttfstd|zt |}t|ttfstd |zt |}t|ttfstd |zt ||k7rt d t ||fzt j| d } t j| d } t j| d} t j| d} t j|d}|d}t j|d}tj|tj}tj|tj}tj|tj}tj|tj}tj |tj}tj |tj}tj|tj}tj|tj}tj|tj}tj | tj} t|t|zt|zt|z||gzt|zt|zt|z| gz}d | d|d|d|d|d | d| d| d|f}t j"d||zdz||||}t j$rt j&d||||dd|dd|zgz|d|zdz}|dd|ddgz}t(j+|}|S)Nr\r]r^r_r`rarbrcrdrerfrUrVrWrXrYTrZrgrhris SdcaOptimizerr@rArGrj)rprqrrr(rsr)r1rtrurvrwr!convert_n_to_tensorrDint64float32rCrr2r4rkrl)rKrLrMrNrOrPrQrRrSrTrUrVrWrXrYrZrrr{r|r}r=r<r7s r>rnrnrs8 *T5M :  ')? @ AA""89 *T5M :  ')? @ AA  $==  E # $&?@ A BB NT5M 2  ')7 8 99 55  E ^ 78 9 :: NT5M 2  ')7 8 99 55  E ^ 78 9 :: )D%= 9  ')> ? @@+..C*D' NT5M 2  ')7 8 99!0 MD%= 1  ')6 7 88 33  = ] 56 7 88 ;7)2t$"2t$" ))*=?TU!**+?AWXJ!!*l;*334JGMMZ334JGMMZ223H'//Z++NGOOL.**?GOOL/)).'//J.++NGMMJ.++NGOOL.**='//J---.@'//R,-5K0LLtTiOjjmqsAnBBFUWeEffimn|i}}@DES@TTW[\iWjjn@mAA, L*2#%J0$D",.D  &   -/H506890:$0C"& ('""$ vw8 BQK71Q)B%BCD DwqSlOlOmGn n' BQK712;- '' & &w /' .rFSdcaOptimizerV2adaptivecL tjxstj}|j}|jrE t j |d||||||||||| d| d|d| d| d| d|}t j|}|St|t t"fst%d |zt'|}t|t t"fst%d |zt'||k7rt)d t'||fzt|t t"fst%d |zt'||k7rt)dt'||fzt|t t"fst%d|zt'||k7rt)dt'||fzt|t t"fst%d|zt'|}t|t t"fst%d|zt'|}t|t t"fst%d|zt'||k7rt)dt'||fzt+j,| d} t+j.| d} t+j.| d} t+j0| d} t+j0|d}|d}t+j2|d}t5j6 d&id|d|d|d|d|d|d|d|d|d| d| d| d| d| d|d|d |\}}}}|dd}t+j8rd|j;dd|j=dd!|j?d!d"|j?d"d#|j?d#d|j;dd|j;dd|j?dd|j?df}|j@}t+jBd||||dd$|d$d$|zgz|d$|zdz}|dd%|d%dgz}t j|}|S#tj$r }tj||Yd}~nd}~wtj$rYnwxYw t|||||||||| | || | | ||| S#tj$rYwxYw)'aDistributed version of Stochastic Dual Coordinate Ascent (SDCA) optimizer for linear models with L1 + L2 regularization. As global optimization objective is strongly-convex, the optimizer optimizes the dual objective at each step. The optimizer applies each update one example at a time. Examples are sampled uniformly, and the optimizer is learning rate free and enjoys linear convergence rate. [Proximal Stochastic Dual Coordinate Ascent](http://arxiv.org/pdf/1211.2717v1.pdf).
Shai Shalev-Shwartz, Tong Zhang. 2012 $$Loss Objective = \sum f_{i} (wx_{i}) + (l2 / 2) * |w|^2 + l1 * |w|$$ [Adding vs. Averaging in Distributed Primal-Dual Optimization](http://arxiv.org/abs/1502.03508).
Chenxin Ma, Virginia Smith, Martin Jaggi, Michael I. Jordan, Peter Richtarik, Martin Takac. 2015 [Stochastic Dual Coordinate Ascent with Adaptive Probabilities](https://arxiv.org/abs/1502.08053).
Dominik Csiba, Zheng Qu, Peter Richtarik. 2015 Args: sparse_example_indices: A list of `Tensor` objects with type `int64`. a list of vectors which contain example indices. sparse_feature_indices: A list with the same length as `sparse_example_indices` of `Tensor` objects with type `int64`. a list of vectors which contain feature indices. sparse_feature_values: A list of `Tensor` objects with type `float32`. a list of vectors which contains feature value associated with each feature group. dense_features: A list of `Tensor` objects with type `float32`. a list of matrices which contains the dense feature values. example_weights: A `Tensor` of type `float32`. a vector which contains the weight associated with each example. example_labels: A `Tensor` of type `float32`. a vector which contains the label/target associated with each example. sparse_indices: A list with the same length as `sparse_example_indices` of `Tensor` objects with type `int64`. a list of vectors where each value is the indices which has corresponding weights in sparse_weights. This field maybe omitted for the dense approach. sparse_weights: A list with the same length as `sparse_example_indices` of `Tensor` objects with type `float32`. a list of vectors where each value is the weight associated with a sparse feature group. dense_weights: A list with the same length as `dense_features` of `Tensor` objects with type `float32`. a list of vectors where the values are the weights associated with a dense feature group. example_state_data: A `Tensor` of type `float32`. a list of vectors containing the example state data. loss_type: A `string` from: `"logistic_loss", "squared_loss", "hinge_loss", "smooth_hinge_loss", "poisson_loss"`. Type of the primal loss. Currently SdcaSolver supports logistic, squared and hinge losses. l1: A `float`. Symmetric l1 regularization strength. l2: A `float`. Symmetric l2 regularization strength. num_loss_partitions: An `int` that is `>= 1`. Number of partitions of the global loss function. num_inner_iterations: An `int` that is `>= 1`. Number of iterations per mini-batch. adaptive: An optional `bool`. Defaults to `True`. Whether to use Adaptive SDCA for the inner loop. name: A name for the operation (optional). Returns: A tuple of `Tensor` objects (out_example_state_data, out_delta_sparse_weights, out_delta_dense_weights). out_example_state_data: A `Tensor` of type `float32`. out_delta_sparse_weights: A list with the same length as `sparse_example_indices` of `Tensor` objects with type `float32`. out_delta_dense_weights: A list with the same length as `dense_features` of `Tensor` objects with type `float32`. rrUrrVrWrXrYN)rUrrVrWrXrYrrVExpected list for 'sparse_example_indices' argument to 'sdca_optimizer_v2' Op, not %r.VExpected list for 'sparse_feature_indices' argument to 'sdca_optimizer_v2' Op, not %r.List argument 'sparse_feature_indices' to 'sdca_optimizer_v2' Op with length %d must match length %d of argument 'sparse_example_indices'.NExpected list for 'sparse_indices' argument to 'sdca_optimizer_v2' Op, not %r.List argument 'sparse_indices' to 'sdca_optimizer_v2' Op with length %d must match length %d of argument 'sparse_example_indices'.NExpected list for 'sparse_weights' argument to 'sdca_optimizer_v2' Op, not %r.List argument 'sparse_weights' to 'sdca_optimizer_v2' Op with length %d must match length %d of argument 'sparse_example_indices'.UExpected list for 'sparse_feature_values' argument to 'sdca_optimizer_v2' Op, not %r.NExpected list for 'dense_features' argument to 'sdca_optimizer_v2' Op, not %r.MExpected list for 'dense_weights' argument to 'sdca_optimizer_v2' Op, not %r.yList argument 'dense_weights' to 'sdca_optimizer_v2' Op with length %d must match length %d of argument 'dense_features'.TrKrLrMrNrOrPrQrRrSrTrrgrhrir@rj)r)"rrrrrr_SdcaOptimizerV2Outputrlrr r!r"r# sdca_optimizer_v2_eager_fallbackr'rprqrrr(rsr)r1rtrurvrwr/r0r2rxryrzr3r4)rKrLrMrNrOrPrQrRrSrTrUrVrWrXrYrrr5r6r7r8r{r|r}r9r:r;r<r=s r>sdca_optimizer_v2rsJ    0h..0$ #\\  11 '= 5~); :$D"&;35I Kg',,W5g n *T5M :  *,B C DD""89 *T5M :  *,B C DD  $==  E # $&?@ A BB NT5M 2  *,: ; << 55  E ^ 78 9 :: NT5M 2  *,: ; << 55  E ^ 78 9 :: )D%= 9  *,A B CC+..C*D' NT5M 2  *,: ; <<!0 MD%= 1  *,9 : ;; 33  = ] 56 7 88 ;7)2t$"2t$" ))*=?TU!**+?AWX H   * 5('8892H92H92G9+9 9 ,; 9 +9 9+99+99*79/A9&/9469;=90C91E9%-9489!QX QK' ""$3<< 4j  ,.C 56/ AB"C$5$56J$KCLL&cll4.@#S%6%67L%M$ 67 9F::L <: BQK71Q)B%BCD DwqSlOlOmGn n' BQK712;- '' " ( ( 1' .W  & &- ##At,,  # #     - "8  ..-  H %83$D JJ  # #   s1AP$$Q+7QQ+*Q+/R R#"R#zraw_ops.SdcaOptimizerV2cD t|ttfstd|zt |}t|ttfstd|zt ||k7rt dt ||fzt|ttfstd|zt ||k7rt dt ||fzt|ttfstd|zt ||k7rt dt ||fzt|ttfstd|zt |}t|ttfstd |zt |}t|ttfstd |zt ||k7rt d t ||fzt j| d } t j| d } t j| d} t j| d} t j|d}|d}t j|d}tj|tj}tj|tj}tj|tj}tj|tj}tj |tj}tj |tj}tj|tj}tj|tj}tj|tj}tj | tj} t|t|zt|zt|z||gzt|zt|zt|z| gz}d | d|d|d|d|d | d| d| d|f}t j"d||zdz||||}t j$rt j&d||||dd|dd|zgz|d|zdz}|dd|ddgz}t(j+|}|S)NrrrrrrrrrrrrUrVrWrXrYTrrgrhrisSdcaOptimizerV2r@rArrj)rprqrrr(rsr)r1rtrurvrwr!rrDrrrCrr2r4rrl)rKrLrMrNrOrPrQrRrSrTrUrVrWrXrYrrrr{r|r}r=r<r7s r>rrs9 *T5M :  *,B C DD""89 *T5M :  *,B C DD  $==  E # $&?@ A BB NT5M 2  *,: ; << 55  E ^ 78 9 :: NT5M 2  *,: ; << 55  E ^ 78 9 :: )D%= 9  *,A B CC+..C*D' NT5M 2  *,: ; <<!0 MD%= 1  *,9 : ;; 33  = ] 56 7 88 ;7)2t$"2t$" ))*=?TU!**+?AWX H   * 5(334JGMMZ334JGMMZ223H'//Z++NGOOL.**?GOOL/)).'//J.++NGMMJ.++NGOOL.**='//J---.@'//R,-5K0LLtTiOjjmqsAnBBFUWeEffimn|i}}@DES@TTW[\iWjjn@mAA, J2#%J0$D",.D  &   /1J526892:$0C"& ('""$ <: BQK71Q)B%BCD DwqSlOlOmGn n' BQK712;- '' " ( ( 1' .rFztrain.sdca_shrink_l1weightsc ztjxstj}|j}|jr t dt ||||fd}|t ur|St|ttfstd|zt|}tj|d}tj|d} tjd||||\}}} } | S#tt f$rJt#j$t&dt)||||}|t"j*j,ur|cYSwxYw) aApplies L1 regularization shrink step on the parameters. Args: weights: A list of `Tensor` objects with type mutable `float32`. a list of vectors where each value is the weight associated with a feature group. l1: A `float`. Symmetric l1 regularization strength. l2: A `float`. Symmetric l2 regularization strength. Should be a positive float. name: A name for the operation (optional). Returns: The created Operation. Ksdca_shrink_l1 op does not support eager execution. Arg 'weights' is a ref.NzDExpected list for 'weights' argument to 'sdca_shrink_l1' Op, not %r.rVrW SdcaShrinkL1)rrVrWrr)rrrr RuntimeError_dispatcher_for_sdca_shrink_l1r%rprqrrr(rsr1rur/r0r)r*r sdca_shrink_l1r,r-r.) rrVrWrr5r6r7_attr_num_featuresr9r:r;s r>rrs?&    0h..0$ #\\ d ee, "b$ $(Gn$ n GdE] +  ')0 1 227|2t$"2t$"  )::B2DBAq#x * Z    "d7rbtL Gi,,::: n  sC!!AD:8D:zraw_ops.SdcaShrinkL1ctd)Nr)r)rrVrWrrs r>sdca_shrink_l1_eager_fallbackrsbccrF)N)TN)B__doc__ collectionstensorflow.pythonrtensorflow.python.eagerrrrrrr1tensorflow.python.frameworkrrDtensorflow.security.fuzzing.pyr_atypesr _op_def_registryr r!r r/"tensorflow.python.util.deprecationr tensorflow.python.utilr r* tensorflow.python.util.tf_exportrtypingrrrtyping_extensionsradd_fallback_dispatch_listadd_type_based_api_dispatcherStringInt64r+ to_raw_opr_tf_type_based_dispatcherDispatchr$r& namedtuplerkFloat32strfloatintboolrorGrmrnrrrrrrrrrrFr>rs^  67179FK3IC86%%' %% (( "#$)*=ygnn!45=YsT[TaTaOaEb=+%)&=~-Y+ ,^T^^K-H I )CCLL iW^^0C&D T]^acjcpcp^pTq .{--UW  %% (( %&',-w9T#Y 5M+Nwhqrvwzr{~E~K~KsKiLwenostwoxzAzIzIoIeJw\efjknfoqxq@q@f@\AwT]^acjcrcr^rTswENORT[TcTcOcEdwv@DEH@IKRKXKX@XvYwktuyz}u~@G@O@OuOkPwajkopsktv}vEvEkEaFw\efikrkzkzfz\{wH K wQ V w\ a wx { wS V wc g w.()&wr3 12>4>>.3QR !/!I!I!R!RW)DIw}}D\:]WxABFGJBKMTMZMZBZx[Wt}~BCF~GIPIXIX~XtYWktuyz}u~@G@O@OuOkPWclmpryrArAmAcBWT]^acjcrcr^rTsWENOSTWOXZaZgZgOgEhWzCDHILDMOVO^O^D^z_Wpyz~BzCELETETzTpUWktuxzA zI zI uI kJ WW Z W` e Wk p WG J Wb e Ws w Wr0//UW }iS 7==8P.Q}ktuyz}u~AHANANvNlO}hqrvwzr{}D}L}LrLhM}_himnqirt{tCtCiC_D}W`adfmfufuauWv}HQRUW^WfWfRfHg}yBCGHKCLNUN[N[C[y\}nwx|}@xACJCRCRxRnS}dmnrsvnwy@yHyHnHdI}_hilnun}n}i}_~}K N }T Y }_ d }{ ~ }V Y }d h }~7)56~t~~FW7XYWYtCyRYR_R_G_=`W{DEIJMENPWP]P]E]{^Ww@AEFIAJLSL[L[A[w\Wnwx|}@xACJCRCRxRnSWfopsu|uDuDpDfEWW`adfmfufuauWvWHQRVWZR[]d]j]jRjHkW}FGKLOGPRYRaRaGa}bWs|}ABE}FHOHWHW}WsXWnwx{}D }L }L xL nM WZ ] Wc h Wn s WJ M We h Wt x Wt %% (( %&',-* Id3i&@A* u* RW* .()&* V1y/01OP !/!I!I!R!Rd9T#Y5O+PdV[dafdrF