o ?eH%@sdZddlZddlmZddlmZddlmZddlm Z ddl m Z ddlmZdd l mZdd l mZdd l mZdd lmZdd lmZddlmZddlmZm Z ej!ej"edgdedd@dej#ej$dej#ej%fddZ&ede'e&Z(e&j)j*Z+dej#ej$dej#ej%fddZ,e-dgdZ.ej!ej"edgdeddAde ej#ej%de ej#ej%de ej#ej/d e ej#ej/d!ej#ej/d"ej#ej/d#e ej#ej%d$e ej#ej/d%e ej#ej/d&ej#ej/d'e0d(e1d)e1d*e2d+e2d,e3f d-d.Z4ed/e'e4Z5e4j)j*Z6de ej#ej%de ej#ej%de ej#ej/d e ej#ej/d!ej#ej/d"ej#ej/d#e ej#ej%d$e ej#ej/d%e ej#ej/d&ej#ej/d'e0d(e1d)e1d*e2d+e2d,e3f d0d1Z7e-d2gdZ8dAde ej#ej%de ej#ej%de ej#ej/d e ej#ej/d!ej#ej/d"ej#ej/d#e ej#ej%d$e ej#ej/d%e ej#ej/d&ej#ej/d'e0d(e1d)e1d*e2d+e2d3e3f d4d5Z9ed6e'e9Z:de ej#ej%de ej#ej%de ej#ej/d e ej#ej/d!ej#ej/d"ej#ej/d#e ej#ej%d$e ej#ej/d%e ej#ej/d&ej#ej/d'e0d(e1d)e1d*e2d+e2d3e3f d7d8Z;ej!ej"ed9gded9d@d:e ej#ej/d(e1d)e1fd;d<Zd:e ej#ej/d(e1d)e1fd>d?Z?dS)BzUPython 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)TypeVarListztrain.sdca_fprint)v1inputreturnc Cstjpt}|j}|jrzz t|d||}|WStjy1}z t ||WYd}~nd}~wtj y:Ynwzt ||fd}|t urJ|WSt |||dWStjy[Yqttfyyttdt||d}|tjjurx|YSwt ||fd}|t ur|Sztjd||d\}}}}Wnttfyttdt||d}|tjjur|YSw|dd}trd} |j} td| | ||\}|S)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_flatrrc/home/www/facesmatcher.com/pyenv/lib/python3.10/site-packages/tensorflow/python/ops/gen_sdca_ops.pyr'sv   r'zraw_ops.SdcaFprintcCsPt|tj}|g}d}tjdd||||d}tr#td||||\}|S)Ns SdcaFprintr/attrsrrr)rconvert_to_tensor_dtypesstringr-rr.r0)rrrr9r8r3rrr:r"_s r" SdcaOptimizer)Zout_example_state_dataZout_delta_sparse_weightsZout_delta_dense_weightsztrain.sdca_optimizerTsparse_example_indicessparse_feature_indicessparse_feature_valuesdense_featuresexample_weightsexample_labelssparse_indicessparse_weights dense_weightsexample_state_data loss_typel1l2num_loss_partitionsnum_inner_iterations adaptativecCstjpt}|j}|jrz%t|d||||||||||| d| d|d| d| d| d|}t|}|WStj yK}z t ||WYd}~nd}~wtj yTYnwz4t |||||||||| | | | | |||fd}|turs|WSt|||||||||| | || | | |||d WStjyYqttfyttd td id |d |d |d|d|d|d|d|d|d| d| d| d| d| d|d|d|}|tjjur|YSwt |||||||||| | | | | |||fd}|tur|St|ttfstd|t|}t|ttfs"td|t||kr3tdt||ft|ttfsAtd|t||krRtdt||ft|ttfs`td|t||krqtdt||ft|ttfstd|t|}t|ttfstd|t|}t|ttfstd|t||krtd t||ft| d} t| d} t| d} t | d} t |d}|durd!}t!|d}z@t"j# d'id |d |d |d|d|d|d|d|d|d| d| d| d| d| d|d|d|\}}}}WnSttfyrttd td id |d |d |d|d|d|d|d|d|d| d| d| d| d| d|d|d|}|tjjurq|YSw|dd}t$rd|%dd|&dd"|'d"d#|'d#d$|'d$d|%dd|%dd|'dd|'df}|j(}t)d||||dd%|d%d%|g|d%|d}|dd&|d&dg}t|}|S)(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`. rArLrQrMrNrOrPN)rLrQrMrNrOrPrrrrBrCrDrErFrGrHrIrJrKrSExpected 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;)rA)*rrrrrr_SdcaOptimizerOutput_makerrrrr_dispatcher_for_sdca_optimizerr!sdca_optimizer_eager_fallbackr#r$r%r&r sdca_optimizerr(r)r* isinstancelisttuplelenr-make_str make_floatmake_int make_boolr+r,r.get_attr_get_attr_bool _get_attr_intr/r0)rBrCrDrErFrGrHrIrJrKrLrMrNrOrPrQrr1r2r3r4_attr_num_sparse_features%_attr_num_sparse_features_with_values_attr_num_dense_featuresr5r6r7r8r9rrr:repsI                                              . rezraw_ops.SdcaOptimizercC*t|ttfs td|t|}t|ttfstd|t||kr.tdt||ft|ttfs;td|t||krKtdt||ft|ttfsXtd|t||krhtdt||ft|ttfsutd|t|}t|ttfstd |t|}t|ttfstd |t||krtd t||ft| d } t| d } t| d} t | d} t |d}|durd}t |d}t |t j}t |t j}t |t j}t |t j}t |t j}t |t j}t |t j}t |t j}t |t j}t | t j} t|t|t|t|||gt|t|t|| g}d | d|d|d|d|d | d| d| d|f}tjd||d||||d}trjtd||||dd|dd|g|d|d}|dd|ddg}t|}|S)NrRrSrTrUrVrWrXrYrZr[r\rLrMrNrOrPTrQr]r^r_s SdcaOptimizerr;r<rAr`)rfrgrhr$rir%r-rjrkrlrmrconvert_n_to_tensorr?int64float32r>rr.r0rarb)rBrCrDrErFrGrHrIrJrKrLrMrNrOrPrQrrrqrrrsr9r8r3rrr:rdq              F  . rdSdcaOptimizerV2adaptivecCstjpt}|j}|jrwz%t|d||||||||||| d| d|d| d| d| d|}t|}|WStj yK}z t ||WYd}~nd}~wtj yTYnwzt |||||||||| | || | | |||d WStjyvYnwt|ttfstd |t|}t|ttfstd |t||krtd t||ft|ttfstd |t||krtdt||ft|ttfstd|t||krtdt||ft|ttfstd|t|}t|ttfstd|t|}t|ttfstd|t||kr tdt||ft| d} t| d} t| d} t| d} t|d}|durEd}t|d}tj d&id|d|d|d|d|d|d|d|d|d| d| d| d| d| d|d|d |\}}}}|dd}trd|dd|dd!|d!d"|d"d#|d#d|dd|dd|dd|df}|j }t!d||||dd$|d$d$|g|d$|d}|dd%|d%dg}t|}|S)'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`. ryrLrzrMrNrOrPN)rLrzrMrNrOrPrrVExpected 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'.TrBrCrDrErFrGrHrIrJrKrr]r^r_r;r`)ry)"rrrrrr_SdcaOptimizerV2Outputrbrrrrr sdca_optimizer_v2_eager_fallbackr#rfrgrhr$rir%r-rjrkrlrmr+r,r.rnrorpr/r0)rBrCrDrErFrGrHrIrJrKrLrMrNrOrPrzrr1r2r3r4rqrrrsr5r6r7r8r9rrr:sdca_optimizer_v2sNE                               . rzraw_ops.SdcaOptimizerV2cCrt)Nr{r|r}r~rrrrrrrrLrMrNrOrPTrzr]r^r_sSdcaOptimizerV2r;r<ryr`)rfrgrhr$rir%r-rjrkrlrmrrur?rvrwr>rr.r0rrb)rBrCrDrErFrGrHrIrJrKrLrMrNrOrPrzrrrqrrrsr9r8r3rrr:rrxrztrain.sdca_shrink_l1weightsc Cstjpt}|j}|jrtdt||||fd}|tur |St|tt fs-t d|t |}t |d}t |d}ztjd||||d\}}} } W| St tfyottdt||||d}|tjjurn|YSw) 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.rMrN SdcaShrinkL1)rrMrNrr)rrrr RuntimeError_dispatcher_for_sdca_shrink_l1r!rfrgrhr$rir-rkr+r,r%r&r sdca_shrink_l1r(r)r*) rrMrNrr1r2r3Z_attr_num_featuresr5r6r7rrr:rs>      rzraw_ops.SdcaShrinkL1cCstd)Nr)r)rrMrNrrrrr:sdca_shrink_l1_eager_fallbacksr)N)TN)@__doc__ collectionsZtensorflow.pythonrZtensorflow.python.eagerrrrrrr-Ztensorflow.python.frameworkrr?Ztensorflow.security.fuzzing.pyrZ_atypesrZ_op_def_registryr rr r+Z"tensorflow.python.util.deprecationr Ztensorflow.python.utilr r&Z tensorflow.python.util.tf_exportr typingrrZadd_fallback_dispatch_listZadd_type_based_api_dispatcherZTensorFuzzingAnnotationStringZInt64r'Z to_raw_oprZ_tf_type_based_dispatcherZDispatchr r" namedtupleraZFloat32strfloatintboolrerArcrdrrryrrrrrrrrr:sh             ,?"  zY@Z ,+&