2VhxddlZddlmZddlmZddlmZddlmZddlm Z ddl m Z ddl m Z d Zed zZed zZed zZed zZ ddZeddg d dZeddg d!dZ d"dZd#dZd#dZd#dZedd#dZedd$dZej8j;dej<ej>e_ ej6j@e_ y)%N)backend)layers) keras_export)imagenet_utils) Functional)operation_utils) file_utilszDhttps://storage.googleapis.com/tensorflow/keras-applications/nasnet/zNASNet-mobile.h5zNASNet-mobile-no-top.h5zNASNet-large.h5zNASNet-large-no-top.h5c tjdk(r td|dvs tj|s td|dk(r|r| dk7r tdt |t r#d|vr|dk(rtd t|zd z| d } tj|| d tj|| }tjdk7rtjdd|tj|}n/tj|stj||}n|}|d|dzzzdk7rtd|tjdk(rdnd}|dz}tj|dddddd|}tj |d d!d"#|}d}t#||||dzzd$%\}}t#||||zd&%\}}t%|D]}t'||||%\}}t#||||zd'|%\}}|s|n|}t%|D]}t'||||z||zdz%\}} t#||||dzzd'd|z%\}}|s|n|}t%|D]$}t'||||dzzd|z|zdz%\}}&tj(d(|}|rOtj*|}tj,| |tj.| | d)*|}n?| d+k(rtj*|}n| d,k(rtj0|}|t3j4|}n|}t7||| -}|dk(r| d.k(rP|rtj8d/t:d0d12}ntj8d3t<d0d42}|j?||S| d k(rP|rtj8d5t@d0d62}ntj8d7tBd0d82}|j?||Std9||j?||S):aInstantiates a NASNet model. Reference: - [Learning Transferable Architectures for Scalable Image Recognition]( https://arxiv.org/abs/1707.07012) (CVPR 2018) For image classification use cases, see [this page for detailed examples]( https://keras.io/api/applications/#usage-examples-for-image-classification-models). For transfer learning use cases, make sure to read the [guide to transfer learning & fine-tuning]( https://keras.io/guides/transfer_learning/). Note: each Keras Application expects a specific kind of input preprocessing. For NasNet, call `keras.applications.nasnet.preprocess_input` on your inputs before passing them to the model. `nasnet.preprocess_input` will scale input pixels between -1 and 1. Args: input_shape: Optional shape tuple, the input shape is by default `(331, 331, 3)` for NASNetLarge and `(224, 224, 3)` for NASNetMobile. It should have exactly 3 input channels, and width and height should be no smaller than 32. E.g. `(224, 224, 3)` would be one valid value. penultimate_filters: Number of filters in the penultimate layer. NASNet models use the notation `NASNet (N @ P)`, where: - N is the number of blocks - P is the number of penultimate filters num_blocks: Number of repeated blocks of the NASNet model. NASNet models use the notation `NASNet (N @ P)`, where: - N is the number of blocks - P is the number of penultimate filters stem_block_filters: Number of filters in the initial stem block skip_reduction: Whether to skip the reduction step at the tail end of the network. filter_multiplier: Controls the width of the network. - If `filter_multiplier` < 1.0, proportionally decreases the number of filters in each layer. - If `filter_multiplier` > 1.0, proportionally increases the number of filters in each layer. - If `filter_multiplier` = 1, default number of filters from the paper are used at each layer. include_top: Whether to include the fully-connected layer at the top of the network. weights: `None` (random initialization) or `imagenet` (ImageNet weights) input_tensor: Optional Keras tensor (i.e. output of `layers.Input()`) to use as image input for the model. pooling: Optional pooling mode for feature extraction when `include_top` is `False`. - `None` means that the output of the model will be the 4D tensor output of the last convolutional block. - `avg` means that global average pooling will be applied to the output of the last convolutional block, and thus the output of the model will be a 2D tensor. - `max` means that global max pooling will be applied. classes: Optional number of classes to classify images into, only to be specified if `include_top` is `True`, and if no `weights` argument is specified. default_size: Specifies the default image size of the model classifier_activation: A `str` or callable. The activation function to use on the "top" layer. Ignored unless `include_top=True`. Set `classifier_activation=None` to return the logits of the "top" layer. When loading pretrained weights, `classifier_activation` can only be `None` or `"softmax"`. name: The name of the model (string). Returns: A model instance. channels_firstzNASNet does not support the `channels_first` image data format. Switch to `channels_last` by editing your local config file at ~/.keras/keras.json>NimagenetzThe `weights` argument should be either `None` (random initialization), `imagenet` (pre-training on ImageNet), or the path to the weights file to be loaded.r zWIf using `weights` as `"imagenet"` with `include_top` as true, `classes` should be 1000NzWhen specifying the input shape of a NASNet and loading `ImageNet` weights, the input_shape argument must be static (no None entries). Got: `input_shape=z`.K ) default_sizemin_size data_formatrequire_flattenweights channels_lastaThe NASNet family of models is only available for the input data format "channels_last" (width, height, channels). However your settings specify the default data format "channels_first" (channels, width, height). You should set `image_data_format="channels_last"` in your Keras config located at ~/.keras/keras.json. The model being returned right now will expect inputs to follow the "channels_last" data format.) stacklevel)shape)tensorrrzuFor NASNet-A models, the `penultimate_filters` must be a multiple of 24 * (`filter_multiplier` ** 2). Current value: rrrvalidF stem_conv1 he_normal)stridespaddinguse_biasnamekernel_initializerZڊ?MbP?stem_bn1axismomentumepsilonr&stem_1block_idstem_2reduce_relu predictions) activationr&avgmaxr&znasnet_mobile.h5models 020fb642bf7360b370c678b08e0adf61) cache_subdir file_hashznasnet_mobile_no_top.h5 1ed92395b5b598bdda52abe5c0dbfd63znasnet_large.h5 11577c9a518f0070763c2b964a382f17znasnet_large_no_top.h5 d81d89dc07e6e56530c4e77faddd61b5zDImageNet weights can only be loaded with NASNetLarge or NASNetMobile)"rimage_data_format ValueErrorr exists isinstancetuplestrrobtain_input_shapewarningswarnrInputis_keras_tensorConv2DBatchNormalization_reduction_a_cellrange_normal_a_cell ActivationGlobalAveragePooling2Dvalidate_activationDenseGlobalMaxPooling2Drget_source_inputsrget_fileNASNET_MOBILE_WEIGHT_PATH NASNET_MOBILE_WEIGHT_PATH_NO_TOP load_weightsNASNET_LARGE_WEIGHT_PATHNASNET_LARGE_WEIGHT_PATH_NO_TOP) input_shapepenultimate_filters num_blocksstem_block_filtersskip_reductionfilter_multiplier include_topr input_tensorpoolingclassesrclassifier_activationr& img_input channel_dimfiltersxpip0inputsmodel weights_paths M/home/dcms/DCMS/lib/python3.12/site-packages/keras/src/applications/nasnet.pyNASNetrts8|  "&66 1  ) )Z->->w-G <  *D 0  ;& K  z ! 47:+6F GJN N   !33!--/# K  "o5  9 LL{3 &&|4 L LI$Ib$5q$89:a? 12 4  0026FF!BK!R'G   &  A !! 64j  A A  1g+Q./( DAq  1g**X DAq: >aG=1>  1g))gj\4J EAr!aA :   ' '"Q*+  1    #Q&&1z>*+  EAr!aA :   '* *J*Q./  1  "&!!$A +F ) ) +A .**+@'J FLL  5M   e /--/2A  +))+A.A 22<@ vqt ,E* 3 )22&-!)@    *22-4!)@     | ,2 L1S )22%,!)@    *22,3!)@     | , L#    7# Lz&keras.applications.nasnet.NASNetMobilezkeras.applications.NASNetMobilectjdk(r td|s|d}t|ddddd|||||d || S) a Instantiates a Mobile NASNet model in ImageNet mode. Reference: - [Learning Transferable Architectures for Scalable Image Recognition]( https://arxiv.org/abs/1707.07012) (CVPR 2018) Optionally loads weights pre-trained on ImageNet. Note that the data format convention used by the model is the one specified in your Keras config at `~/.keras/keras.json`. Note: each Keras Application expects a specific kind of input preprocessing. For NASNet, call `keras.applications.nasnet.preprocess_input` on your inputs before passing them to the model. Args: input_shape: Optional shape tuple, only to be specified if `include_top` is False (otherwise the input shape has to be `(224, 224, 3)` for NASNetMobile It should have exactly 3 inputs channels, and width and height should be no smaller than 32. E.g. `(224, 224, 3)` would be one valid value. include_top: Whether to include the fully-connected layer at the top of the network. weights: `None` (random initialization) or `imagenet` (ImageNet weights). For loading `imagenet` weights, `input_shape` should be (224, 224, 3) input_tensor: Optional Keras tensor (i.e. output of `layers.Input()`) to use as image input for the model. pooling: Optional pooling mode for feature extraction when `include_top` is `False`. - `None` means that the output of the model will be the 4D tensor output of the last convolutional layer. - `avg` means that global average pooling will be applied to the output of the last convolutional layer, and thus the output of the model will be a 2D tensor. - `max` means that global max pooling will be applied. classes: Optional number of classes to classify images into, only to be specified if `include_top` is `True`, and if no `weights` argument is specified. classifier_activation: A `str` or callable. The activation function to use on the "top" layer. Ignored unless `include_top=True`. Set `classifier_activation=None` to return the logits of the "top" layer. When loading pretrained weights, `classifier_activation` can only be `None` or `"softmax"`. name: The name of the model (string). Returns: A Keras model instance. torchz}NASNetMobile is not available with the torch backend at this time due to an outstanding bug. If interested, please open a PR.)r:r:ri rFrr: r_r`rarbrcrdrrerfrgrrhr&)rrCrtr^rdrrerfrgrhr&s rs NASNetMobiler{;slLG# /  ;.#  !3  ruz%keras.applications.nasnet.NASNetLargezkeras.applications.NASNetLargec4t|ddddd|||||d||S)a Instantiates a NASNet model in ImageNet mode. Reference: - [Learning Transferable Architectures for Scalable Image Recognition]( https://arxiv.org/abs/1707.07012) (CVPR 2018) Optionally loads weights pre-trained on ImageNet. Note that the data format convention used by the model is the one specified in your Keras config at `~/.keras/keras.json`. Note: each Keras Application expects a specific kind of input preprocessing. For NASNet, call `keras.applications.nasnet.preprocess_input` on your inputs before passing them to the model. Args: input_shape: Optional shape tuple, only to be specified if `include_top` is False (otherwise the input shape has to be `(331, 331, 3)` for NASNetLarge. It should have exactly 3 inputs channels, and width and height should be no smaller than 32. E.g. `(224, 224, 3)` would be one valid value. include_top: Whether to include the fully-connected layer at the top of the network. weights: `None` (random initialization) or `imagenet` (ImageNet weights). For loading `imagenet` weights, `input_shape` should be (331, 331, 3) input_tensor: Optional Keras tensor (i.e. output of `layers.Input()`) to use as image input for the model. pooling: Optional pooling mode for feature extraction when `include_top` is `False`. - `None` means that the output of the model will be the 4D tensor output of the last convolutional layer. - `avg` means that global average pooling will be applied to the output of the last convolutional layer, and thus the output of the model will be a 2D tensor. - `max` means that global max pooling will be applied. classes: Optional number of classes to classify images into, only to be specified if `include_top` is `True`, and if no `weights` argument is specified. classifier_activation: A `str` or callable. The activation function to use on the "top" layer. Ignored unless `include_top=True`. Set `classifier_activation=None` to return the logits of the "top" layer. When loading pretrained weights, `classifier_activation` can only be `None` or `"softmax"`. name: The name of the model (string). Returns: A Keras model instance. `Trrry)rtrzs rs NASNetLargers<L  !3  ruc tjdk(rdnd}tjd|5tjd|}|dk(r7tj t j||d||}d }nd }tj|||d ||d |}tj|ddd||}tjd|}tj||d|d d |}tj|ddd||}ddd|S#1swYSxYw)aFAdds 2 blocks of [relu-separable conv-batchnorm]. Args: ip: Input tensor filters: Number of output filters per layer kernel_size: Kernel size of separable convolutions strides: Strided convolution for downsampling block_id: String block_id Returns: A Keras tensor r rrseparable_conv_block_r4rseparable_conv_1_pad_r$r&r sameseparable_conv_1_F)r#r&r$r%r(r)separable_conv_1_bn_r+separable_conv_2_)r&r$r%separable_conv_2_bn_N) rrB name_scoperrR ZeroPadding2Dr correct_padSeparableConv2DrN)iprk kernel_sizer#r1rjrlconv_pads rs_separable_conv_blockrs0026FF!BK   3H:> ?% %F  f %b ) f $$&221kB,XJ7AHH F " "  $XJ/     F % %'z2    &F  f %a ( F " "  $XJ/    F % %'z2   A% L HM% L Hs DD>>Ec tjdk(rdnd}tjdk(rdnd}tjd5||}n|j||j|k7rGtjd|5t j d d | |}t j d d dd||}t j|dzd ddd|d|}t jd|}t jd|}t j d d dd||}t j|dzd ddd|d|}t j||g|}t j|ddd| |}dddn|j||k7rtjd!|5t j d |}t j|d d dd"|dd#|}t j|ddd| |}dddddd|S#1swYxYw#1swYxYw#1swY|SxYw)$aAdjusts the input `previous path` to match the shape of the `input`. Used in situations where the output number of filters needs to be changed. Args: p: Input tensor which needs to be modified ip: Input tensor whose shape needs to be matched filters: Number of output filters to be matched block_id: String block_id Returns: Adjusted Keras tensor r rrr adjust_blockNadjust_reduction_block_r4adjust_relu_1_r9rrrr adjust_avg_pool_1_r#r$r&rFadjust_conv_1_r")r$r%r&r')rrr)r$)rrr)croppingadjust_avg_pool_2_adjust_conv_2_)r,r(r) adjust_bn_r+adjust_projection_block_adjust_conv_projection_r#r$r&r%r') rrBrrrrRAveragePooling2DrMr Cropping2D concatenaterN)rmrrkr1rjimg_dimp1p2s rs _adjust_blockr-s0026FF!BK,,.2BBaG   N +B 9A WWW '!2 2##&=hZ$HI* OF%%f^H:3NOV,,"#-hZ8    V]]qL"")(4'2 DV))2BCAFAV&&0@A"EV,,"#-hZ8    V]]qL"")(4'2 &&BxkBF--$# %hZ0   K* * XWW[ !W ,##&>xj$IJ -F%%f-a0FMM""28*="'2F--$# %hZ0    eBF H}* * Z  eBF Hs> ?I1 D&I13I1$A#I%I1I" I1%I. *I11I;c tjdk(rdnd}tjd|5t||||}t j d|}t j |dddd|d d |}t j|d d d||}tjd5t||dd|}t||d|}t j||gd|}dddtjd5t||dd|} t||dd|} t j| | gd|} dddtjd5t jdddd| |} t j| |gd!|} dddtjd"5t jdddd#| |} t jdddd$| |}t j| |gd%|}dddtjd&5t||d'|}t j||gd(|}dddt j|  g|d)|*}ddd||fS#1swYxYw#1swYixYw#1swYxYw#1swYxYw#1swYixYw#1swY|fSxYw)+zAdds a Normal cell for NASNet-A (Fig. 4 in the paper). Args: ip: Input tensor `x` p: Input tensor `p` filters: Number of output filters block_id: String block_id Returns: A Keras tensor r rrnormal_A_block_r4rrnormal_conv_1_Fr"rr(r) normal_bn_1_r+block_1r normal_left1_)rr1normal_right1_r0 normal_add_1_r9Nblock_2 normal_left2_rnormal_right2_ normal_add_2_block_3 normal_left3_r normal_add_3_block_4 normal_left4_normal_right4_ normal_add_4_block_5 normal_left5_ normal_add_5_normal_concat_r,r&) rrBrrrrRrMrNraddrr)rrmrkr1rjhx1_1x1_2x1x2_1x2_2x2x3x4_1x4_2x4x5rls rsrQrQs0026FF!BK   ohZ8 9K !R( 3 %F  f %b ) FMM  !(,*    F % %z*      * K("( 3 D )7~hZ%@DT4Lxj/IJB K   * K(7F}XJ-GD)7F~hZ-HDT4Lxj/IJB K   * F(($XJ/   B RGM(*DEB F   * K6**$XJ/   D 6**%hZ0   D T4Lxj/IJB K   * F&7}XJ%?BRGM(*DEB  F    BB #!(, OK X b5Lq K K K K F F K K F FCK X b5LsBK1:AJ2:K1AJ?K15?K 4K1A!K2K1.K%=)K12J< 7K1?K K1 K K1K" K1%K. *K11K=c tjdk(rdnd}tjd|5t||||}t j d|}t j |dddd|d d |}t j|d d d||}t jtj|dd||}tjd5t||ddd|}t||ddd|}t j||gd|} dddtjd5t jdddd |!|} t||ddd"|} t j| | gd#|} dddtjd$5t jdddd%|!|} t||ddd&|}t j| |gd'|}dddtjd(5t jdddd)|! }t j |g}dddtjd*5t |dd)|+}t jdddd,|!|}t j||gd-|}dddt j g|d.|/}||fcdddS#1swYxYw#1swYwxYw#1swYxYw#1swYxYw#1swYhxYw#1swYyxYw)0zAdds a Reduction cell for NASNet-A (Fig. 4 in the paper). Args: ip: Input tensor `x` p: Input tensor `p` filters: Number of output filters block_id: String block_id Returns: A Keras tensor r rrreduction_A_block_r4rrreduction_conv_1_Fr"rr(r)reduction_bn_1_r+rreduction_pad_1_rrrrreduction_left1_)r#r1)rreduction_right1_reduction_add_1_r9Nrrr reduction_left2_rreduction_right2_reduction_add_2_rreduction_left3_reduction_right3_reduction_add3_rreduction_left4_rr0reduction_right5_reduction_add4_reduction_concat_r)rrBrrrrRrMrNrrrrr MaxPooling2Drr)rrmrkr1rjrh3rrrrrrx3_1x3_2rrx5_1x5_2rrls rsrOrOs0026FF!BK   0 ; <c !R( 3 %F  f %b ) FMM  $XJ/*    F % %"8*-     V ! !"..q!4#H:.       * N(+H:6 D),XJ7 DT4L1A(/LMB N"   * N6&&'z2   D ),XJ7 DT4L1A(/LMB N    * M6**'z2   D ),XJ7 DT4L /KLB M    * &(('z2   B RH%B &   * M(GV0@ .KD6&&( 3   D T4L /KLB M    R $XJ/  "uGcc0 N N" N N  M M  & & M MeccsB:M.AL1MAL M=AL"M,:L/&MAL;*ML ML M"L, 'M/L8 4M;M MMz*keras.applications.nasnet.preprocess_inputc2tj||dS)Ntf)rmode)rpreprocess_input)rlrs rsrrUs  * * { ruz,keras.applications.nasnet.decode_predictionsc0tj||S)N)top)rdecode_predictions)predsrs rsrr\s  , ,U <rs-1')&K.0BB#47P#P ,/@@"36N"N   # dN 0)   # W  Wt/(   # O  Of?C7 tT nZzqh:;< <==>=*>>EE 22  3 3F ,>>FFru