6biSrSSKrSSKJs Jr SSKJr SSK J r SSK J r SSK Jr SSKJr SSKJr S rS rS r\"S S 5"SS\ R,55r\"SS5"SS\ R,55r\"SS/S9"SS\ R255r\"SS5"SS\ R,55rSrSrSr\"S S!/S9"S"S#\ R255r\"S$S%/S9"S&S'\ R255r SDS(jr!SES)jr"SDS*jr#\"S+S,/S9"S-S.\ R255r$\"S/S0/S9"S1S2\ R255r%SDS3jr&\"S4S5/S9"S6S7\ R255r'\"S8/S9"S9S:\ R255r(\"S;S\ R255r)\"S?S@/S9"SASB\ R255r*SDSCjr+g)Fz!Keras image preprocessing layers.N) keras_export)backend) base_layer)preprocessing_utils) image_utils)tf_utilscXUS;a[SUS35eUS;a[SUS35eg)N>wrapnearestreflectconstantzUnknown `fill_mode` zA. Only `reflect`, `wrap`, `constant` and `nearest` are supported.>r bilinearzUnknown `interpolation` z.. Only `nearest` and `bilinear` are supported.)NotImplementedError) fill_mode interpolations o/home/james-whalen/.local/lib/python3.13/site-packages/tf_keras/src/layers/preprocessing/image_preprocessing.py!check_fill_mode_and_interpolationrsWBB!"9+.6 6  33!&}o6( (  4zkeras.layers.Resizingz0keras.layers.experimental.preprocessing.ResizingcL^\rSrSrSrSU4SjjrSrSrU4SjrSr U=r $) Resizing,aA preprocessing layer which resizes images. This layer resizes an image input to a target height and width. The input should be a 4D (batched) or 3D (unbatched) tensor in `"channels_last"` format. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. This layer can be called on tf.RaggedTensor batches of input images of distinct sizes, and will resize the outputs to dense tensors of uniform size. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Args: height: Integer, the height of the output shape. width: Integer, the width of the output shape. interpolation: String, the interpolation method. Supports `"bilinear"`, `"nearest"`, `"bicubic"`, `"area"`, `"lanczos3"`, `"lanczos5"`, `"gaussian"`, `"mitchellcubic"`. Defaults to `"bilinear"`. crop_to_aspect_ratio: If True, resize the images without aspect ratio distortion. When the original aspect ratio differs from the target aspect ratio, the output image will be cropped so as to return the largest possible window in the image (of size `(height, width)`) that matches the target aspect ratio. By default (`crop_to_aspect_ratio=False`), aspect ratio may not be preserved. c >XlX lX0lX@l[R "U5Ul[TU] "S0UD6 gN) heightwidthrcrop_to_aspect_ratiorget_interpolation_interpolation_methodsuper__init__)selfrrrrkwargs __class__s rr#Resizing.__init__OsC  *$8!%0%B%B & " "6"rc^^^^TRS:Xa TRnO[Rn[ TUS9mTR TR /mTR(aUUU4Sjn[R"T5(aU[R"T5nUTRSS-n[R"XR5n[R"UTUS9nO2U"T5nO)[RRTTTR S9n[R""UTR5$)Nr dtypec>[R"T5(aUR5n[R"UTTR S9$)N)sizer)r is_ragged to_tensorr smart_resizer!)xinputsr$r,s rresize_to_aspect'Resizing.call..resize_to_aspectms?%%f-- A"//D0J0Jr)fn_output_signature)r,method)r compute_dtypetffloat32convert_inputsrrrrr- TensorShapeshape TensorSpecmap_fnimageresizer!cast) r$r1 input_dtyper2 size_as_shaper<specoutputsr,s `` @rcall Resizing.call`s    *,,K**Kk: TZZ(  $ $ !!&)) "t 4 % RS(99}}U8))$f$+62hhooT$*D*D&Gwww 2 233rc[R"U5R5nURU['UR U[ '[R"U5$Nr8r;as_listrH_AXISrW_AXISr$ input_shapes rcompute_output_shapeResizing.compute_output_shapeDnn[199; "kk F"jj F~~k**rc>URURURURS.n[TU]5n[ [UR55[UR55-5$)N)rrrr) rrrrr" get_configdictlistitemsr$config base_configr&s rrTResizing.get_configsbkkZZ!//$($=$=   g(* D**,-V\\^0DDEEr)r!rrrr)rF __name__ __module__ __qualname____firstlineno____doc__r#rFrPrT__static_attributes__ __classcell__r&s@rrr,s-F!" #"!4F+ FFrrzkeras.layers.CenterCropz2keras.layers.experimental.preprocessing.CenterCropcD^\rSrSrSrU4SjrSrSrU4SjrSr U=r $) CenterCropa=A preprocessing layer which crops images. This layers crops the central portion of the images to a target size. If an image is smaller than the target size, it will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., target_height, target_width, channels)`. If the input height/width is even and the target height/width is odd (or inversely), the input image is left-padded by 1 pixel. Args: height: Integer, the height of the output shape. width: Integer, the width of the output shape. c F>XlX l[TU] "S0UDSS0D6 g)NautocastFr)rrr"r#)r$rrr%r&s rr#CenterCrop.__init__s"   262E2rc@^^^^[TTR5m[R"T5nU[TR - mU[ TR- mUUUU4SjnUU4Sjn[R"[R"TS:TS:45X45$)Nc>[R"TS- [R5n[R"TS- [R5n[RR TXTR TR 5$)N)r8rAint32r?crop_to_bounding_boxrr)h_startw_starth_diffr1r$w_diffs r center_crop$CenterCrop.call..center_crops[ggfqj"((3Gggfqj"((3G8800$++tzz rc>[R"TTRTR/5n[R "UTR 5$rIrr/rrr8rAr7rEr1r$s rupsizeCenterCrop.call..upsize>!..djj1G777D$6$67 7rr r:r7r8r<rLrrMrcond reduce_all)r$r1rOrtryrrrss`` @@rrFCenterCrop.calls(:(:;hhv& V$t{{2V$tzz1   8ww MM6Q;! 4 5{  rc[R"U5R5nURU['UR U[ '[R"U5$rIrJrNs rrPCenterCrop.compute_output_shaperRrc>URURS.n[TU] 5n[ [ UR 55[ UR 55-5$)Nrr)rrr"rTrUrVrWrXs rrTCenterCrop.get_configsPkkZZ g(* D**,-V\\^0DDEErrr\rds@rrfrfs# <3  0+ FFrrfzkeras.layers.RandomCropz2keras.layers.experimental.preprocessing.RandomCrop)v1cL^\rSrSrSrSU4SjjrS SjrSrU4SjrSr U=r $) RandomCropaA preprocessing layer which randomly crops images during training. During training, this layer will randomly choose a location to crop images down to a target size. The layer will crop all the images in the same batch to the same cropping location. At inference time, and during training if an input image is smaller than the target size, the input will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. If you need to apply random cropping at inference time, set `training` to True when calling the layer. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., target_height, target_width, channels)`. Args: height: Integer, the height of the output shape. width: Integer, the width of the output shape. seed: Integer. Used to create a random seed. c V>[TU]"S0UDSUSS.D6 XlX lX0lg)NFT)riseedforce_generatorr)r"r#rrr)r$rrrr%r&s rr#RandomCrop.__init__ s5   $4    rcD^^^^^[TTRS9m[R"T5mT[TR - mT[ TR- mUUUUU4SjnUU4Sjn[R"[R"UTS:TS:45UU5$)Nr)c>TRnTRRS/SURU5nUSTS--nUSTS--n[R R TX#TRTR5$)Nrmr) r*_random_generatorrandom_uniformmaxr8r?rorr) r*randsrprqrrrOr1r$rss r random_crop$RandomCrop.call..random_crops%%E**99Q 5EAh&1*-GAh&1*-G8800$++tzz rc>[R"TTRTR/5n[R "UTR 5$rIrwrxs rr@RandomCrop.call..resize#r{rrr|)r$r1trainingrr@rrrOrss`` @@@rrFRandomCrop.callsd.@.@Ahhv& V$t{{2V$tzz1   8ww MM8Vq[&A+> ?    rc[R"U5R5nURU['UR U[ '[R"U5$rIrJrNs rrPRandomCrop.compute_output_shape0rRrc>URURURS.n[TU]5n[ [ UR55[ UR55-5$)N)rrr)rrrr"rTrUrVrWrXs rrTRandomCrop.get_config6sWkkZZII  g(* D**,-V\\^0DDEEr)rrrrITr\rds@rrrs$ B <+ FFrrzkeras.layers.Rescalingz1keras.layers.experimental.preprocessing.RescalingcH^\rSrSrSrSU4SjjrSrSrU4SjrSr U=r $) Rescalingi@aA preprocessing layer which rescales input values to a new range. This layer rescales every value of an input (often an image) by multiplying by `scale` and adding `offset`. For instance: 1. To rescale an input in the `[0, 255]` range to be in the `[0, 1]` range, you would pass `scale=1./255`. 2. To rescale an input in the `[0, 255]` range to be in the `[-1, 1]` range, you would pass `scale=1./127.5, offset=-1`. The rescaling is applied both during training and inference. Inputs can be of integer or floating point dtype, and by default the layer will output floats. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Input shape: Arbitrary. Output shape: Same as input. Args: scale: Float, the scale to apply to the inputs. offset: Float, the offset to apply to the inputs. c >>XlX l[TU] "S0UD6 gr)scaleoffsetr"r#)r$rrr%r&s rr#Rescaling.__init__ds   "6"rcURn[XS9n[R"URU5n[R"UR U5n[R"X5U-U-$Nr))r7r:r8rArr)r$r1r*rrs rrFRescaling.callisX""4 E*e,wwv%-66rcU$rIrrNs rrPRescaling.compute_output_shapeprc>URURS.n[TU] 5n[ [ UR 55[ UR 55-5$)N)rr)rrr"rTrUrVrWrXs rrTRescaling.get_configssPZZkk g(* D**,-V\\^0DDEEr)rr)r\rds@rrr@s# ># 7FFrr horizontalverticalhorizontal_and_verticalzkeras.layers.RandomFlipz2keras.layers.experimental.preprocessing.RandomFlipcP^\rSrSrSr\S4U4SjjrS SjrSrU4Sjr Sr U=r $) RandomFlipiaA preprocessing layer which randomly flips images during training. This layer will flip the images horizontally and or vertically based on the `mode` attribute. During inference time, the output will be identical to input. Call the layer with `training=True` to flip the input. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Args: mode: String indicating which flip mode to use. Can be `"horizontal"`, `"vertical"`, or `"horizontal_and_vertical"`. `"horizontal"` is a left-right flip and `"vertical"` is a top-bottom flip. Defaults to `"horizontal_and_vertical"` seed: Integer. Used to create a random seed. Nc >[TU]"SUSS.UD6 XlU[:XaSUlSUlOMU[ :XaSUlSUlO4U[:XaSUlSUlO[SURSU35eX l g)NTrrFzRandomFlip layer z# received an unknown mode argument r) r"r#mode HORIZONTALrrVERTICALHORIZONTAL_AND_VERTICAL ValueErrornamer)r$rrr%r&s rr#RandomFlip.__init__s CdDCFC : "DO!DM X #DO DM , ,"DO DM#DII;/ 6#  rc^^[UTR5nU4SjnU(aU"U5$U$)NcR>UnTR(asTRR5nUb[RR XS9nO8[RR UTRR55nTR(asTRR5nUb[RRXS9nO8[RRUTRR55nURUR5 U$)Nr) rrmake_seed_for_stateless_opr8r? stateless_random_flip_left_rightrandom_flip_left_rightmake_legacy_seedrstateless_random_flip_up_downrandom_flip_up_down set_shaper<)r1flipped_outputsrr$s rrandom_flipped_inputs.RandomFlip.call..random_flipped_inputss$O--HHJ#&(hh&O&O''P'O')hh&E&E'..??A'O}}--HHJ#&(hh&L&L''M'O')hh&B&B'..??A'O  % %fll 3" "rr:r7)r$r1rrs` rrFRandomFlip.calls/(:(:; #6 (0 0MrcU$rIrrNs rrPRandomFlip.compute_output_shaperrc>URURS.n[TU] 5n[ [ UR 55[ UR 55-5$)N)rr)rrr"rTrUrVrWrXs rrTRandomFlip.get_configsPIIII g(* D**,-V\\^0DDEEr)rrrrr) r]r^r_r`rarr#rFrPrTrbrcrds@rrrs+ :4$&!FFFrrzkeras.layers.RandomTranslationz9keras.layers.experimental.preprocessing.RandomTranslationcT^\rSrSrSrSU4SjjrS SjrSrU4SjrSr U=r $) RandomTranslationia; A preprocessing layer which randomly translates images during training. This layer will apply random translations to each image during training, filling empty space according to `fill_mode`. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Args: height_factor: a float represented as fraction of value, or a tuple of size 2 representing lower and upper bound for shifting vertically. A negative value means shifting image up, while a positive value means shifting image down. When represented as a single positive float, this value is used for both the upper and lower bound. For instance, `height_factor=(-0.2, 0.3)` results in an output shifted by a random amount in the range `[-20%, +30%]`. `height_factor=0.2` results in an output height shifted by a random amount in the range `[-20%, +20%]`. width_factor: a float represented as fraction of value, or a tuple of size 2 representing lower and upper bound for shifting horizontally. A negative value means shifting image left, while a positive value means shifting image right. When represented as a single positive float, this value is used for both the upper and lower bound. For instance, `width_factor=(-0.2, 0.3)` results in an output shifted left by 20%, and shifted right by 30%. `width_factor=0.2` results in an output height shifted left or right by 20%. fill_mode: Points outside the boundaries of the input are filled according to the given mode (one of `{"constant", "reflect", "wrap", "nearest"}`). - *reflect*: `(d c b a | a b c d | d c b a)` The input is extended by reflecting about the edge of the last pixel. - *constant*: `(k k k k | a b c d | k k k k)` The input is extended by filling all values beyond the edge with the same constant value k = 0. - *wrap*: `(a b c d | a b c d | a b c d)` The input is extended by wrapping around to the opposite edge. - *nearest*: `(a a a a | a b c d | d d d d)` The input is extended by the nearest pixel. interpolation: Interpolation mode. Supported values: `"nearest"`, `"bilinear"`. seed: Integer. Used to create a random seed. fill_value: a float represents the value to be filled outside the boundaries when `fill_mode="constant"`. Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. c ">[TU]"S USS.UD6 Xl[U[[ 45(aUSUlUSUlOU*UlXlURUR :a[SU35e[UR 5S:d[UR5S:a[SU35eX l [U[[ 45(aUSUl USUl OU*Ul X l URUR:a[SU35e[UR5S:d[UR5S:a[S U35e[X45 X0lX`lX@lXPlg) NTrrrzC`height_factor` cannot have upper bound less than lower bound, got ?S`height_factor` argument must have values between [-1, 1]. Received: height_factor=zB`width_factor` cannot have upper bound less than lower bound, got z5`width_factor` must have values between [-1, 1], got r)r"r# height_factor isinstancetuplerV height_lower height_upperrabs width_factor width_lower width_upperrr fill_valuerr r$rrrrrrr%r&s rr#RandomTranslation.__init__%s CdDCFC* meT] 3 3 -a 0D  -a 0D !.D  -    t00 0$$1?4  t  !C '3t/@/@+AC+G++8/;  ) lUDM 2 2+AD +AD  ,}D +    d.. .$$0>3  t 3 &#d.>.>*?#*E#n&  *)C"$* rc^^[UTR5nU4SjnU(aU"U5$U$)Nc>URnURRS:HnU(a[R"US5n[R"U5nUSn[R"U[ [R 5n[R"U[[R 5nT RRUS/T RT R[R S9nXu-nT RRUS/T RT R[R S9nX-n[R"[R"X/SS9[R S9n [U[!U 5T R"T R$T R&S9n U(a[R("U S5n U R+U5 U $)z"Translated inputs with random ops.rr)r<minvalmaxvalr*axisr))rrr)r<rankr8 expand_dimsrArLr9rMrrrrrrconcat transformget_translation_matrixrrrsqueezer) r1original_shape unbatched inputs_shape batch_sizeimg_hdimg_wdheight_translatewidth_translate translationsoutputr$s rrandom_translated_inputs8RandomTranslation.call..random_translated_inputs^s $\\N ))Q.I288F+L%aJWW\&12::>FWW\&12::>F#55DD!1o((((jj E   08 "44CC!1o''''jj DO .6O77 ?=AFjjL&|4"00..?? FFA.   ^ ,Mrr)r$r1rrs` rrFRandomTranslation.call[s0(:(:;* X +F3 3MrcU$rIrrNs rrP&RandomTranslation.compute_output_shaperrc">URURURURURUR S.n[ TU]5n[[UR55[UR55-5$N)rrrrrr rrrrrrr"rTrUrVrWrXs rrTRandomTranslation.get_configr!// --//!//II  g(* D**,-V\\^0DDEEr rrrrrrrrrrrrNrrr\rds@rrrs4 6x 4l2h F Frrc v[R"U=(d S5 [R"U5Sn[R"[R "US4[R 5[R"US4[R 5USS2SS4*[R"US4[R 5[R "US4[R 5USS2SS4*[R"US4[R 5/SS9sSSS5 $!,(df  g=f)aYReturns projective transform(s) for the given translation(s). Args: translations: A matrix of 2-element lists representing `[dx, dy]` to translate for each image (for a batch of images). name: The name of the op. Returns: A tensor of shape `(num_images, 8)` projective transforms which can be given to `transform`. translation_matrixrrNrmvaluesr)r name_scoper8r<ronesr9zeros)rrnum_translationss rrrs   D8$8 988L1!4yy)1-rzz:*A. ;aDj))*A. ;)1-rzz:aDj))*A. ;   : 9 9s DD** D8c [R"U=(d S5 UcN[R"U5SSn[R"5(d[R "U5nUbUn[R "U[RSS9nUR5RS/5(d[SU35e[R "U[RS S9n[RRUUUUUR5UR5S 9sSSS5 $!,(df  g=f) aApplies the given transform(s) to the image(s). Args: images: A tensor of shape `(num_images, num_rows, num_columns, num_channels)` (NHWC). The rank must be statically known (the shape is not `TensorShape(None)`). transforms: Projective transform matrix/matrices. A vector of length 8 or tensor of size N x 8. If one row of transforms is [a0, a1, a2, b0, b1, b2, c0, c1], then it maps the *output* point `(x, y)` to a transformed *input* point `(x', y') = ((a0 x + a1 y + a2) / k, (b0 x + b1 y + b2) / k)`, where `k = c0 x + c1 y + 1`. The transforms are *inverted* compared to the transform mapping input points to output points. Note that gradients are not backpropagated into transformation parameters. fill_mode: Points outside the boundaries of the input are filled according to the given mode (one of `{"constant", "reflect", "wrap", "nearest"}`). fill_value: a float represents the value to be filled outside the boundaries when `fill_mode="constant"`. interpolation: Interpolation mode. Supported values: `"nearest"`, `"bilinear"`. output_shape: Output dimension after the transform, `[height, width]`. If `None`, output is the same size as input image. name: The name of the op. Fill mode behavior for each valid value is as follows: - `"reflect"`: `(d c b a | a b c d | d c b a)` The input is extended by reflecting about the edge of the last pixel. - `"constant"`: `(k k k k | a b c d | k k k k)` The input is extended by filling all values beyond the edge with the same constant value k = 0. - `"wrap"`: `(a b c d | a b c d | a b c d)` The input is extended by wrapping around to the opposite edge. - `"nearest"`: `(a a a a | a b c d | d d d d)` The input is extended by the nearest pixel. Input shape: 4D tensor with shape: `(samples, height, width, channels)`, in `"channels_last"` format. Output shape: 4D tensor with shape: `(samples, height, width, channels)`, in `"channels_last"` format. Returns: Image(s) with the same type and shape as `images`, with the given transform(s) applied. Transformed coordinates outside of the input image will be filled with zeros. rNrr output_shape)rrmzaoutput_shape must be a 1-D Tensor of 2 elements: new_height, new_width, instead got output_shape=r)imagesr r transformsrr)rrr8r<executing_eagerlyget_static_valueconvert_to_tensorrn get_shapeis_compatible_withrr9raw_opsImageProjectiveTransformV3upper)r rrrrr routput_shape_values rrrsB   D/K 0  88F+Aa0L''))%'%8%8%F"%1#5L++ "(( %%'::A3?? ,~/  ))   zz44%!!oo''--/ 5 / 1 0 0s D D44 Ec ([R"U=(d S5 US- [R"U5US- -[R"U5US- -- - S- nUS- [R"U5US- -[R"U5US- --- S- n[R "U5Sn[R "[R"U5SS2S4[R"U5SS2S4*USS2S4[R"U5SS2S4[R"U5SS2S4USS2S4[R"US4[R5/SS9sSSS5 $!,(df  g=f)a`Returns projective transform(s) for the given angle(s). Args: angles: A scalar angle to rotate all images by, or (for batches of images) a vector with an angle to rotate each image in the batch. The rank must be statically known (the shape is not `TensorShape(None)`). image_height: Height of the image(s) to be transformed. image_width: Width of the image(s) to be transformed. name: The name of the op. Returns: A tensor of shape (num_images, 8). Projective transforms which can be given to operation `image_projective_transform_v2`. If one row of transforms is [a0, a1, a2, b0, b1, b2, c0, c1], then it maps the *output* point `(x, y)` to a transformed *input* point `(x', y') = ((a0 x + a1 y + a2) / k, (b0 x + b1 y + b2) / k)`, where `k = c0 x + c1 y + 1`. rotation_matrixr@rNrmr) rrr8cossinr<rr r9)angles image_height image_widthrx_offsety_offset num_angless rget_rotation_matrixr##s_,   D5$5 6 1_v+/2&&.L1$456   A v+/2&&.L1$456   XXf%a( yyvq$w'4((D!vq$w'vq$w'D!*a"**5  ! 7 6 6s EF Fzkeras.layers.RandomRotationz6keras.layers.experimental.preprocessing.RandomRotationcT^\rSrSrSrSU4SjjrS SjrSrU4SjrSr U=r $) RandomRotationiWa: A preprocessing layer which randomly rotates images during training. This layer will apply random rotations to each image, filling empty space according to `fill_mode`. By default, random rotations are only applied during training. At inference time, the layer does nothing. If you need to apply random rotations at inference time, set `training` to True when calling the layer. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format Args: factor: a float represented as fraction of 2 Pi, or a tuple of size 2 representing lower and upper bound for rotating clockwise and counter-clockwise. A positive values means rotating counter clock-wise, while a negative value means clock-wise. When represented as a single float, this value is used for both the upper and lower bound. For instance, `factor=(-0.2, 0.3)` results in an output rotation by a random amount in the range `[-20% * 2pi, 30% * 2pi]`. `factor=0.2` results in an output rotating by a random amount in the range `[-20% * 2pi, 20% * 2pi]`. fill_mode: Points outside the boundaries of the input are filled according to the given mode (one of `{"constant", "reflect", "wrap", "nearest"}`). - *reflect*: `(d c b a | a b c d | d c b a)` The input is extended by reflecting about the edge of the last pixel. - *constant*: `(k k k k | a b c d | k k k k)` The input is extended by filling all values beyond the edge with the same constant value k = 0. - *wrap*: `(a b c d | a b c d | a b c d)` The input is extended by wrapping around to the opposite edge. - *nearest*: `(a a a a | a b c d | d d d d)` The input is extended by the nearest pixel. interpolation: Interpolation mode. Supported values: `"nearest"`, `"bilinear"`. seed: Integer. Used to create a random seed. fill_value: a float represents the value to be filled outside the boundaries when `fill_mode="constant"`. c J>[TU]"SUSS.UD6 Xl[U[[ 45(aUSUlUSUlOU*UlXlURUR :a[SU35e[X#5 X l XPl X0l X@l g)NTrrrzA`factor` argument cannot have a negative value. Received: factor=r)r"r#factorrrrVlowerrrrrrrr)r$r'rrrrr%r&s rr#RandomRotation.__init__s CdDCFC fudm , ,DJDJ DJJ :: "$$*8-  *)C"$* rc^^[UTR5nU4SjnU(aU"U5$U$)Nc>URnURRS:HnU(a[R"US5n[R"U5nUSn[R"U[ [R 5n[R"U[[R 5nT RS-[R-nT RS-[R-nT RRU/XxS9n [U[XU5T R T R"T R$S9n U(a[R&"U S5n U R)U5 U $)zRotated inputs with random ops.rrrr<rrrrr)r<rr8rrArLr9rMr(nppirrrrr#rrrrr) r1rrrrrr min_angle max_anglerrr$s rrandom_rotated_inputs2RandomRotation.call..random_rotated_inputss&#\\N ))Q.I288F+L%aJWW\&12::>FWW\&12::>F S(2550I S(2550I++::!l9;F#FF;..??"00 FFA.   ^ ,Mrr)r$r1rr2s` rrFRandomRotation.calls/(:(:; < (0 0MrcU$rIrrNs rrP#RandomRotation.compute_output_shaperrc >URURURURURS.n[ TU]5n[[UR55[UR55-5$)N)r'rrrr) r'rrrrr"rTrUrVrWrXs rrTRandomRotation.get_configsgkk//!//II  g(* D**,-V\\^0DDEEr)r'rrrr(rrrrr\rds@rr%r%Ws3 9|  8$L F Frr%zkeras.layers.RandomZoomz2keras.layers.experimental.preprocessing.RandomZoomcV^\rSrSrSrSU4SjjrS SjrSrU4SjrSr U=r $) RandomZoomiac A preprocessing layer which randomly zooms images during training. This layer will randomly zoom in or out on each axis of an image independently, filling empty space according to `fill_mode`. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Args: height_factor: a float represented as fraction of value, or a tuple of size 2 representing lower and upper bound for zooming vertically. When represented as a single float, this value is used for both the upper and lower bound. A positive value means zooming out, while a negative value means zooming in. For instance, `height_factor=(0.2, 0.3)` result in an output zoomed out by a random amount in the range `[+20%, +30%]`. `height_factor=(-0.3, -0.2)` result in an output zoomed in by a random amount in the range `[+20%, +30%]`. width_factor: a float represented as fraction of value, or a tuple of size 2 representing lower and upper bound for zooming horizontally. When represented as a single float, this value is used for both the upper and lower bound. For instance, `width_factor=(0.2, 0.3)` result in an output zooming out between 20% to 30%. `width_factor=(-0.3, -0.2)` result in an output zooming in between 20% to 30%. `None` means i.e., zooming vertical and horizontal directions by preserving the aspect ratio. Defaults to `None`. fill_mode: Points outside the boundaries of the input are filled according to the given mode (one of `{"constant", "reflect", "wrap", "nearest"}`). - *reflect*: `(d c b a | a b c d | d c b a)` The input is extended by reflecting about the edge of the last pixel. - *constant*: `(k k k k | a b c d | k k k k)` The input is extended by filling all values beyond the edge with the same constant value k = 0. - *wrap*: `(a b c d | a b c d | a b c d)` The input is extended by wrapping around to the opposite edge. - *nearest*: `(a a a a | a b c d | d d d d)` The input is extended by the nearest pixel. interpolation: Interpolation mode. Supported values: `"nearest"`, `"bilinear"`. seed: Integer. Used to create a random seed. fill_value: a float represents the value to be filled outside the boundaries when `fill_mode="constant"`. Example: >>> input_img = np.random.random((32, 224, 224, 3)) >>> layer = tf.keras.layers.RandomZoom(.5, .2) >>> out_img = layer(input_img) >>> out_img.shape TensorShape([32, 224, 224, 3]) Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. c d>[TU]"S USS.UD6 Xl[U[[ 45(aUSUlUSUlOU*UlXl[UR 5S:d[UR5S:a[SU35eX l Ubl[U[[ 45(aUSUl USUl OU*Ul X l URS:dURS:a[SU35e[X45 X0lX`lX@lXPlg) NTrrrrrzP`width_factor` argument must have values larger than -1. Received: width_factor=r)r"r#rrrrVrrrrrrrrrrrrrs rr#RandomZoom.__init__7s; CdDCFC* meT] 3 3 -a 0D  -a 0D !.D  -  t  !C '3t/@/@+AC+G++8/;  )  #, 66#/? #/? $0= #/ $&$*:*:T*A ..:^= *)C"$* rc^^[UTR5nU4SjnU(aU"U5$U$)Nc>URnURRS:HnU(a[R"US5n[R"U5nUSn[R"U[ [R 5n[R"U[[R 5nT RRUS/ST R-ST R-S9nT Rb8T RRUS/ST R-ST R-S9nOUn[R"[R"X/SS9[R S9n [!U[#XU5T R$T R&T R(S9n U(a[R*"U S5n U R-U5 U $) zZoomed inputs with random ops.rrrrr,rr)r-)r<rr8rrArLr9rMrrrrrrrrrget_zoom_matrixrrrrr) r1rrrrrr height_zoom width_zoomzoomsrr$s rrandom_zoomed_inputs-RandomZoom.call..random_zoomed_inputsis#\\N ))Q.I288F+L%aJWW\&12::>FWW\&12::>F00??!1oT...T...@K   ,!33BB%q/!1!11!1!11C ) GG :3!URURURURURUR S.n[ TU]5n[[UR55[UR55-5$rrrXs rrTRandomZoom.get_configrrr)NrrNrrr\rds@rr:r:s8 FV -^/b F Frr:c 6[R"U=(d S5 [R"U5SnUS- S- SUSS2SS4- -nUS- S- SUSS2SS4- -n[R"USS2SS4[R "US4[R 5U[R "US4[R 5USS2SS4U[R "US4[R 5/SS9sSSS5 $!,(df  g=f) aReturns projective transform(s) for the given zoom(s). Args: zooms: A matrix of 2-element lists representing `[zx, zy]` to zoom for each image (for a batch of images). image_height: Height of the image(s) to be transformed. image_width: Width of the image(s) to be transformed. name: The name of the op. Returns: A tensor of shape `(num_images, 8)`. Projective transforms which can be given to operation `image_projective_transform_v2`. If one row of transforms is `[a0, a1, a2, b0, b1, b2, c0, c1]`, then it maps the *output* point `(x, y)` to a transformed *input* point `(x', y') = ((a0 x + a1 y + a2) / k, (b0 x + b1 y + b2) / k)`, where `k = c0 x + c1 y + 1`. zoom_matrixrrrNrrmr)rrr8r<rr r9)rCrrr num_zoomsr r!s rr@r@s&   D1M 2HHUOA& !3&#-#aDj8I2IJ!C'3.3q!Tz9J3JKyyaDj!)Q4)Q4aDj!)Q4   3 2 2s C D  Dzkeras.layers.RandomContrastz6keras.layers.experimental.preprocessing.RandomContrastcL^\rSrSrSrSU4SjjrS SjrSrU4SjrSr U=r $) RandomContrastiaA preprocessing layer which randomly adjusts contrast during training. This layer will randomly adjust the contrast of an image or images by a random factor. Contrast is adjusted independently for each channel of each image during training. For each channel, this layer computes the mean of the image pixels in the channel and then adjusts each component `x` of each pixel to `(x - mean) * contrast_factor + mean`. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and in integer or floating point dtype. By default, the layer will output floats. The output value will be clipped to the range `[0, 255]`, the valid range of RGB colors. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Args: factor: a positive float represented as fraction of value, or a tuple of size 2 representing lower and upper bound. When represented as a single float, lower = upper. The contrast factor will be randomly picked between `[1.0 - lower, 1.0 + upper]`. For any pixel x in the channel, the output will be `(x - mean) * factor + mean` where `mean` is the mean value of the channel. seed: Integer. Used to create a random seed. c <>[TU]"SUSS.UD6 Xl[U[[ 45(aUSUlUSUlOU=UlUlUR S:d URS:dUR S:a[SU35eX l g) NTrrrrrzX`factor` argument cannot have negative values or values greater than 1.Received: factor=r) r"r#r'rrrVr(rrr)r$r'rr%r&s rr#RandomContrast.__init__s CdDCFC fudm , ,DJDJ&, ,DJ :: tzzC/4::3C$$*8-   rc^^[UTR5nU4SjnU(aU"U5$U$)Nc>TRR5nUb;[RR USTR - STR -US9nOR[RRUSTR - STR -TRR5S9n[R"USS5nURUR5 U$)Nrrr) rrr8r?stateless_random_contrastr(rrandom_contrastr clip_by_valuerr<)r1rrr$s rrandom_contrasted_inputs5RandomContrast.call..random_contrasted_inputss))DDFD;;C$**,cDJJ.>T<11$**$$**$//@@B 2 %%fa5F   V\\ *Mrr)r$r1rrXs` rrFRandomContrast.calls/(:(:; " +F3 3MrcU$rIrrNs rrP#RandomContrast.compute_output_shape'rrc>URURS.n[TU] 5n[ [ UR 55[ UR 55-5$)N)r'r)r'rr"rTrUrVrWrXs rrTRandomContrast.get_config*sPkkII g(* D**,-V\\^0DDEEr)r'r(rrrIrr\rds@rrOrOs$ $L 2FFrrOzkeras.layers.RandomBrightnesscf^\rSrSrSrSrSrS U4SjjrSrSr Sr SS jr S r U4S jr S rU=r$)RandomBrightnessi3a A preprocessing layer which randomly adjusts brightness during training. This layer will randomly increase/reduce the brightness for the input RGB images. At inference time, the output will be identical to the input. Call the layer with `training=True` to adjust the brightness of the input. Note that different brightness adjustment factors will be apply to each the images in the batch. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Args: factor: Float or a list/tuple of 2 floats between -1.0 and 1.0. The factor is used to determine the lower bound and upper bound of the brightness adjustment. A float value will be chosen randomly between the limits. When -1.0 is chosen, the output image will be black, and when 1.0 is chosen, the image will be fully white. When only one float is provided, eg, 0.2, then -0.2 will be used for lower bound and 0.2 will be used for upper bound. value_range: Optional list/tuple of 2 floats for the lower and upper limit of the values of the input data. To make no change, use [0.0, 1.0], e.g., if the image input has been scaled before this layer. Defaults to [0.0, 255.0]. The brightness adjustment will be scaled to this range, and the output values will be clipped to this range. seed: optional integer, for fixed RNG behavior. Inputs: 3D (HWC) or 4D (NHWC) tensor, with float or int dtype. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) Output: 3D (HWC) or 4D (NHWC) tensor with brightness adjusted based on the `factor`. By default, the layer will output floats. The output value will be clipped to the range `[0, 255]`, the valid range of RGB colors, and rescaled based on the `value_range` if needed. Sample usage: ```python random_bright = tf.keras.layers.RandomBrightness(factor=0.2) # An image with shape [2, 2, 3] image = [[[1, 2, 3], [4 ,5 ,6]], [[7, 8, 9], [10, 11, 12]]] # Assume we randomly select the factor to be 0.1, then it will apply # 0.1 * 255 to all the channel output = random_bright(image, training=True) # output will be int64 with 25.5 added to each channel and round down. tf.Tensor([[[26.5, 27.5, 28.5] [29.5, 30.5, 31.5]] [[32.5, 33.5, 34.5] [35.5, 36.5, 37.5]]], shape=(2, 2, 3), dtype=int64) ``` z^The `factor` argument should be a number (or a list of two numbers) in the range [-1.0, 1.0]. z[TU]"SUSS.UD6 URU5 URU5 X0lg)NTrr)r"r# _set_factor_set_value_range_seed)r$r' value_rangerr%r&s rr#RandomBrightness.__init__ys; CdDCFC   k* rc[U[[45(d[URSU3-5e[ U5S:wa[URSU3-5e[ U5Ulg)NGot rm)rrrVr_VALUE_RANGE_VALIDATION_ERRORlensorted _value_range)r$res rrc!RandomBrightness._set_value_rangesu+t}5522tK=5II  { q 22tK=5II #;/rc[U[[45(ac[U5S:wa[ UR SU3-5eUR US5 UR US5 [U5Ulg[U[[45(a'UR U5 [U5nU*U/Ulg[ UR SU3-5e)Nrmrhrr) rrrVrjr_FACTOR_VALIDATION_ERROR_check_factor_rangerk_factorintfloatr)r$r's rrbRandomBrightness._set_factors fudm , ,6{a 11d6(OC  $ $VAY /  $ $VAY /!&>DL e - -  $ $V ,[F#GV,DLT::tF8_LM MrcRUS:dUS:a[URSU3-5eg)Nrr<rh)rro)r$ input_numbers rrp$RandomBrightness._check_factor_ranges8 # !4--$|n0EE "5rc\[XRS9nU(aURU5$U$r)r:r7_brightness_adjust)r$r1rs rrFRandomBrightness.calls+.@.@A **62 2MrcURRnUS:XaSnO  rc>URURURS.n[TU]5n[ [ UR55[ UR55-5$)N)r'rer)rqrlrdr"rTrUrVrWrXs rrTRandomBrightness.get_configsYll,,JJ  g(* D**,-V\\^0DDEEr)rqrdrl))rrTNr)r]r^r_r`rarorir#rcrbrprFryrTrbrcrds@rr`r`3sI:z % G" 0N   2FFrr`zkeras.layers.RandomHeightz4keras.layers.experimental.preprocessing.RandomHeightcL^\rSrSrSrSU4SjjrS SjrSrU4SjrSr U=r $) RandomHeightiaA preprocessing layer which randomly varies image height during training. This layer adjusts the height of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the `"channels_last"` image data format. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Args: factor: A positive float (fraction of original height), or a tuple of size 2 representing lower and upper bound for resizing vertically. When represented as a single float, this value is used for both the upper and lower bound. For instance, `factor=(0.2, 0.3)` results in an output with height changed by a random amount in the range `[20%, 30%]`. `factor=(-0.2, 0.3)` results in an output with height changed by a random amount in the range `[-20%, +30%]`. `factor=0.2` results in an output with height changed by a random amount in the range `[-20%, +20%]`. interpolation: String, the interpolation method. Supports `"bilinear"`, `"nearest"`, `"bicubic"`, `"area"`, `"lanczos3"`, `"lanczos5"`, `"gaussian"`, `"mitchellcubic"`. Defaults to `"bilinear"`. seed: Integer. Used to create a random seed. Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., random_height, width, channels)`. c >[TU]"SUSS.UD6 Xl[U[[ 45(aUSUlUSUlOU*UlXlURUR :a[SU35eUR S:dURS:a[SU35eX l [R"U5Ul X0l g) NTrrrz[`factor` argument cannot have an upper bound lesser than the lower bound. Received: factor=r<D`factor` argument must have values larger than -1. Received: factor=r)r"r#r'rrrVrrrrrr r!rr$r'rrr%r&s rr#RandomHeight.__init__s CdDCFC fudm , , &q D  &q D !'D  &    t00 0117:    t #t'8'84'?$$*8- +%0%B%B & " rcH^[U5nU4SjnU(aU"U5$U$)Nc>[R"U5n[R"U[[R5nU[ nT R R/ST R-ST R-S9n[R"XB-[R5n[R"XS/5n[RRUUT RS9n[R"UT R5nURR!5nSU['UR#U5 U$)z'Inputs height-adjusted with random ops.rr,r r,r6N)r8r<rArLr9rMrrrrrnstackr?r@r!r7rKr) r1rrrradjusted_height adjusted_sizerr r$s rrandom_height_inputs/RandomHeight.call..random_height_inputss88F+LWW\&12::>F!&)F 22AAd///d///BM !ggm&?MXX__"11%F WWVT%7%78F!<<//1L#'L   \ *Mrr:)r$r1rrs` rrFRandomHeight.calls'' 0 '/ /Mrc[R"U5R5nSU['[R"U5$rI)r8r;rKrLrNs rrP!RandomHeight.compute_output_shape32nn[199; " F~~k**rc>URURURS.n[TU]5n[ [ UR55[ UR55-5$N)r'rrr'rrr"rTrUrVrWrXs rrTRandomHeight.get_config8Ykk!//II  g(* D**,-V\\^0DDEEr)r!r'rrrrrNrr\rds@rrrs% 'R4@+ FFrrzkeras.layers.RandomWidthz3keras.layers.experimental.preprocessing.RandomWidthcL^\rSrSrSrSU4SjjrS SjrSrU4SjrSr U=r $) RandomWidthiBaA preprocessing layer which randomly varies image width during training. This layer will randomly adjusts the width of a batch of images of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the `"channels_last"` image data format. Input pixel values can be of any range (e.g. `[0., 1.)` or `[0, 255]`) and of integer or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference. For an overview and full list of preprocessing layers, see the preprocessing [guide](https://www.tensorflow.org/guide/tf_keras/preprocessing_layers). Args: factor: A positive float (fraction of original width), or a tuple of size 2 representing lower and upper bound for resizing horizontally. When represented as a single float, this value is used for both the upper and lower bound. For instance, `factor=(0.2, 0.3)` results in an output with width changed by a random amount in the range `[20%, 30%]`. `factor=(-0.2, 0.3)` results in an output with width changed by a random amount in the range `[-20%, +30%]`. `factor=0.2` results in an output with width changed by a random amount in the range `[-20%, +20%]`. interpolation: String, the interpolation method. Supports `"bilinear"`, `"nearest"`, `"bicubic"`, `"area"`, `"lanczos3"`, `"lanczos5"`, `"gaussian"`, `"mitchellcubic"`. Defaults to `bilinear`. seed: Integer. Used to create a random seed. Input shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, width, channels)`, in `"channels_last"` format. Output shape: 3D (unbatched) or 4D (batched) tensor with shape: `(..., height, random_width, channels)`. c >[TU]"SUSS.UD6 Xl[U[[ 45(aUSUlUSUlOU*UlXlURUR :a[SU35eUR S:dURS:a[SU35eX l [R"U5Ul X0l g) NTrrrzY`factor` argument cannot have an upper bound less than the lower bound. Received: factor=r<rr)r"r#r'rrrVrrrrrr r!rrs rr#RandomWidth.__init__ps CdDCFC fudm , ,%ayD %ayD  &wD %    d.. .117:    d "d&6&6&=$$*8- +%0%B%B & " rcH^[U5nU4SjnU(aU"U5$U$)Nc>[R"U5nU[n[R"U[[R 5nT R R/ST R-ST R-S9n[R"XC-[R5n[R"X%/5n[RRUUT RS9n[R"UT R5nURR!5nSU['UR#U5 U$)z&Inputs width-adjusted with random ops.rr,rN)r8r<rLrArMr9rrrrrnrr?r@r!r7rKr) r1rrrradjusted_widthrrr r$s rrandom_width_inputs-RandomWidth.call..random_width_inputss88F+L!&)FWW\&12::>F11@@d...d...AL  WW\%:BHHENHHf%=>MXX__"11%F WWVT%7%78F!<<//1L#'L   \ *Mrr)r$r1rrs` rrFRandomWidth.calls'' 0 &v. .Mrc[R"U5R5nSU['[R"U5$rI)r8r;rKrMrNs rrP RandomWidth.compute_output_shaperrc>URURURS.n[TU]5n[ [ UR55[ UR55-5$rrrXs rrTRandomWidth.get_configrr)r!r'rrrrrrr\rds@rrrBs% &P2@+ FFrrc[U[5(a[S[U5S35e[R "XS9nU$)NzhThis layer can only process a tensor representing an image or a batch of images. Received: type(inputs)=z.If you need to pass a dict containing images, labels, and bounding boxes, you should instead use the preprocessing and augmentation layers from `keras_cv.layers`. See docs at https://keras.io/api/keras_cv/layers/r))rrUrtypeutils ensure_tensor)r1r*s rr:r:sL&$ 99=fG4 4    5F MrrI)rrrNN),ranumpyr.tensorflow.compat.v2compatv2r8 tensorflow.python.util.tf_exportr tf_keras.srcrtf_keras.src.enginer!tf_keras.src.layers.preprocessingrrtf_keras.src.utilsrrrLrMrLayerrrfBaseRandomLayerrrrrrrrrrr#r%r:r@rOr`rrr:rrrrs(!!9 *J*'    ObFzbFbFJ8HF!!HF HFV8  UF++UF  UFp75F   5F 5Fp 38  ]F++]F  ]FB$?  pF 22pF  pFf J _ D1 h!<  JFZ//JF  JFZ8  vF++vF  vFr( V!<  YFZ//YF  YFx-"5SFz11SF6SFl:  pF:--pF  pFf9  nF*,,nF  nFb r