~WhCdZddlmcmZddlmZddlmZddl m Z ddl m Z ddl mZddl mZdd lmZd Zed zZed zZed zZedzZedzZedzZe ZdZdZdZ d+dZedd d+dZedd d+dZedd  d+d!Z ed"d,d#Z!ed$d-d&Z"ej#$d'ej%ej&(e!_ej"je"_d)Z'e(ed*eje'ze(ed*eje'ze(e d*e je'zdS).zDenseNet models for Keras. Reference: - [Densely Connected Convolutional Networks]( https://arxiv.org/abs/1608.06993) (CVPR 2017) N)backend)imagenet_utils)training)VersionAwareLayers) data_utils) layer_utils) keras_exportzFhttps://storage.googleapis.com/tensorflow/keras-applications/densenet/1densenet121_weights_tf_dim_ordering_tf_kernels.h57densenet121_weights_tf_dim_ordering_tf_kernels_notop.h51densenet169_weights_tf_dim_ordering_tf_kernels.h57densenet169_weights_tf_dim_ordering_tf_kernels_notop.h51densenet201_weights_tf_dim_ordering_tf_kernels.h57densenet201_weights_tf_dim_ordering_tf_kernels_notop.h5c zt|D]*}t|d|dzt|dzz}+|S)zA dense block. Args: x: input tensor. blocks: integer, the number of building blocks. name: string, block label. Returns: Output tensor for the block. _blockname)range conv_blockstr)xblocksris a/var/www/html/movieo_spanner_bot/venv/lib/python3.11/site-packages/keras/applications/densenet.py dense_blockr?sI6]]AA q"4(?SQZZ#? @ @ @ Hctjdkrdnd}t|d|dz|}td|dz |}tt tj|||zdd |d z |}td d |dz|}|S)zA transition block. Args: x: input tensor. reduction: float, compression rate at transition layers. name: string, block label. Returns: output tensor for the block. channels_lastr>_bnaxisepsilonrrelu_relurF_convuse_biasr_poolstridesr) rimage_data_formatlayersBatchNormalization ActivationConv2Dint int_shapeAveragePooling2D)r reductionrbn_axiss rtransition_blockr:Os,../AAaaqG!! hTE\ "    A &tg~66q99A  G a  )I 566  G^       A 14'>BB1EEA Hrc vtjdkrdnd}t|d|dz|}td|dz |}td |zdd |d z |}t|d|dz|}td|dz |}t|ddd |dz|}t||dz||g}|S)zA building block for a dense block. Args: x: input tensor. growth_rate: float, growth rate at dense layers. name: string, block label. Returns: Output tensor for the block. r r!rr"_0_bnr$r'_0_relurF_1_convr*_1_bn_1_relusame_2_conv)paddingr+r_concat)r%r)rr0r1r2r3r4 Concatenate)r growth_raterr9x1s rrrisp,../AAaaqG  " " hTG^ #    B   6y(8  9 9" = =B  KU 1A     B  " " hTG^ #    B   6y(8  9 9" = =B QTI=M     B dY.>??BHHA HrTimagenetsoftmaxc |dvs3tjj|st d|dkr|r|dkrt dt j|ddtj||}|t | }n3tj |st || }n|}tjd krd nd} t d|} t ddddd| } t| dd| } tdd| } t d| } td dd| } t#| |d d!} t%| d"d#} t#| |dd$} t%| d"d%} t#| |dd&} t%| d"d'} t#| |d d(} t| dd)| } tdd| } |r^td*| } t j||t||d+,| } nS|d-kr$td*| } n)|d.kr#td/| } |t/j|} n|} |gd0krt3j| | d1} nW|gd2krt3j| | d3} n7|gd4krt3j| | d5} nt3j| | d6} |dkr|rr|gd0krt7jd7t:d8d9:} n|gd2krt7jd;t<d8d<:} n|gd4krt7jd=t>d8d>:} nq|gd0krt7jd?t@d8d@:} nK|gd2krt7jdAtBd8dB:} n%|gd4krt7jdCtDd8dD:} | #| n|| #|| S)Eaz Instantiates the DenseNet architecture. Reference: - [Densely Connected Convolutional Networks]( https://arxiv.org/abs/1608.06993) (CVPR 2017) This function returns a Keras image classification model, optionally loaded with weights pre-trained on ImageNet. 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 DenseNet, call `tf.keras.applications.densenet.preprocess_input` on your inputs before passing them to the model. `densenet.preprocess_input` will scale pixels between 0 and 1 and then will normalize each channel with respect to the ImageNet dataset statistics. Args: blocks: numbers of building blocks for the four dense layers. include_top: whether to include the fully-connected layer at the top of the network. weights: one of `None` (random initialization), 'imagenet' (pre-training on ImageNet), or the path to the weights file to be loaded. input_tensor: optional Keras tensor (i.e. output of `layers.Input()`) to use as image input for the model. input_shape: optional shape tuple, only to be specified if `include_top` is False (otherwise the input shape has to be `(224, 224, 3)` (with `'channels_last'` data format) or `(3, 224, 224)` (with `'channels_first'` data format). It should have exactly 3 inputs channels, and width and height should be no smaller than 32. E.g. `(200, 200, 3)` would be one valid value. 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. 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"`. Returns: A `keras.Model` instance. >NrIzThe `weights` argument should be either `None` (random initialization), `imagenet` (pre-training on ImageNet), or the path to the weights file to be loaded.rIrJzWIf using `weights` as `"imagenet"` with `include_top` as true, `classes` should be 1000r) default_sizemin_size data_formatrequire_flattenweightsN)shape)tensorrSr r!r)r!r!rU)rD@r,Fz conv1/conv)r/r+rr"zconv1/bnr$r'z conv1/relur)rrrXpool1r.rconv2g?pool2conv3pool3conv4pool4conv5bnavg_pool predictions) activationravgmaxmax_pool  densenet121rirjrr densenet169rirj0r densenet201densenetr models 9d60b8095a5708f2dcce2bca79d332c7) cache_subdir file_hashr d699b8f76981ab1b30698df4c175e90br 1ceb130c1ea1b78c3bf6114dbdfd8807r 30ee3e1110167f948a6b9946edeeb738r b8c4d4c20dd625c148057b9ff1c1176br c13680b51ded0fb44dff2d8f86ac8bb1)$tfiogfileexists ValueErrorrobtain_input_shaperr0r1Inputis_keras_tensor ZeroPadding2Dr4r2r3 MaxPooling2Drr:GlobalAveragePooling2Dvalidate_activationDenseGlobalMaxPooling2Drget_source_inputsrModelrget_fileDENSENET121_WEIGHT_PATHDENSENET169_WEIGHT_PATHDENSENET201_WEIGHT_PATHDENSENET121_WEIGHT_PATH_NO_TOPDENSENET169_WEIGHT_PATH_NO_TOPDENSENET201_WEIGHT_PATH_NO_TOP load_weights)r include_toprR input_tensor input_shapepoolingclassesclassifier_activation img_inputr9rinputsmodel weights_paths rDenseNetrsR ) ) )RU[-?-?-H-H ) <   *D 1   !3-//# KLL{L33 &|44 % L LLII$I,../AAaaqG%566yAAA b!Q\ JJ1MMA!! hZ "    A &|44Q77A%566q99AAqw77::AAvayw///ACg...AAvayw///ACg...AAvayw///ACg...AAvayw///A!!wt!LLQOOA&v..q11A >  ) )z ) : :1 = =*+@'JJJ LL  5M      e  --:->>qAAAA   ))z)::1==A.|<<  vq}=== ??? " "vq}=== ??? " "vq}===vqz:::* + (()2G+!)@     ???**)2G+!)@     ???**)2G+!)@     (()2M2!)@     ???**)2M2!)@     ???**)2M2!)@     <((((   7### Lrz'keras.applications.densenet.DenseNet121zkeras.applications.DenseNet121c 2tgd|||||||S)z*Instantiates the Densenet121 architecture.rhrrrRrrrrrs r DenseNet121rY2    rz'keras.applications.densenet.DenseNet169zkeras.applications.DenseNet169c 2tgd|||||||S)z*Instantiates the Densenet169 architecture.rnrrs r DenseNet169rrrrz'keras.applications.densenet.DenseNet201zkeras.applications.DenseNet201c 2tgd|||||||S)z*Instantiates the Densenet201 architecture.rprrs r DenseNet201rrrz,keras.applications.densenet.preprocess_inputc0tj||dS)Ntorch)rPmode)rpreprocess_input)rrPs rrrs#  * {   rz.keras.applications.densenet.decode_predictionsc.tj||S)N)top)rdecode_predictions)predsrs rrrs  ,U < < <rss "!!!!!!!!------!!!!!!++++++""""""######:99999MKK?@ KK?@ KK?@           4   @   #OOOOd-/O   #,-/O   #,-/O   #,<==>= >??===@?=*>EE 5  3F ,>F3j Y 3c 9::: Y 3c 9::: Y 3c 9:::::r