6bi=SrSSKJr SSKJr SSKJr SSKJr SSKJr SSKJ r SSKJ r SS K J r SS K Jr SS KJr \"S /S 9"SS\ R$55r\"S/S 9"SS\ 55rg)z Long Short-Term Memory V1 layer.) activations) constraints) initializers) regularizers) InputSpec)lstm) rnn_utils)RNN) tf_logging) keras_exportzkeras.layers.LSTMCell)v1cN^\rSrSrSrSU4SjjrSrU=r$)LSTMCell aCell class for the LSTM layer. Args: units: Positive integer, dimensionality of the output space. activation: Activation function to use. Default: hyperbolic tangent (`tanh`). If you pass `None`, no activation is applied (ie. "linear" activation: `a(x) = x`). recurrent_activation: Activation function to use for the recurrent step. Default: hard sigmoid (`hard_sigmoid`). If you pass `None`, no activation is applied (ie. "linear" activation: `a(x) = x`). use_bias: Boolean, whether the layer uses a bias vector. kernel_initializer: Initializer for the `kernel` weights matrix, used for the linear transformation of the inputs. recurrent_initializer: Initializer for the `recurrent_kernel` weights matrix, used for the linear transformation of the recurrent state. bias_initializer: Initializer for the bias vector. unit_forget_bias: Boolean. If True, add 1 to the bias of the forget gate at initialization. Setting it to true will also force `bias_initializer="zeros"`. This is recommended in [Jozefowicz et al., 2015]( http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf) kernel_regularizer: Regularizer function applied to the `kernel` weights matrix. recurrent_regularizer: Regularizer function applied to the `recurrent_kernel` weights matrix. bias_regularizer: Regularizer function applied to the bias vector. kernel_constraint: Constraint function applied to the `kernel` weights matrix. recurrent_constraint: Constraint function applied to the `recurrent_kernel` weights matrix. bias_constraint: Constraint function applied to the bias vector. dropout: Float between 0 and 1. Fraction of the units to drop for the linear transformation of the inputs. recurrent_dropout: Float between 0 and 1. Fraction of the units to drop for the linear transformation of the recurrent state. Call arguments: inputs: A 2D tensor. states: List of state tensors corresponding to the previous timestep. training: Python boolean indicating whether the layer should behave in training mode or in inference mode. Only relevant when `dropout` or `recurrent_dropout` is used. c >[TU]"U40SU_SU_SU_SU_SU_SU_SU_SU _S U _S U _S U _S U _S U_SU_SU_SURSS5_UD6 g)N activationrecurrent_activationuse_biaskernel_initializerrecurrent_initializerbias_initializerunit_forget_biaskernel_regularizerrecurrent_regularizerbias_regularizerkernel_constraintrecurrent_constraintbias_constraintdropoutrecurrent_dropoutimplementation)super__init__pop)selfunitsrrrrrrrrrrrrrrr kwargs __class__s Y/home/james-whalen/.local/lib/python3.13/site-packages/tf_keras/src/layers/rnn/lstm_v1.pyr$LSTMCell.__init__Ts(   ! "6   2  #8  . .  2 #8 . 0 "6 ,  0! ""::&6:% )tanh hard_sigmoidTglorot_uniform orthogonalzerosTNNNNNNr3)__name__ __module__ __qualname____firstlineno____doc__r$__static_attributes__ __classcell__r)s@r*rr sB0j++* "!#' ' r,rzkeras.layers.LSTMc^\rSrSrSrSU4SjjrSU4Sjjr\S5r\S5r \S5r \S5r \S 5r \S 5r \S 5r\S 5r\S 5r\S5r\S5r\S5r\S5r\S5r\S5r\S5r\S5rU4Sjr\S5rSrU=r$)LSTM~aLong Short-Term Memory layer - Hochreiter 1997. Note that this cell is not optimized for performance on GPU. Please use `tf.compat.v1.keras.layers.CuDNNLSTM` for better performance on GPU. Args: units: Positive integer, dimensionality of the output space. activation: Activation function to use. Default: hyperbolic tangent (`tanh`). If you pass `None`, no activation is applied (ie. "linear" activation: `a(x) = x`). recurrent_activation: Activation function to use for the recurrent step. Default: hard sigmoid (`hard_sigmoid`). If you pass `None`, no activation is applied (ie. "linear" activation: `a(x) = x`). use_bias: Boolean, whether the layer uses a bias vector. kernel_initializer: Initializer for the `kernel` weights matrix, used for the linear transformation of the inputs.. recurrent_initializer: Initializer for the `recurrent_kernel` weights matrix, used for the linear transformation of the recurrent state. bias_initializer: Initializer for the bias vector. unit_forget_bias: Boolean. If True, add 1 to the bias of the forget gate at initialization. Setting it to true will also force `bias_initializer="zeros"`. This is recommended in [Jozefowicz et al., 2015]( http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf). kernel_regularizer: Regularizer function applied to the `kernel` weights matrix. recurrent_regularizer: Regularizer function applied to the `recurrent_kernel` weights matrix. bias_regularizer: Regularizer function applied to the bias vector. activity_regularizer: Regularizer function applied to the output of the layer (its "activation"). kernel_constraint: Constraint function applied to the `kernel` weights matrix. recurrent_constraint: Constraint function applied to the `recurrent_kernel` weights matrix. bias_constraint: Constraint function applied to the bias vector. dropout: Float between 0 and 1. Fraction of the units to drop for the linear transformation of the inputs. recurrent_dropout: Float between 0 and 1. Fraction of the units to drop for the linear transformation of the recurrent state. return_sequences: Boolean. Whether to return the last output in the output sequence, or the full sequence. return_state: Boolean. Whether to return the last state in addition to the output. go_backwards: Boolean (default False). If True, process the input sequence backwards and return the reversed sequence. stateful: Boolean (default False). If True, the last state for each sample at index i in a batch will be used as initial state for the sample of index i in the following batch. unroll: Boolean (default False). If True, the network will be unrolled, else a symbolic loop will be used. Unrolling can speed-up a RNN, although it tends to be more memory-intensive. Unrolling is only suitable for short sequences. time_major: The shape format of the `inputs` and `outputs` tensors. If True, the inputs and outputs will be in shape `(timesteps, batch, ...)`, whereas in the False case, it will be `(batch, timesteps, ...)`. Using `time_major = True` is a bit more efficient because it avoids transposes at the beginning and end of the RNN calculation. However, most TensorFlow data is batch-major, so by default this function accepts input and emits output in batch-major form. Call arguments: inputs: A 3D tensor. mask: Binary tensor of shape `(samples, timesteps)` indicating whether a given timestep should be masked. An individual `True` entry indicates that the corresponding timestep should be utilized, while a `False` entry indicates that the corresponding timestep should be ignored. training: Python boolean indicating whether the layer should behave in training mode or in inference mode. This argument is passed to the cell when calling it. This is only relevant if `dropout` or `recurrent_dropout` is used. initial_state: List of initial state tensors to be passed to the first call of the cell. c >URSS5nUS:Xa[R"S5 SU;aSURS50nO0n[U40SU_SU_SU_S U_S U_S U_S U_S U _SU _SU _SU _SU_SU_SU_SU_SU_SUR S5_SUR SS5_SS_UD6n[ TU]"U4UUUUUS.UD6 [R"U 5Ul[SS9/Ul g)Nr!r"rzm`implementation=0` has been deprecated, and now defaults to `implementation=1`.Please update your layer call.enable_caching_devicerrrrrrrrrrrrrrr dtype trainableTname lstm_cell)return_sequences return_state go_backwardsstatefulunroll)ndim) r%loggingwarningrgetr#r$ractivity_regularizerr input_spec)r&r'rrrrrrrrrrrOrrrrr rErFrGrHrIr(r! cell_kwargscellr)s r*r$ LSTM.__init__s4 $4a8 Q  OO1  #f ,'4K)LKK  ! "6   2  #8  . .  2 #8 . 0 "6 ,  0! "*# $**W%% &jjd3' (+ .   -%%   %1$4$45I$J!$!,-r,c >[TU]XX4S9$)N)masktraining initial_state)r#call)r&inputsrUrVrWr)s r*rX LSTM.callsw|   r,c.URR$N)rRr'r&s r*r' LSTM.units$syyr,c.URR$r\)rRrr]s r*rLSTM.activation(syy###r,c.URR$r\)rRrr]s r*rLSTM.recurrent_activation,yy---r,c.URR$r\)rRrr]s r*r LSTM.use_bias0syy!!!r,c.URR$r\)rRrr]s r*rLSTM.kernel_initializer4yy+++r,c.URR$r\)rRrr]s r*rLSTM.recurrent_initializer8yy...r,c.URR$r\)rRrr]s r*rLSTM.bias_initializer<yy)))r,c.URR$r\)rRrr]s r*rLSTM.unit_forget_bias@rnr,c.URR$r\)rRrr]s r*rLSTM.kernel_regularizerDrhr,c.URR$r\)rRrr]s r*rLSTM.recurrent_regularizerHrkr,c.URR$r\)rRrr]s r*rLSTM.bias_regularizerLrnr,c.URR$r\)rRrr]s r*rLSTM.kernel_constraintPyy***r,c.URR$r\)rRrr]s r*rLSTM.recurrent_constraintTrcr,c.URR$r\)rRrr]s r*rLSTM.bias_constraintXsyy(((r,c.URR$r\)rRrr]s r*r LSTM.dropout\syy   r,c.URR$r\)rRr r]s r*r LSTM.recurrent_dropout`ryr,c.URR$r\)rRr!r]s r*r!LSTM.implementationdsyy'''r,c>0SUR_S[R"UR5_S[R"UR5_SUR _S[ R"UR5_S[ R"UR5_S[ R"UR5_SUR_S [R"UR5_S [R"UR5_S [R"UR5_S [R"UR5_S [ R"UR"5_S[ R"UR$5_S[ R"UR&5_SUR(_SUR*_SUR,0EnUR/[0R2"UR455 [6TU]q5nUS [;[=UR?55[=UR?55-5$)Nr'rrrrrrrrrrrOrrrrr r!rR) r'r serializerrrrrrrrrrrrrOrrrrrr r!updater config_for_enable_caching_devicerRr# get_configdictlistitems)r&config base_configr)s r*rLSTM.get_confighs9! TZZ! +//@!  #K$9$9))%!  !  !,"8"8''#!  $\%;%;**&!   6 6t7L7L M!   5 5!  !,"8"8''#! $ $\%;%;**&%! *  6 6t7L7L M+! , #L$:$:))%-! 2 !6!6t7M7M!N3! 4 #K$9$9))%5! : {44T5I5IJ;! < t||=! > !7!7?! @ d11A! D  i@@KLg(*  D**,-V\\^0DDEEr,c:SU;aUSS:XaSUS'U"S0UD6$)Nr!rr"r-r-)clsrs r* from_configLSTM.from_configs. v %&1A*Ba*G'(F# $}V}r,)rOrP)r.r/Tr0r1r2TNNNNNNNr3r3FFFFF)NNN)r4r5r6r7r8r$rXpropertyr'rrrrrrrrrrrrrrr r!r classmethodrr9r:r;s@r*r=r=~sSp++* "!!/H.T $$.."",,//****,,//**++..))!!++((&FPr,r=N)r8 tf_keras.srcrrrrtf_keras.src.engine.input_specrtf_keras.src.layers.rnnrr tf_keras.src.layers.rnn.base_rnnr tensorflow.python.platformr rL tensorflow.python.util.tf_exportr rr=r-r,r*rs'%$%%4(-0=9)*+Z t}}Z ,Z z%&'U3U(Ur,