σ ΎχXc@@sfdZddlmZddlmZddlZddlmZddlmZddlm Z dd lm Z dd lm Z dd lm Z dd lm Z dd lmZddlmZddlmZddlmZddlmZddlmZddlmZddlmZddlmZdZdZedddddd„Zd„Z dS(sZXception V1 model for Keras. On ImageNet, this model gets to a top-1 validation accuracy of 0.790 and a top-5 validation accuracy of 0.945. Do note that the input image format for this model is different than for the VGG16 and ResNet models (299x299 instead of 224x224), and that the input preprocessing function is also different (same as Inception V3). Also do note that this model is only available for the TensorFlow backend, due to its reliance on `SeparableConvolution` layers. # Reference - [Xception: Deep Learning with Depthwise Separable Convolutions](https://arxiv.org/abs/1610.02357) i(tprint_function(tabsolute_importNi(tModel(tlayers(tDense(tInput(tBatchNormalization(t Activation(tConv2D(tSeparableConv2D(t MaxPooling2D(tGlobalAveragePooling2D(tGlobalMaxPooling2D(tget_source_inputs(tget_file(tbackendi(tdecode_predictions(t_obtain_input_shapesvhttps://github.com/fchollet/deep-learning-models/releases/download/v0.4/xception_weights_tf_dim_ordering_tf_kernels.h5s|https://github.com/fchollet/deep-learning-models/releases/download/v0.4/xception_weights_tf_dim_ordering_tf_kernels_notop.h5timagenetiθc C@s* |ddahkr!tdƒ‚n|dkrN|rN|dkrNtdƒ‚ntjƒdkrotdƒ‚ntjƒdkr€tjdƒtjdƒd }nda}t |d d d d dtjƒd|ƒ}|dakrςt d|ƒ}n-tj |ƒst d|d|ƒ}n|}t ddbddcdt ddƒ|ƒ}tddƒ|ƒ}tdddƒ|ƒ}t ddddt ddƒ|ƒ}tddƒ|ƒ}tdddƒ|ƒ}t d deddfd"d#dt ƒ|ƒ} tƒ| ƒ} td dgd"d#dt dd$ƒ|ƒ}tdd%ƒ|ƒ}tddd&ƒ|ƒ}td dhd"d#dt dd'ƒ|ƒ}tdd(ƒ|ƒ}tdiddjd"d#dd)ƒ|ƒ}tj|| gƒ}t d*dkddld"d#dt ƒ|ƒ} tƒ| ƒ} tddd+ƒ|ƒ}td*dmd"d#dt dd,ƒ|ƒ}tdd-ƒ|ƒ}tddd.ƒ|ƒ}td*dnd"d#dt dd/ƒ|ƒ}tdd0ƒ|ƒ}tdoddpd"d#dd1ƒ|ƒ}tj|| gƒ}t d2dqddrd"d#dt ƒ|ƒ} tƒ| ƒ} tddd3ƒ|ƒ}td2dsd"d#dt dd4ƒ|ƒ}tdd5ƒ|ƒ}tddd6ƒ|ƒ}td2dtd"d#dt dd7ƒ|ƒ}tdd8ƒ|ƒ}tduddvd"d#dd9ƒ|ƒ}tj|| gƒ}xctd:ƒD]U} |} d;t| d<ƒ} tdd| d=ƒ|ƒ}td2dwd"d#dt d| d>ƒ|ƒ}td| d?ƒ|ƒ}tdd| d@ƒ|ƒ}td2dxd"d#dt d| dAƒ|ƒ}td| dBƒ|ƒ}tdd| dCƒ|ƒ}td2dyd"d#dt d| dDƒ|ƒ}td| dEƒ|ƒ}tj|| gƒ}qϋWt dFdzdd{d"d#dt ƒ|ƒ} tƒ| ƒ} tdddGƒ|ƒ}td2d|d"d#dt ddHƒ|ƒ}tddIƒ|ƒ}tdddJƒ|ƒ}tdFd}d"d#dt ddKƒ|ƒ}tddLƒ|ƒ}td~ddd"d#ddMƒ|ƒ}tj|| gƒ}tdNd€d"d#dt ddOƒ|ƒ}tddPƒ|ƒ}tdddQƒ|ƒ}tdRdd"d#dt ddSƒ|ƒ}tddTƒ|ƒ}tdddUƒ|ƒ}|rOtddVƒ|ƒ}t|dWdXddYƒ|ƒ}n<|dZkrmtƒ|ƒ}n|d[kr‹tƒ|ƒ}n|dak r¦t|ƒ} n|} t| |dd\ƒ} |dkr |rλtd]td^d_ƒ}ntd`td^d_ƒ}| j|ƒn|r& tj|ƒn| S(‚swInstantiates the Xception architecture. Optionally loads weights pre-trained on ImageNet. This model is available for TensorFlow only, and can only be used with inputs following the TensorFlow data format `(width, height, channels)`. You should set `image_data_format="channels_last"` in your Keras config located at ~/.keras/keras.json. Note that the default input image size for this model is 299x299. # Arguments include_top: whether to include the fully-connected layer at the top of the network. weights: one of `None` (random initialization) or "imagenet" (pre-training on ImageNet). 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 `(299, 299, 3)`. It should have exactly 3 inputs channels, and width and height should be no smaller than 71. E.g. `(150, 150, 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 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. # Returns A Keras model instance. # Raises ValueError: in case of invalid argument for `weights`, or invalid input shape. RuntimeError: If attempting to run this model with a backend that does not support separable convolutions. RspThe `weights` argument should be either `None` (random initialization) or `imagenet` (pre-training on ImageNet).iθsSIf using `weights` as imagenet with `include_top` as true, `classes` should be 1000t tensorflowsAThe Xception model is only available with the TensorFlow backend.t channels_lasts”The Xception model 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.tchannels_firstt default_sizei+tmin_sizeiGt data_formatt include_toptshapettensori itstridesituse_biastnamet block1_conv1tblock1_conv1_bntrelutblock1_conv1_acti@t block1_conv2tblock1_conv2_bntblock1_conv2_acti€itpaddingtsametblock2_sepconv1tblock2_sepconv1_bntblock2_sepconv2_acttblock2_sepconv2tblock2_sepconv2_bnt block2_poolitblock3_sepconv1_acttblock3_sepconv1tblock3_sepconv1_bntblock3_sepconv2_acttblock3_sepconv2tblock3_sepconv2_bnt block3_pooliΨtblock4_sepconv1_acttblock4_sepconv1tblock4_sepconv1_bntblock4_sepconv2_acttblock4_sepconv2tblock4_sepconv2_bnt block4_poolitblockit _sepconv1_actt _sepconv1t _sepconv1_bnt _sepconv2_actt _sepconv2t _sepconv2_bnt _sepconv3_actt _sepconv3t _sepconv3_bnitblock13_sepconv1_acttblock13_sepconv1tblock13_sepconv1_bntblock13_sepconv2_acttblock13_sepconv2tblock13_sepconv2_bnt block13_poolitblock14_sepconv1tblock14_sepconv1_bntblock14_sepconv1_actitblock14_sepconv2tblock14_sepconv2_bntblock14_sepconv2_acttavg_poolt activationtsoftmaxt predictionstavgtmaxtxceptions.xception_weights_tf_dim_ordering_tf_kernels.h5t cache_subdirtmodelss4xception_weights_tf_dim_ordering_tf_kernels_notop.h5N(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(ii(tNonet ValueErrortKRt RuntimeErrortimage_data_formattwarningstwarntset_image_data_formatRRtis_keras_tensorRtFalseRRR R RtaddtrangetstrR RR R RRtTF_WEIGHTS_PATHtTF_WEIGHTS_PATH_NO_TOPt load_weights(Rtweightst input_tensort input_shapetpoolingtclassestold_data_formatt img_inputtxtresidualtitprefixtinputstmodelt weights_path((s:/tmp/pip-build-isqEY4/keras/keras/applications/xception.pytXception/sΰ4      '!''$''$''$+++''$''!      cC@s"|d:}|d8}|d9}|S(Ngΰo@gΰ?g@((Rs((s:/tmp/pip-build-isqEY4/keras/keras/applications/xception.pytpreprocess_inputs   (!t__doc__t __future__RRRaR[RtRRRRRRR R R R tengine.topologyR tutils.data_utilsRRR^timagenet_utilsRRRiRjtTrueR\RzR{(((s:/tmp/pip-build-isqEY4/keras/keras/applications/xception.pyts4  Τ