2VhTfddlZddlmZddlmZddlmZddlmZddlmZddlm Z ddl m Z dd l m Z dd lmZdd lmZdd lmZdd lmZdZddddddZgdgdddgdgdddgdgdddgdgdddgdgddddZdZGdd e ZGd!d"e Z d@d#ZdAd$ZdBd%Z dCd&Ze d'd(g dDd)Ze d*d+g dEd,Z e d-d.g dFd/Z!e d0d1g dGd2Z"e d3d4g dHd5Z#ejId67e_%ejId87e _%ejId97e!_%ejId:7e"_%ejId;7e#_%e d<dAd=Z&e d>dId?Z'e jNjJe'_%y)JN)backend) initializers)layers)ops)random) keras_export)imagenet_utils)Layer) Functional) Sequential)operation_utils) file_utilszFhttps://storage.googleapis.com/tensorflow/keras-applications/convnext/)@8ae6e78ce2933352b1ef4008e6dd2f17bc40771563877d156bc6426c7cf503ff@d547c096cabd03329d7be5562c5e14798aa39ed24b474157cef5e85ab9e49ef1)@ce1277d8f1ee5a0ef0e171469089c18f5233860ceaf9b168049cb9263fd7483c@6fc8009faa2f00c1c1dfce59feea9b0745eb260a7dd11bee65c8e20843da6eab)@52cbb006d3dadd03f6e095a8ca1aca47aecdd75acb4bc74bce1f5c695d0086e6@40a20c5548a5e9202f69735ecc06c990e6b7c9d2de39f0361e27baeb24cb7c45)@070c5ed9ed289581e477741d3b34beffa920db8cf590899d6d2c67fba2a198a6@96f02b6f0753d4f543261bc9d09bed650f24dd6bc02ddde3066135b63d23a1cd)@c1f5ccab661354fc3a79a10fa99af82f0fbf10ec65cb894a3ae0815f17a889ee@de3f8a54174130e0cecdc71583354753d557fcf1f4487331558e2a16ba0cfe05) convnext_tinyconvnext_small convnext_baseconvnext_largeconvnext_xlarge)r r)`)depthsprojection_dims default_size)rrr))r!r"r#i)r*r+r,i)tinysmallbaselargexlargea Instantiates the {name} architecture. References: - [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) (CVPR 2022) 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/). The `base`, `large`, and `xlarge` models were first pre-trained on the ImageNet-21k dataset and then fine-tuned on the ImageNet-1k dataset. The pre-trained parameters of the models were assembled from the [official repository](https://github.com/facebookresearch/ConvNeXt). To get a sense of how these parameters were converted to Keras compatible parameters, please refer to [this repository](https://github.com/sayakpaul/keras-convnext-conversion). Note: Each Keras Application expects a specific kind of input preprocessing. For ConvNeXt, preprocessing is included in the model using a `Normalization` layer. ConvNeXt models expect their inputs to be float or uint8 tensors of pixels with values in the [0-255] range. When calling the `summary()` method after instantiating a ConvNeXt model, prefer setting the `expand_nested` argument `summary()` to `True` to better investigate the instantiated model. Args: include_top: Whether to include the fully-connected layer at the top of the network. Defaults to `True`. weights: One of `None` (random initialization), `"imagenet"` (pre-training on ImageNet-1k), or the path to the weights file to be loaded. Defaults to `"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`. It should have exactly 3 inputs channels. pooling: Optional pooling mode for feature extraction when `include_top` is `False`. Defaults to None. - `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. Defaults to 1000 (number of ImageNet classes). 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. Defaults to `"softmax"`. When loading pretrained weights, `classifier_activation` can only be `None` or `"softmax"`. name: The name of the model (string). Returns: A model instance. c4eZdZdZfdZddZfdZxZS)StochasticDepthaStochastic Depth module. It performs batch-wise dropping rather than sample-wise. In libraries like `timm`, it's similar to `DropPath` layers that drops residual paths sample-wise. References: - https://github.com/rwightman/pytorch-image-models Args: drop_path_rate (float): Probability of dropping paths. Should be within [0, 1]. Returns: Tensor either with the residual path dropped or kept. c 2t|di|||_yN)super__init__drop_path_rate)selfr9kwargs __class__s O/home/dcms/DCMS/lib/python3.12/site-packages/keras/src/applications/convnext.pyr8zStochasticDepth.__init__s "6",c|rd|jz }tj|dfdttj|dz zz}|t j |ddz}tj |}||z |zS|S)Nr)r@)r9rshapelenruniformfloor)r:xtraining keep_probrA random_tensors r=callzStochasticDepth.calls~ D///IYYq\!_&SYYq\1BQ1F)GGE%ua(CCMIIm4M M]2 2r>c^t|}|jd|ji|S)Nr9)r7 get_configupdater9r:configr<s r=rKzStochasticDepth.get_configs-#% ')<)<=> r>N)__name__ __module__ __qualname____doc__r8rIrK __classcell__r<s@r=r3r3s"-r>r3c8eZdZdZfdZdZdZfdZxZS) LayerScalea,Layer scale module. References: - https://arxiv.org/abs/2103.17239 Args: init_values (float): Initial value for layer scale. Should be within [0, 1]. projection_dim (int): Projection dimensionality. Returns: Tensor multiplied to the scale. c @t|di|||_||_yr5)r7r8 init_valuesprojection_dim)r:rYrZr;r<s r=r8zLayerScale.__init__s# "6"&,r>c|j|jftj|jd|_y)NT)rA initializer trainable) add_weightrZrConstantrYgamma)r:_s r=buildzLayerScale.builds:__&&($--d.>.>?%  r>c ||jzSrO)r`)r:rEs r=rIzLayerScale.calls4::~r>ctt|}|j|j|jd|S)N)rYrZ)r7rKrLrYrZrMs r=rKzLayerScale.get_configs:#% #//"&"5"5   r>) rPrQrRrSr8rbrIrKrTrUs@r=rWrWs! -  r>rWcd!dttjdzfd}|S)afConvNeXt block. References: - https://arxiv.org/abs/2201.03545 - https://github.com/facebookresearch/ConvNeXt/blob/main/models/convnext.py Notes: In the original ConvNeXt implementation (linked above), the authors use `Dense` layers for pointwise convolutions for increased efficiency. Following that, this implementation also uses the same. Args: projection_dim (int): Number of filters for convolution layers. In the ConvNeXt paper, this is referred to as projection dimension. drop_path_rate (float): Probability of dropping paths. Should be within [0, 1]. layer_scale_init_value (float): Layer scale value. Should be a small float number. name: name to path to the keras layer. Returns: A function representing a ConvNeXtBlock block. prestemc |}tjdddz|}tjddz|}tjdzd z |}tjd d z |}tjd z |}t dz |}rt dz }ntjddz }|||zS)Nsame_depthwise_conv)filters kernel_sizepaddinggroupsnameư> _layernormepsilonro_pointwise_conv_1rogelu_gelu_pointwise_conv_2 _layer_scale_stochastic_depthlinear _identity)rConv2DLayerNormalizationDense ActivationrWr3)inputsrElayerr9layer_scale_init_valuerorZs r=applyzConvNeXtBlock..applys$  FMM"!))    NF % %d 9L Ma P MFLL^+$9L2L Ma P :F  f4'> :1 = IFLLd5H.H I! L ! - &N* A #T,?%?E%%hTK5GHEa  r>strrget_uid)rZr9rrors```` r= ConvNeXtBlockrs.4 |3wy9::!< Lr>cX!dttjdzfd}|S)z0Normalizes inputs with ImageNet-1k mean and std.rfcRtjgdgddz|}|S)N)g33333^@gR]@gRY@)g)~@gec}@gȷ@_prestem_normalization)meanvariancero)r Normalization)rEros r=rzPreStem..applys; F 8 00   r>r)rors` r=PreStemrs, |3wy9::  Lr>cZttjdfd}|S)aImplementation of classification head of ConvNeXt. Args: num_classes: number of classes for Dense layer classifier_activation: activation function for the Dense layer name: name prefix Returns: Classification head function. headctjdz|}tjddz|}tjdz|}|S)N _head_gaprvrp_head_layernormrr _head_dense) activationro)rGlobalAveragePooling2Drr)rEclassifier_activationro num_classess r=rzHead..apply<sw BF ) )tk/A B1 E F % %t&77   FLL , %    r>r)rrrors``` r=Headr.s) |7??6*+  Lr>c ztjdk(r td|dvs tj|s td|dk(r|r| dk7rtd| t j | |dtj|| } | tj| }n/tj| stj| | }n| }| tj| d }| }n|}|}|r:tjd k(rdnd}| |dz }|dk(rt||}ttj|d dd|dztjd|dzg|dz}g}|j!|d}t#|D]}ttjd|dzt%|ztj||dzdd|dzt%|zg|dzt%|z}|j!|t'j(d|t+|Dcgc] }t-|}}d }d}t#|D]O}||}t#||D])}t/|||||z||d|d|z|}+|||z }Q|r+t j0| |t3| | | }n[| d!k(rtj4}n| d"k(rtj6}tjd#}t9|||$}|dk(rwt:|j=d%d&}||d'k7s||d(k7r!td)|d*|d'd+|d,|d(d+|d- ||vr'td.|d/|d0|j=d%d&d1|dk(rW|rd2}t>|d } nd3}t>|d} ||z}!tj@|!tB|!zd4| 5}"|jE|"|S||jE||Scc}w)6a Instantiates ConvNeXt architecture given specific configuration. Args: depths: An iterable containing depths for each individual stages. projection_dims: An iterable containing output number of channels of each individual stages. drop_path_rate: Stochastic depth probability. If 0.0, then stochastic depth won't be used. layer_scale_init_value: Layer scale coefficient. If 0.0, layer scaling won't be used. default_size: Default input image size. name: An optional name for the model. include_preprocessing: boolean denoting whether to include preprocessing in the model. When `weights="imagenet"` this should always be `True`. But for other models (e.g., randomly initialized) you should set it to `False` and apply preprocessing to data accordingly. include_top: Boolean denoting whether to include classification head to the model. weights: one of `None` (random initialization), `"imagenet"` (pre-training on ImageNet-1k), 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`. It should have exactly 3 inputs channels. 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. Returns: A model instance. channels_firstzConvNeXt 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.rzbIf using `weights="imagenet"` with `include_top=True`, `classes` should be 1000. Received classes= )r'min_size data_formatrequire_flattenweights)rA)tensorrAr channels_lastrr@rvrt _stem_conv)rlstridesrorp_stem_layernormrr_stem_downsampling_layernorm__downsampling_conv__downsampling_block__stage__block_)rZr9rro)rrroavgmax)rs)routputsro convnext_r%r&z*Architecture configuration does not match z variant. When using pre-trained weights, the model architecture must match the pre-trained configuration exactly. Expected depths: z, got: z. Expected projection_dims: .z Model name "z"" does not match weights variant "z]". When using imagenet weights, model name must contain the weights variant (e.g., "convnext_z").z.h5z _notop.h5models) cache_subdir file_hash)#rimage_data_format ValueErrorrexistsr obtain_input_shaperInputis_keras_tensorr get_source_inputsrr r~rappendrangernplinspacesumfloatrvalidate_activationrrGlobalMaxPooling2Dr MODEL_CONFIGSsplitWEIGHTS_HASHESget_fileBASE_WEIGHTS_PATH load_weights)#r%r&r9rr'roinclude_preprocessing include_topr input_tensor input_shapepoolingclassesr weights_name img_inputrrE channel_axis num_channelsstemdownsample_layersnum_downsample_layersidownsample_layerdepth_drop_ratescurnum_convnext_blocksjmodelexpected_config file_suffixr file_name weights_paths# r=ConvNeXtrKsF~  "&66 1  ) )Z->->w-G <  *D 'y *  !33!--/# KLL{3 &&|4 L LI$I 22<@C  **,?AQ #  ..Q7    !12!3,++c>3v;Ga C & '  a  #vay! A .q1/a8'=gaSs33   A  vay **+@'J D"7    e /--/2A  +))+A.A 3F % %d 3A 6 fad ;E*' (:(:;(G(KL oh/ //2C"DD<\NK-.=X-F,GHx;"#456go=Na Q  t #tf% >"E%%k2267s< * K&|4Q7I%K&|4Q7I;& !**   )!  <( L   7# L_sP8z(keras.applications.convnext.ConvNeXtTinyzkeras.applications.ConvNeXtTinyc rtdtddtddddtdd|||||||||S) Nrr-r%r&rrpr'rr%r&r9rr'rorrrrrrrrrr rrrrrrrrros r= ConvNeXtTinyr?]" $V$X.%f-.?@#"6*>: 3!3 r>z)keras.applications.convnext.ConvNeXtSmallz keras.applications.ConvNeXtSmallc rtdtddtddddtdd|||||||||S) Nrr.r%r&rrpr'rrrs r= ConvNeXtSmallrc]" %W%h/%g./@A#"7+N; 3!3 r>z(keras.applications.convnext.ConvNeXtBasezkeras.applications.ConvNeXtBasec rtdtddtddddtdd|||||||||S) Nrr/r%r&rrpr'rrrs r= ConvNeXtBaserrr>z)keras.applications.convnext.ConvNeXtLargez keras.applications.ConvNeXtLargec rtdtddtddddtdd|||||||||S) Nrr0r%r&rrpr'rrrs r= ConvNeXtLargerrr>z*keras.applications.convnext.ConvNeXtXLargez!keras.applications.ConvNeXtXLargec rtdtddtddddtdd|||||||||S) Nrr1r%r&rrpr'rrrs r=ConvNeXtXLargers]" &X&x0%h/0AB#"8,^< 3!3 r>rrvrrrrz,keras.applications.convnext.preprocess_inputc|S)aA placeholder method for backward compatibility. The preprocessing logic has been included in the convnext model implementation. Users are no longer required to call this method to normalize the input data. This method does nothing and only kept as a placeholder to align the API surface between old and new version of model. Args: x: A floating point `numpy.array` or a tensor. data_format: Optional data format of the image tensor/array. Defaults to None, in which case the global setting `keras.backend.image_data_format()` is used (unless you changed it, it defaults to `"channels_last"`).{mode} Returns: Unchanged `numpy.array` or tensor. r6)rErs r=preprocess_inputrs & Hr>z.keras.applications.convnext.decode_predictionsc0tj||S)N)top)r decode_predictions)predsrs r=rrs  , ,U <)rrpNrO)rNN) rrpr$convnextTTNNNNrsoftmaxN) TTrNNNrrr) TTrNNNrrr) TTrNNNrrr) TTrNNNrrr) TTrNNNrrr))(numpyr keras.srcrrrrrkeras.src.api_exportrkeras.src.applicationsr keras.src.layers.layerr keras.src.modelsr r keras.src.opsr keras.src.utilsrrrrBASE_DOCSTRINGr3rWrrrrrrrrrformatrSrrr6r>r=r s"-1('')&M #4.   .  0   0  1+ 8BJ"e"J''VKO;|(@    #nh2)   #   <3*   #   <2)   #   <3*   #   <4+   #   <&,,.,A &--?-C %,,.,A &--?-C '..4D.E<= > *>?=@=,>>FFr>