2VhO\tddlZddlmZddlmZddlmZddlmZddl m Z GddeZ ed d gGd d e Z ed dgGdde Z eddgGdde ZeddgGdde ZeddgGddeZeddgGdd eZed!d"gGd#d$eZed%d&gGd'd(eZed)d*gGd+d,eZed-d.gGd/d0eZd1Zegd2Gd3d4e Zy)5N)ops) keras_export)random) Initializer)serialization_libceZdZddZdZy)RandomInitializerNc||_|tj}||_ yt|trt j |}||_ yt|ttjfstd|||_ y)Nzf`seed` argument should be an instance of `keras.random.SeedGenerator()` or an integer. Received: seed=) _init_seedrmake_default_seed isinstancedictrdeserialize_keras_objectint SeedGenerator ValueErrorseed)selfrs Z/home/dcms/DCMS/lib/python3.12/site-packages/keras/src/initializers/random_initializers.py__init__zRandomInitializer.__init__ s <++-D d #$==dCD D3(<(<"=>""&)   cHtj|j}d|iSNrrserialize_keras_objectr )r seed_configs r get_configzRandomInitializer.get_configs!'>>tO  $$rN)__name__ __module__ __qualname__rrrrr r s  %rr zkeras.initializers.RandomNormalz keras.initializers.random_normalc6eZdZdZdfd ZddZfdZxZS) RandomNormalaERandom normal initializer. Draws samples from a normal distribution for given parameters. Examples: >>> # Standalone usage: >>> initializer = RandomNormal(mean=0.0, stddev=1.0) >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = RandomNormal(mean=0.0, stddev=1.0) >>> layer = Dense(3, kernel_initializer=initializer) Args: mean: A python scalar or a scalar keras tensor. Mean of the random values to generate. stddev: A python scalar or a scalar keras tensor. Standard deviation of the random values to generate. seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. cB||_||_t| |yNrmeanstddevsuperrrr)r*r __class__s rrzRandomNormal.__init__B!   d#rcrtj||j|j|j|SN)shaper)r*rdtype)rnormalr)r*rrr1r2s r__call__zRandomNormal.__call__Gs.}};;   rc^t|}|j|jd}i||SN)r)r*r+rr)r*r base_configconfigr-s rrzRandomNormal.get_configP2g(* ))t{{;(+(((r皙?Nrrr r!__doc__rr5r __classcell__r-s@rr$r$:$  ))rr$z"keras.initializers.TruncatedNormalz#keras.initializers.truncated_normalc6eZdZdZdfd ZddZfdZxZS)TruncatedNormalaInitializer that generates a truncated normal distribution. The values generated are similar to values from a `RandomNormal` initializer, except that values more than two standard deviations from the mean are discarded and re-drawn. Examples: >>> # Standalone usage: >>> initializer = TruncatedNormal(mean=0., stddev=1.) >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = TruncatedNormal(mean=0., stddev=1.) >>> layer = Dense(3, kernel_initializer=initializer) Args: mean: A python scalar or a scalar keras tensor. Mean of the random values to generate. stddev: A python scalar or a scalar keras tensor. Standard deviation of the random values to generate. seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. cB||_||_t| |yr&r(r,s rrzTruncatedNormal.__init__}r.rcrtj||j|j|j|Sr0)rtruncated_normalr)r*rr4s rr5zTruncatedNormal.__call__s0&&;;   rc^t|}|j|jd}i||Sr7r8r9s rrzTruncatedNormal.get_configr<rr=rr@rCs@rrFrFVs@$  ))rrFz keras.initializers.RandomUniformz!keras.initializers.random_uniformc6eZdZdZdfd ZddZfdZxZS) RandomUniformazRandom uniform initializer. Draws samples from a uniform distribution for given parameters. Examples: >>> # Standalone usage: >>> initializer = RandomUniform(minval=0.0, maxval=1.0) >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = RandomUniform(minval=0.0, maxval=1.0) >>> layer = Dense(3, kernel_initializer=initializer) Args: minval: A python scalar or a scalar keras tensor. Lower bound of the range of random values to generate (inclusive). maxval: A python scalar or a scalar keras tensor. Upper bound of the range of random values to generate (exclusive). seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. cB||_||_t| |yr&)minvalmaxvalr+r)rrNrOrr-s rrzRandomUniform.__init__s!   d#rcrtj||j|j|j|S)N)r1rNrOrr2)runiformrNrOrr4s rr5zRandomUniform.__call__s.~~;;;;   rc^t|}|j|jd}i||S)N)rNrO)r+rrNrOr9s rrzRandomUniform.get_configs2g(*  KK4;;?(+(((r)gr?Nrr@rCs@rrLrLrDrrLz"keras.initializers.VarianceScalingz#keras.initializers.variance_scalingc>eZdZdZ dfd ZddZfdZxZS)VarianceScalingaInitializer that adapts its scale to the shape of its input tensors. With `distribution="truncated_normal" or "untruncated_normal"`, samples are drawn from a truncated/untruncated normal distribution with a mean of zero and a standard deviation (after truncation, if used) `stddev = sqrt(scale / n)`, where `n` is: - number of input units in the weight tensor, if `mode="fan_in"` - number of output units, if `mode="fan_out"` - average of the numbers of input and output units, if `mode="fan_avg"` With `distribution="uniform"`, samples are drawn from a uniform distribution within `[-limit, limit]`, where `limit = sqrt(3 * scale / n)`. Examples: >>> # Standalone usage: >>> initializer = VarianceScaling( scale=0.1, mode='fan_in', distribution='uniform') >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = VarianceScaling( scale=0.1, mode='fan_in', distribution='uniform') >>> layer = Dense(3, kernel_initializer=initializer) Args: scale: Scaling factor (positive float). mode: One of `"fan_in"`, `"fan_out"`, `"fan_avg"`. distribution: Random distribution to use. One of `"truncated_normal"`, `"untruncated_normal"`, or `"uniform"`. seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. c|dkrtd|hd}||vrtd|d||j}|dk(rd}hd}||vrtd |d |||_||_||_t || y) Nr>z9Argument `scale` must be positive float. Received: scale=>fan_infan_avgfan_outzInvalid `mode` argument: z. Please use one of r3rI>rQrIuntruncated_normalz!Invalid `distribution` argument: z.Please use one of r')rlowerscalemode distributionr+r)rr[r\r]r allowed_modesallowed_distributionsr-s rrzVarianceScaling.__init__s C<##('+ 9 } $+D62%%2O5 $))+ 8 #-L!  4 43L>B%%:$;=   ( d#rc||j}t|\}}|jdk(r|td|z}n4|jdk(r|td|z}n|td||zdz z}|jdk(r)r)r*r2rrYg@)rNrOr2r) r[ compute_fansr\maxr]mathsqrtrrIrr3rQ)rr1r2r[rVrXr*limits rr5zVarianceScaling.__call__s( &u- 99 Sf% %E YY) # Sg& &E Sv/367 7E    2 2YYu%(;;F**Ce$))   "6 6YYu%F==Ce$)) IIcEk*E>>ufU%dii rctt|}|j|j|jd}i||S)N)r[r\r])r+rr[r\r]r9s rrzVarianceScaling.get_config6s@g(* ZZII --  )+(((r)rarVrINrr@rCs@rrTrTs,'V '  "$H2))rrTz keras.initializers.GlorotUniformz!keras.initializers.glorot_uniformc*eZdZdZdfd ZdZxZS) GlorotUniformahThe Glorot uniform initializer, also called Xavier uniform initializer. Draws samples from a uniform distribution within `[-limit, limit]`, where `limit = sqrt(6 / (fan_in + fan_out))` (`fan_in` is the number of input units in the weight tensor and `fan_out` is the number of output units). Examples: >>> # Standalone usage: >>> initializer = GlorotUniform() >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = GlorotUniform() >>> layer = Dense(3, kernel_initializer=initializer) Args: seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. Reference: - [Glorot et al., 2010](http://proceedings.mlr.press/v9/glorot10a.html) c,t|ddd|y)NrarWrQr[r\r]rr+rrrr-s rrzGlorotUniform.__init__fs IID  rcDdtj|jiSrrrs rrzGlorotUniform.get_configk! %<  rrjzkeras.initializers.GlorotNormalz keras.initializers.glorot_normalc*eZdZdZdfd ZdZxZS) GlorotNormala~The Glorot normal initializer, also called Xavier normal initializer. Draws samples from a truncated normal distribution centered on 0 with `stddev = sqrt(2 / (fan_in + fan_out))` where `fan_in` is the number of input units in the weight tensor and `fan_out` is the number of output units in the weight tensor. Examples: >>> # Standalone usage: >>> initializer = GlorotNormal() >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = GlorotNormal() >>> layer = Dense(3, kernel_initializer=initializer) Args: seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. Reference: - [Glorot et al., 2010](http://proceedings.mlr.press/v9/glorot10a.html) c,t|ddd|y)NrarWrIrlrmrns rrzGlorotNormal.__init__s! +  rcDdtj|jiSrrrps rrzGlorotNormal.get_configrqrrrrrCs@rruruqs@  rruzkeras.initializers.LecunNormalzkeras.initializers.lecun_normalc*eZdZdZdfd ZdZxZS) LecunNormalaLecun normal initializer. Initializers allow you to pre-specify an initialization strategy, encoded in the Initializer object, without knowing the shape and dtype of the variable being initialized. Draws samples from a truncated normal distribution centered on 0 with `stddev = sqrt(1 / fan_in)` where `fan_in` is the number of input units in the weight tensor. Examples: >>> # Standalone usage: >>> initializer = LecunNormal() >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = LecunNormal() >>> layer = Dense(3, kernel_initializer=initializer) Args: seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. Reference: - [Klambauer et al., 2017](https://arxiv.org/abs/1706.02515) c,t|ddd|y)NrarVrIrlrmrns rrzLecunNormal.__init__ H3ED  rcDdtj|jiSrrrps rrzLecunNormal.get_configrqrrrrrCs@rryrys!F  rryzkeras.initializers.LecunUniformz keras.initializers.lecun_uniformc*eZdZdZdfd ZdZxZS) LecunUniformaLecun uniform initializer. Draws samples from a uniform distribution within `[-limit, limit]`, where `limit = sqrt(3 / fan_in)` (`fan_in` is the number of input units in the weight tensor). Examples: >>> # Standalone usage: >>> initializer = LecunUniform() >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = LecunUniform() >>> layer = Dense(3, kernel_initializer=initializer) Args: seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. Reference: - [Klambauer et al., 2017](https://arxiv.org/abs/1706.02515) c,t|ddd|y)NrarVrQrlrmrns rrzLecunUniform.__init__ H94  rcDdtj|jiSrrrps rrzLecunUniform.get_configrqrrrrrCs@rr~r~rsrr~zkeras.initializers.HeNormalzkeras.initializers.he_normalc*eZdZdZdfd ZdZxZS)HeNormalaHe normal initializer. It draws samples from a truncated normal distribution centered on 0 with `stddev = sqrt(2 / fan_in)` where `fan_in` is the number of input units in the weight tensor. Examples: >>> # Standalone usage: >>> initializer = HeNormal() >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = HeNormal() >>> layer = Dense(3, kernel_initializer=initializer) Args: seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. Reference: - [He et al., 2015](https://arxiv.org/abs/1502.01852) c,t|ddd|y)NrbrVrIrlrmrns rrzHeNormal.__init__-r{rcDdtj|jiSrrrps rrzHeNormal.get_config2rqrrrrrCs@rrr >  rrzkeras.initializers.HeUniformzkeras.initializers.he_uniformc*eZdZdZdfd ZdZxZS) HeUniformaHe uniform variance scaling initializer. Draws samples from a uniform distribution within `[-limit, limit]`, where `limit = sqrt(6 / fan_in)` (`fan_in` is the number of input units in the weight tensor). Examples: >>> # Standalone usage: >>> initializer = HeUniform() >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = HeUniform() >>> layer = Dense(3, kernel_initializer=initializer) Args: seed: A Python integer or instance of `keras.backend.SeedGenerator`. Used to make the behavior of the initializer deterministic. Note that an initializer seeded with an integer or `None` (unseeded) will produce the same random values across multiple calls. To get different random values across multiple calls, use as seed an instance of `keras.backend.SeedGenerator`. Reference: - [He et al., 2015](https://arxiv.org/abs/1502.01852) c,t|ddd|y)NrbrVrQrlrmrns rrzHeUniform.__init__YrrcDdtj|jiSrrrps rrzHeUniform.get_config^rqrrrrrCs@rrr8rrrc t|}t|dkrdx}}nPt|dk(r|dx}}n:t|dk(r |d}|d}n!d}|ddD]}||z} |d|z}|d|z}t|t|fS)zComputes the number of input and output units for a weight shape. Args: shape: Integer shape tuple. Returns: A tuple of integer scalars: `(fan_in, fan_out)`. rN)tuplelenr)r1rVrXreceptive_field_sizedims rrcrcds %LE 5zA~ Uq 8# Uqq( !": (C C '  (r11)22 v;G $$r)zkeras.initializers.Orthogonalzkeras.initializers.orthogonalz(keras.initializers.OrthogonalInitializerc6eZdZdZdfd ZddZfdZxZS) OrthogonalaInitializer that generates an orthogonal matrix. If the shape of the tensor to initialize is two-dimensional, it is initialized with an orthogonal matrix obtained from the QR decomposition of a matrix of random numbers drawn from a normal distribution. If the matrix has fewer rows than columns then the output will have orthogonal rows. Otherwise, the output will have orthogonal columns. If the shape of the tensor to initialize is more than two-dimensional, a matrix of shape `(shape[0] * ... * shape[n - 2], shape[n - 1])` is initialized, where `n` is the length of the shape vector. The matrix is subsequently reshaped to give a tensor of the desired shape. Examples: >>> # Standalone usage: >>> initializer = keras.initializers.Orthogonal() >>> values = initializer(shape=(2, 2)) >>> # Usage in a Keras layer: >>> initializer = keras.initializers.Orthogonal() >>> layer = keras.layers.Dense(3, kernel_initializer=initializer) Args: gain: Multiplicative factor to apply to the orthogonal matrix. seed: A Python integer. Used to make the behavior of the initializer deterministic. Reference: - [Saxe et al., 2014](https://openreview.net/forum?id=_wzZwKpTDF_9C) c4||_t| |yr&)gainr+r)rrrr-s rrzOrthogonal.__init__s  d#rct|dkrtd|dt|dd}|ddD]}||z} |d}t||t||f}t j ||j |}tj|\}} tj| } |tj| z}||krtj|}|jtj||zS)NrzKThe tensor to initialize must be at least two-dimensional. Received: shape=z of rank .rr)rr2)rrrdminrr3rrqrdiagsign transposerreshape) rr1r2num_rowsrnum_cols flat_shapeaqrds rr5zOrthogonal.__call__s u:>yU A7 ": C OH 9(H-s8X/FG  MM*499E Bvvay1 HHQK SXXa[ h  a Ayy3;;q%000rcHt|}d|ji}i||S)Nr)r+rrr9s rrzOrthogonal.get_configs.g(* $))$(+(((r)raNrr@rCs@rrrsB$16))rr)re keras.srcrkeras.src.api_exportrkeras.src.backendr"keras.src.initializers.initializerrkeras.src.savingrr r$rFrLrTrjruryr~rrrcrr"rrrs  -$:.% %()* /)$/)  /)d,- 2)'2)  2)j*+ /)%/)  /)d,- n)'n)  n)b*+ ( O(  ( V)* , ?,  , ^() , /,  , ^)* ( ?(  ( V,.LMN( ( O( V-/NOP( ( Q( V%8D)"D)D)r