hhj dZddlZddlZddlZddlmcmZddlm Z ddl m Z ddl m ZddlmZddlmZddlmZdd lmZdd lmZdd lmZdd lmZdd lmZddlmZedGddZedgGddeZedGddeZ edGddeZ!edGddeZ"edGd d!eZ#ed"Gd#d$eZ$ed%Gd&d'eZ%ed(Gd)d*eZ&ed+Gd,d-eZ'ed.Gd/d0eZ(ed1Gd2d3eZ)ed4Gd5d6eZ*ed7Gd8d9eZ+ed:Gd;dd?eZ-ed@GdAdBeZ.edCGdDdEeZ/edFGdGdHeZ0edIdJdKdLdMdNejbj2jddOZ3ddPZ4ejje3ejldQZ7edRdSdTdUdVdWejbj2jddXZ8ejje8ejldYZ9edZd[d\d]d^d_ejbj2jdd`Z:ejje:ejldaZ;edbdcdddedfdgejbj2jddhZ<ejjeedkdlejbj2jddmZ?edndoejbj2jddpZ@edqejbj2jddrZAedsgejbj2jdddtZBedudvdwdxejbj2jddyZCedzd{ejbj2jd dd|ZDejjeDejl dd}ZEed~dejbj2jd ddZFejjeFejl ddZGeddejbj2jd ddZHejjeHejl ddZIeddejbj2jd ddZJejjeJejl ddZKeddejbj2jd ddZLejjeLejl ddZMeddddddddejbj2jddZNeddejbj2jddZOedgdejbj2jdddZPeJxZQZRe3xZSZTe8xZUZVe:xZWZXelambda)r __class____name__strip _name_scopers rrzLoss._set_name_scope\sT 99 #~~66<>IIr88KKMVV,F"++F3G#,,V4H"x';)$%++-I(99 tYM 55 5       s#E&DE> E&E E&&E/c|di|S)zInstantiates a `Loss` from its config (output of `get_config()`). Args: config: Output of `get_config()`. Returns: A `Loss` instance. r=)clsconfigs r from_configzLoss.from_configs}V}rc4|j|jdS)z4Returns the config dictionary for a `Loss` instance.rrrBr#s r get_configzLoss.get_configs!^^TYY??rctd)aInvokes the `Loss` instance. Args: y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`, except sparse loss functions such as sparse categorical crossentropy where shape = `[batch_size, d0, .. dN-1]` y_pred: The predicted values. shape = `[batch_size, d0, .. dN]` Returns: Loss values with the shape `[batch_size, d0, .. dN-1]`. z"Must be implemented in subclasses.)NotImplementedError)rr2r3s rr)z Loss.calls""FGGrc|jswtjjrY|jt j jk(s'|jt j jk(r td|jt j jk(rt j jS|jS)z?Handles `AUTO` reduction cases and returns the reduction value.aZPlease use `tf.keras.losses.Reduction.SUM` or `tf.keras.losses.Reduction.NONE` for loss reduction when losses are used with `tf.distribute.Strategy`, except for specifying losses in `Model.compile()` for use by the built-in training looop `Model.fit()`. Please see https://www.tensorflow.org/tutorials/distribute/custom_training for more details.) rr' distribute has_strategyrrrAUTOSUM_OVER_BATCH_SIZE ValueErrorr#s rr/zLoss._get_reductions// **,,":":"?"??>>++??@@  >>\55:: :++?? ?~~rN)r __module__ __qualname____doc__rrrIrrr; classmethodr@rCabcabstractmethodrfor_subclass_implementersr)r/r=rrrr)ss2".!9!9!>!>T04=~  @ ++ H, Hrrz-keras.__internal__.losses.LossFunctionWrapper)v1cpeZdZdZej j dffd ZdZfdZ e dZ xZ S)LossFunctionWrapperz*Wraps a loss function in the `Loss` class.Nc Dt|||||_||_y)aInitializes `LossFunctionWrapper` class. Args: fn: The loss function to wrap, with signature `fn(y_true, y_pred, **kwargs)`. reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. **kwargs: The keyword arguments that are passed on to `fn`. rBN)superrfn _fn_kwargs)rrYrrkwargsrs rrzLossFunctionWrapper.__init__s%* 948 rcdtj|r.tj|rtj||\}}tjj j |jtjj j}|||fi|jS)zInvokes the `LossFunctionWrapper` instance. Args: y_true: Ground truth values. y_pred: The predicted values. Returns: Loss values per sample. ) r' is_tensorrsqueeze_or_expand_dimensionsr*r+r,rYr-rZ)rr2r3ag_fns rr)zLossFunctionWrapper.calls << BLL$8)FFNFF))44 GGR__..AAC VV7t77rci}|jjD]4\}}tj|rt j |n|||<6t jrddlm }||j|d<t|1}tt|jt|jzS)Nr)get_registered_namerY)rZitemsr is_tensor_or_variablerevalrsaving_v3_enabledtf_keras.src.utilsrarYrXrCdictlist)rr?kvra base_configrs rrCzLossFunctionWrapper.get_configsOO))+ DAq#+#A#A!#D Q! 1I   ' ' ) >.tww7F4Lg(* D**,-V\\^0DDEErctjr*|jdd}|r|turt ||d<|di|S)zInstantiates a `Loss` from its config (output of `get_config()`). Args: config: Output of `get_config()`. Returns: A `keras.losses.Loss` instance. rYNr=)rrepoprVget)r>r?fn_names rr@zLossFunctionWrapper.from_configsD  ' ' )jjt,G3"55"7|t }V}r) r rMrNrOrrrIrr)rCrPr@ __classcell__rs@rrVrVs=4)4499!28( F  rrVzkeras.losses.MeanSquaredErrorcPeZdZdZej j dffd ZxZS)MeanSquaredErroraComputes the mean of squares of errors between labels and predictions. `loss = mean(square(y_true - y_pred))` Standalone usage: >>> y_true = [[0., 1.], [0., 0.]] >>> y_pred = [[1., 1.], [1., 0.]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> mse = tf.keras.losses.MeanSquaredError() >>> mse(y_true, y_pred).numpy() 0.5 >>> # Calling with 'sample_weight'. >>> mse(y_true, y_pred, sample_weight=[0.7, 0.3]).numpy() 0.25 >>> # Using 'sum' reduction type. >>> mse = tf.keras.losses.MeanSquaredError( ... reduction=tf.keras.losses.Reduction.SUM) >>> mse(y_true, y_pred).numpy() 1.0 >>> # Using 'none' reduction type. >>> mse = tf.keras.losses.MeanSquaredError( ... reduction=tf.keras.losses.Reduction.NONE) >>> mse(y_true, y_pred).numpy() array([0.5, 0.5], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.MeanSquaredError()) ``` mean_squared_errorc2t|t||y)aAInitializes `MeanSquaredError` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'mean_squared_error'. rrN)rXrrtrrrrs rrzMeanSquaredError.__init__Vs& +$)Lr r rMrNrOrrrIrrprqs@rrsrs0s)"J%0055>> y_true = [[0., 1.], [0., 0.]] >>> y_pred = [[1., 1.], [1., 0.]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> mae = tf.keras.losses.MeanAbsoluteError() >>> mae(y_true, y_pred).numpy() 0.5 >>> # Calling with 'sample_weight'. >>> mae(y_true, y_pred, sample_weight=[0.7, 0.3]).numpy() 0.25 >>> # Using 'sum' reduction type. >>> mae = tf.keras.losses.MeanAbsoluteError( ... reduction=tf.keras.losses.Reduction.SUM) >>> mae(y_true, y_pred).numpy() 1.0 >>> # Using 'none' reduction type. >>> mae = tf.keras.losses.MeanAbsoluteError( ... reduction=tf.keras.losses.Reduction.NONE) >>> mae(y_true, y_pred).numpy() array([0.5, 0.5], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.MeanAbsoluteError()) ``` mean_absolute_errorc2t|t||y)aCInitializes `MeanAbsoluteError` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'mean_absolute_error'. rvN)rXrr{rws rrzMeanAbsoluteError.__init__s* ,49Mrrxrqs@rrzrzls)"L**// "NNrrzz(keras.losses.MeanAbsolutePercentageErrorcPeZdZdZej j dffd ZxZS)MeanAbsolutePercentageErroraComputes the mean absolute percentage error between `y_true` & `y_pred`. Formula: `loss = 100 * abs((y_true - y_pred) / y_true)` Note that to avoid dividing by zero, a small epsilon value is added to the denominator. Standalone usage: >>> y_true = [[2., 1.], [2., 3.]] >>> y_pred = [[1., 1.], [1., 0.]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> mape = tf.keras.losses.MeanAbsolutePercentageError() >>> mape(y_true, y_pred).numpy() 50. >>> # Calling with 'sample_weight'. >>> mape(y_true, y_pred, sample_weight=[0.7, 0.3]).numpy() 20. >>> # Using 'sum' reduction type. >>> mape = tf.keras.losses.MeanAbsolutePercentageError( ... reduction=tf.keras.losses.Reduction.SUM) >>> mape(y_true, y_pred).numpy() 100. >>> # Using 'none' reduction type. >>> mape = tf.keras.losses.MeanAbsolutePercentageError( ... reduction=tf.keras.losses.Reduction.NONE) >>> mape(y_true, y_pred).numpy() array([25., 75.], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.MeanAbsolutePercentageError()) ``` mean_absolute_percentage_errorc2t|t||y)aXInitializes `MeanAbsolutePercentageError` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'mean_absolute_percentage_error'. rvN)rXrrrws rrz$MeanAbsolutePercentageError.__init__*  *  rrxrqs@rr~r~s'(X**// -  rr~z(keras.losses.MeanSquaredLogarithmicErrorcPeZdZdZej j dffd ZxZS)MeanSquaredLogarithmicErroraZComputes the mean squared logarithmic error between `y_true` & `y_pred`. `loss = square(log(y_true + 1.) - log(y_pred + 1.))` Standalone usage: >>> y_true = [[0., 1.], [0., 0.]] >>> y_pred = [[1., 1.], [1., 0.]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> msle = tf.keras.losses.MeanSquaredLogarithmicError() >>> msle(y_true, y_pred).numpy() 0.240 >>> # Calling with 'sample_weight'. >>> msle(y_true, y_pred, sample_weight=[0.7, 0.3]).numpy() 0.120 >>> # Using 'sum' reduction type. >>> msle = tf.keras.losses.MeanSquaredLogarithmicError( ... reduction=tf.keras.losses.Reduction.SUM) >>> msle(y_true, y_pred).numpy() 0.480 >>> # Using 'none' reduction type. >>> msle = tf.keras.losses.MeanSquaredLogarithmicError( ... reduction=tf.keras.losses.Reduction.NONE) >>> msle(y_true, y_pred).numpy() array([0.240, 0.240], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.MeanSquaredLogarithmicError()) ``` mean_squared_logarithmic_errorc2t|t||y)aXInitializes `MeanSquaredLogarithmicError` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'mean_squared_logarithmic_error'. rvN)rXrrrws rrz$MeanSquaredLogarithmicError.__init__rrrxrqs@rrrs'#N**// -  rrzkeras.losses.BinaryCrossentropycVeZdZdZdddej j dffd ZxZS)BinaryCrossentropya! Computes the cross-entropy loss between true labels and predicted labels. Use this cross-entropy loss for binary (0 or 1) classification applications. The loss function requires the following inputs: - `y_true` (true label): This is either 0 or 1. - `y_pred` (predicted value): This is the model's prediction, i.e, a single floating-point value which either represents a [logit](https://en.wikipedia.org/wiki/Logit), (i.e, value in [-inf, inf] when `from_logits=True`) or a probability (i.e, value in [0., 1.] when `from_logits=False`). **Recommended Usage:** (set `from_logits=True`) With `tf.keras` API: ```python model.compile( loss=tf.keras.losses.BinaryCrossentropy(from_logits=True), .... ) ``` As a standalone function: >>> # Example 1: (batch_size = 1, number of samples = 4) >>> y_true = [0, 1, 0, 0] >>> y_pred = [-18.6, 0.51, 2.94, -12.8] >>> bce = tf.keras.losses.BinaryCrossentropy(from_logits=True) >>> bce(y_true, y_pred).numpy() 0.865 >>> # Example 2: (batch_size = 2, number of samples = 4) >>> y_true = [[0, 1], [0, 0]] >>> y_pred = [[-18.6, 0.51], [2.94, -12.8]] >>> # Using default 'auto'/'sum_over_batch_size' reduction type. >>> bce = tf.keras.losses.BinaryCrossentropy(from_logits=True) >>> bce(y_true, y_pred).numpy() 0.865 >>> # Using 'sample_weight' attribute >>> bce(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy() 0.243 >>> # Using 'sum' reduction` type. >>> bce = tf.keras.losses.BinaryCrossentropy(from_logits=True, ... reduction=tf.keras.losses.Reduction.SUM) >>> bce(y_true, y_pred).numpy() 1.730 >>> # Using 'none' reduction type. >>> bce = tf.keras.losses.BinaryCrossentropy(from_logits=True, ... reduction=tf.keras.losses.Reduction.NONE) >>> bce(y_true, y_pred).numpy() array([0.235, 1.496], dtype=float32) **Default Usage:** (set `from_logits=False`) >>> # Make the following updates to the above "Recommended Usage" section >>> # 1. Set `from_logits=False` >>> tf.keras.losses.BinaryCrossentropy() # OR ...('from_logits=False') >>> # 2. Update `y_pred` to use probabilities instead of logits >>> y_pred = [0.6, 0.3, 0.2, 0.8] # OR [[0.6, 0.3], [0.2, 0.8]] Fbinary_crossentropycFt|t|||||||_y)aInitializes `BinaryCrossentropy` instance. Args: from_logits: Whether to interpret `y_pred` as a tensor of [logit](https://en.wikipedia.org/wiki/Logit) values. By default, we assume that `y_pred` contains probabilities (i.e., values in [0, 1]). label_smoothing: Float in [0, 1]. When 0, no smoothing occurs. When > 0, we compute the loss between the predicted labels and a smoothed version of the true labels, where the smoothing squeezes the labels towards 0.5. Larger values of `label_smoothing` correspond to heavier smoothing. axis: The axis along which to compute crossentropy (the features axis). Defaults to -1. reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Name for the op. Defaults to 'binary_crossentropy'. rr from_logitslabel_smoothingaxisN)rXrrrrrrrrrrs rrzBinaryCrossentropy.__init__qs4D  #+  'rrxrqs@rrr1s0<@ **// " *'*'rrz$keras.losses.BinaryFocalCrossentropycfeZdZdZddddddej j dffd Zfd ZxZ S) BinaryFocalCrossentropyaComputes focal cross-entropy loss between true labels and predictions. Binary cross-entropy loss is often used for binary (0 or 1) classification tasks. The loss function requires the following inputs: - `y_true` (true label): This is either 0 or 1. - `y_pred` (predicted value): This is the model's prediction, i.e, a single floating-point value which either represents a [logit](https://en.wikipedia.org/wiki/Logit), (i.e, value in [-inf, inf] when `from_logits=True`) or a probability (i.e, value in [0., 1.] when `from_logits=False`). According to [Lin et al., 2018](https://arxiv.org/pdf/1708.02002.pdf), it helps to apply a "focal factor" to down-weight easy examples and focus more on hard examples. By default, the focal tensor is computed as follows: `focal_factor = (1 - output) ** gamma` for class 1 `focal_factor = output ** gamma` for class 0 where `gamma` is a focusing parameter. When `gamma=0`, this function is equivalent to the binary crossentropy loss. With the `compile()` API: ```python model.compile( loss=tf.keras.losses.BinaryFocalCrossentropy(gamma=2.0, from_logits=True), .... ) ``` As a standalone function: >>> # Example 1: (batch_size = 1, number of samples = 4) >>> y_true = [0, 1, 0, 0] >>> y_pred = [-18.6, 0.51, 2.94, -12.8] >>> loss = tf.keras.losses.BinaryFocalCrossentropy(gamma=2, ... from_logits=True) >>> loss(y_true, y_pred).numpy() 0.691 >>> # Apply class weight >>> loss = tf.keras.losses.BinaryFocalCrossentropy( ... apply_class_balancing=True, gamma=2, from_logits=True) >>> loss(y_true, y_pred).numpy() 0.51 >>> # Example 2: (batch_size = 2, number of samples = 4) >>> y_true = [[0, 1], [0, 0]] >>> y_pred = [[-18.6, 0.51], [2.94, -12.8]] >>> # Using default 'auto'/'sum_over_batch_size' reduction type. >>> loss = tf.keras.losses.BinaryFocalCrossentropy(gamma=3, ... from_logits=True) >>> loss(y_true, y_pred).numpy() 0.647 >>> # Apply class weight >>> loss = tf.keras.losses.BinaryFocalCrossentropy( ... apply_class_balancing=True, gamma=3, from_logits=True) >>> loss(y_true, y_pred).numpy() 0.482 >>> # Using 'sample_weight' attribute with focal effect >>> loss = tf.keras.losses.BinaryFocalCrossentropy(gamma=3, ... from_logits=True) >>> loss(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy() 0.133 >>> # Apply class weight >>> loss = tf.keras.losses.BinaryFocalCrossentropy( ... apply_class_balancing=True, gamma=3, from_logits=True) >>> loss(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy() 0.097 >>> # Using 'sum' reduction` type. >>> loss = tf.keras.losses.BinaryFocalCrossentropy(gamma=4, ... from_logits=True, ... reduction=tf.keras.losses.Reduction.SUM) >>> loss(y_true, y_pred).numpy() 1.222 >>> # Apply class weight >>> loss = tf.keras.losses.BinaryFocalCrossentropy( ... apply_class_balancing=True, gamma=4, from_logits=True, ... reduction=tf.keras.losses.Reduction.SUM) >>> loss(y_true, y_pred).numpy() 0.914 >>> # Using 'none' reduction type. >>> loss = tf.keras.losses.BinaryFocalCrossentropy( ... gamma=5, from_logits=True, ... reduction=tf.keras.losses.Reduction.NONE) >>> loss(y_true, y_pred).numpy() array([0.0017 1.1561], dtype=float32) >>> # Apply class weight >>> loss = tf.keras.losses.BinaryFocalCrossentropy( ... apply_class_balancing=True, gamma=5, from_logits=True, ... reduction=tf.keras.losses.Reduction.NONE) >>> loss(y_true, y_pred).numpy() array([0.0004 0.8670], dtype=float32) Args: apply_class_balancing: A bool, whether to apply weight balancing on the binary classes 0 and 1. alpha: A weight balancing factor for class 1, default is `0.25` as mentioned in reference [Lin et al., 2018]( https://arxiv.org/pdf/1708.02002.pdf). The weight for class 0 is `1.0 - alpha`. gamma: A focusing parameter used to compute the focal factor, default is `2.0` as mentioned in the reference [Lin et al., 2018](https://arxiv.org/pdf/1708.02002.pdf). from_logits: Whether to interpret `y_pred` as a tensor of [logit](https://en.wikipedia.org/wiki/Logit) values. By default, we assume that `y_pred` are probabilities (i.e., values in `[0, 1]`). label_smoothing: Float in `[0, 1]`. When `0`, no smoothing occurs. When > `0`, we compute the loss between the predicted labels and a smoothed version of the true labels, where the smoothing squeezes the labels towards `0.5`. Larger values of `label_smoothing` correspond to heavier smoothing. axis: The axis along which to compute crossentropy (the features axis). Defaults to `-1`. reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Name for the op. Defaults to 'binary_focal_crossentropy'. F?@rrbinary_focal_crossentropyc vt |t|||||||| ||_||_||_||_y)z/Initializes `BinaryFocalCrossentropy` instance.)apply_class_balancingalphagammarrrrrN)rXrrrrrr) rrrrrrrrrrs rrz BinaryFocalCrossentropy.__init__&sR  %"7#+  '%:"  rc|j|j|jd}t|}t t |jt |jzS)N)rrr)rrrrXrCrgrhrbrr?rkrs rrCz"BinaryFocalCrossentropy.get_configBsY%)%?%?ZZZZ  g(* D**,-V\\^0DDEEr r rMrNrOrrrIrrCrprqs@rrrsADP$ **// (8FFrrz$keras.losses.CategoricalCrossentropycVeZdZdZdddej j dffd ZxZS)CategoricalCrossentropyaComputes the crossentropy loss between the labels and predictions. Use this crossentropy loss function when there are two or more label classes. We expect labels to be provided in a `one_hot` representation. If you want to provide labels as integers, please use `SparseCategoricalCrossentropy` loss. There should be `# classes` floating point values per feature. In the snippet below, there is `# classes` floating pointing values per example. The shape of both `y_pred` and `y_true` are `[batch_size, num_classes]`. Standalone usage: >>> y_true = [[0, 1, 0], [0, 0, 1]] >>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> cce = tf.keras.losses.CategoricalCrossentropy() >>> cce(y_true, y_pred).numpy() 1.177 >>> # Calling with 'sample_weight'. >>> cce(y_true, y_pred, sample_weight=tf.constant([0.3, 0.7])).numpy() 0.814 >>> # Using 'sum' reduction type. >>> cce = tf.keras.losses.CategoricalCrossentropy( ... reduction=tf.keras.losses.Reduction.SUM) >>> cce(y_true, y_pred).numpy() 2.354 >>> # Using 'none' reduction type. >>> cce = tf.keras.losses.CategoricalCrossentropy( ... reduction=tf.keras.losses.Reduction.NONE) >>> cce(y_true, y_pred).numpy() array([0.0513, 2.303], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.CategoricalCrossentropy()) ``` Frrcategorical_crossentropyc8t|t|||||y)aInitializes `CategoricalCrossentropy` instance. Args: from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in [0, 1]. When > 0, label values are smoothed, meaning the confidence on label values are relaxed. For example, if `0.1`, use `0.1 / num_classes` for non-target labels and `0.9 + 0.1 / num_classes` for target labels. axis: The axis along which to compute crossentropy (the features axis). Defaults to -1. reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'categorical_crossentropy'. rN)rXrrrs rrz CategoricalCrossentropy.__init__{s*@  $#+  rrxrqs@rrrLs0+^ **// ' ' ' rrz)keras.losses.CategoricalFocalCrossentropycdeZdZdZdddddej j dffd Zfd ZxZ S) CategoricalFocalCrossentropyuGComputes the alpha balanced focal crossentropy loss. Use this crossentropy loss function when there are two or more label classes and if you want to handle class imbalance without using `class_weights`. We expect labels to be provided in a `one_hot` representation. According to [Lin et al., 2018](https://arxiv.org/pdf/1708.02002.pdf), it helps to apply a focal factor to down-weight easy examples and focus more on hard examples. The general formula for the focal loss (FL) is as follows: `FL(p_t) = (1 − p_t)^gamma * log(p_t)` where `p_t` is defined as follows: `p_t = output if y_true == 1, else 1 - output` `(1 − p_t)^gamma` is the `modulating_factor`, where `gamma` is a focusing parameter. When `gamma` = 0, there is no focal effect on the cross entropy. `gamma` reduces the importance given to simple examples in a smooth manner. The authors use alpha-balanced variant of focal loss (FL) in the paper: `FL(p_t) = −alpha * (1 − p_t)^gamma * log(p_t)` where `alpha` is the weight factor for the classes. If `alpha` = 1, the loss won't be able to handle class imbalance properly as all classes will have the same weight. This can be a constant or a list of constants. If alpha is a list, it must have the same length as the number of classes. The formula above can be generalized to: `FL(p_t) = alpha * (1 − p_t)^gamma * CrossEntropy(y_true, y_pred)` where minus comes from `CrossEntropy(y_true, y_pred)` (CE). Extending this to multi-class case is straightforward: `FL(p_t) = alpha * (1 − p_t)^gamma * CategoricalCE(y_true, y_pred)` In the snippet below, there is `# classes` floating pointing values per example. The shape of both `y_pred` and `y_true` are `[batch_size, num_classes]`. Standalone usage: >>> y_true = [[0., 1., 0.], [0., 0., 1.]] >>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> cce = tf.keras.losses.CategoricalFocalCrossentropy() >>> cce(y_true, y_pred).numpy() 0.23315276 >>> # Calling with 'sample_weight'. >>> cce(y_true, y_pred, sample_weight=tf.constant([0.3, 0.7])).numpy() 0.1632 >>> # Using 'sum' reduction type. >>> cce = tf.keras.losses.CategoricalFocalCrossentropy( ... reduction=tf.keras.losses.Reduction.SUM) >>> cce(y_true, y_pred).numpy() 0.46631 >>> # Using 'none' reduction type. >>> cce = tf.keras.losses.CategoricalFocalCrossentropy( ... reduction=tf.keras.losses.Reduction.NONE) >>> cce(y_true, y_pred).numpy() array([3.2058331e-05, 4.6627346e-01], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='adam', loss=tf.keras.losses.CategoricalFocalCrossentropy()) ``` Args: alpha: A weight balancing factor for all classes, default is `0.25` as mentioned in the reference. It can be a list of floats or a scalar. In the multi-class case, alpha may be set by inverse class frequency by using `compute_class_weight` from `sklearn.utils`. gamma: A focusing parameter, default is `2.0` as mentioned in the reference. It helps to gradually reduce the importance given to simple (easy) examples in a smooth manner. from_logits: Whether `output` is expected to be a logits tensor. By default, we consider that `output` encodes a probability distribution. label_smoothing: Float in [0, 1]. When > 0, label values are smoothed, meaning the confidence on label values are relaxed. For example, if `0.1`, use `0.1 / num_classes` for non-target labels and `0.9 + 0.1 / num_classes` for target labels. axis: The axis along which to compute crossentropy (the features axis). Defaults to -1. reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'categorical_focal_crossentropy'. rrFrrcategorical_focal_crossentropyc ft|t|||||||||_||_||_y)z4Initializes `CategoricalFocalCrossentropy` instance.)rrrrrrrN)rXrrrrr) rrrrrrrrrs rrz%CategoricalFocalCrossentropy.__init__sG  *#+  '  rc|j|jd}t| }t t |j t |j zS)N)rr)rrrXrCrgrhrbrs rrCz'CategoricalFocalCrossentropy.get_config(sPZZZZ g(* D**,-V\\^0DDEErrrqs@rrrs>fT **// -2FFrrz*keras.losses.SparseCategoricalCrossentropycTeZdZdZddej j dffd ZxZS)SparseCategoricalCrossentropyaComputes the crossentropy loss between the labels and predictions. Use this crossentropy loss function when there are two or more label classes. We expect labels to be provided as integers. If you want to provide labels using `one-hot` representation, please use `CategoricalCrossentropy` loss. There should be `# classes` floating point values per feature for `y_pred` and a single floating point value per feature for `y_true`. In the snippet below, there is a single floating point value per example for `y_true` and `# classes` floating pointing values per example for `y_pred`. The shape of `y_true` is `[batch_size]` and the shape of `y_pred` is `[batch_size, num_classes]`. Standalone usage: >>> y_true = [1, 2] >>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> scce = tf.keras.losses.SparseCategoricalCrossentropy() >>> scce(y_true, y_pred).numpy() 1.177 >>> # Calling with 'sample_weight'. >>> scce(y_true, y_pred, sample_weight=tf.constant([0.3, 0.7])).numpy() 0.814 >>> # Using 'sum' reduction type. >>> scce = tf.keras.losses.SparseCategoricalCrossentropy( ... reduction=tf.keras.losses.Reduction.SUM) >>> scce(y_true, y_pred).numpy() 2.354 >>> # Using 'none' reduction type. >>> scce = tf.keras.losses.SparseCategoricalCrossentropy( ... reduction=tf.keras.losses.Reduction.NONE) >>> scce(y_true, y_pred).numpy() array([0.0513, 2.303], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.SparseCategoricalCrossentropy()) ``` FNsparse_categorical_crossentropyc6t|t||||y)alInitializes `SparseCategoricalCrossentropy` instance. Args: from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. ignore_class: Optional integer. The ID of a class to be ignored during loss computation. This is useful, for example, in segmentation problems featuring a "void" class (commonly -1 or 255) in segmentation maps. By default (`ignore_class=None`), all classes are considered. reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction ption will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'sparse_categorical_crossentropy'. )rrr ignore_classN)rXrr)rrrrrrs rrz&SparseCategoricalCrossentropy.__init__bs&>  +#%  rrxrqs@rrr1s--b**// . % % rrzkeras.losses.CosineSimilaritycReZdZdZdej j dffd ZxZS)CosineSimilaritya Computes the cosine similarity between labels and predictions. Note that it is a number between -1 and 1. When it is a negative number between -1 and 0, 0 indicates orthogonality and values closer to -1 indicate greater similarity. The values closer to 1 indicate greater dissimilarity. This makes it usable as a loss function in a setting where you try to maximize the proximity between predictions and targets. If either `y_true` or `y_pred` is a zero vector, cosine similarity will be 0 regardless of the proximity between predictions and targets. `loss = -sum(l2_norm(y_true) * l2_norm(y_pred))` Standalone usage: >>> y_true = [[0., 1.], [1., 1.]] >>> y_pred = [[1., 0.], [1., 1.]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> cosine_loss = tf.keras.losses.CosineSimilarity(axis=1) >>> # l2_norm(y_true) = [[0., 1.], [1./1.414, 1./1.414]] >>> # l2_norm(y_pred) = [[1., 0.], [1./1.414, 1./1.414]] >>> # l2_norm(y_true) . l2_norm(y_pred) = [[0., 0.], [0.5, 0.5]] >>> # loss = mean(sum(l2_norm(y_true) . l2_norm(y_pred), axis=1)) >>> # = -((0. + 0.) + (0.5 + 0.5)) / 2 >>> cosine_loss(y_true, y_pred).numpy() -0.5 >>> # Calling with 'sample_weight'. >>> cosine_loss(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy() -0.0999 >>> # Using 'sum' reduction type. >>> cosine_loss = tf.keras.losses.CosineSimilarity(axis=1, ... reduction=tf.keras.losses.Reduction.SUM) >>> cosine_loss(y_true, y_pred).numpy() -0.999 >>> # Using 'none' reduction type. >>> cosine_loss = tf.keras.losses.CosineSimilarity(axis=1, ... reduction=tf.keras.losses.Reduction.NONE) >>> cosine_loss(y_true, y_pred).numpy() array([-0., -0.999], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.CosineSimilarity(axis=1)) ``` Args: axis: The axis along which the cosine similarity is computed (the features axis). Defaults to -1. reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'cosine_similarity'. rcosine_similarityc4t|t|||y)N)rrr)rXrr)rrrrrs rrzCosineSimilarity.__init__s  D  rrxrqs@rrrs*>D**//   rrzkeras.losses.HingecPeZdZdZej j dffd ZxZS)Hingea/Computes the hinge loss between `y_true` & `y_pred`. `loss = maximum(1 - y_true * y_pred, 0)` `y_true` values are expected to be -1 or 1. If binary (0 or 1) labels are provided we will convert them to -1 or 1. Standalone usage: >>> y_true = [[0., 1.], [0., 0.]] >>> y_pred = [[0.6, 0.4], [0.4, 0.6]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> h = tf.keras.losses.Hinge() >>> h(y_true, y_pred).numpy() 1.3 >>> # Calling with 'sample_weight'. >>> h(y_true, y_pred, sample_weight=[1, 0]).numpy() 0.55 >>> # Using 'sum' reduction type. >>> h = tf.keras.losses.Hinge( ... reduction=tf.keras.losses.Reduction.SUM) >>> h(y_true, y_pred).numpy() 2.6 >>> # Using 'none' reduction type. >>> h = tf.keras.losses.Hinge( ... reduction=tf.keras.losses.Reduction.NONE) >>> h(y_true, y_pred).numpy() array([1.1, 1.5], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.Hinge()) ``` hingec2t|t||y)aInitializes `Hinge` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction ption will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'hinge'. rvN)rXrrrws rrzHinge.__init__s TY?rrxrqs@rrrs(%N".!9!9!>!>W@@rrzkeras.losses.SquaredHingecPeZdZdZej j dffd ZxZS) SquaredHingea_Computes the squared hinge loss between `y_true` & `y_pred`. `loss = square(maximum(1 - y_true * y_pred, 0))` `y_true` values are expected to be -1 or 1. If binary (0 or 1) labels are provided we will convert them to -1 or 1. Standalone usage: >>> y_true = [[0., 1.], [0., 0.]] >>> y_pred = [[0.6, 0.4], [0.4, 0.6]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> h = tf.keras.losses.SquaredHinge() >>> h(y_true, y_pred).numpy() 1.86 >>> # Calling with 'sample_weight'. >>> h(y_true, y_pred, sample_weight=[1, 0]).numpy() 0.73 >>> # Using 'sum' reduction type. >>> h = tf.keras.losses.SquaredHinge( ... reduction=tf.keras.losses.Reduction.SUM) >>> h(y_true, y_pred).numpy() 3.72 >>> # Using 'none' reduction type. >>> h = tf.keras.losses.SquaredHinge( ... reduction=tf.keras.losses.Reduction.NONE) >>> h(y_true, y_pred).numpy() array([1.46, 2.26], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.SquaredHinge()) ``` squared_hingec2t|t||y)a'Initializes `SquaredHinge` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction ption will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'squared_hinge'. rvN)rXrrrws rrzSquaredHinge.__init__<s$ TYGrrxrqs@rrrs(%P%0055OHHrrzkeras.losses.CategoricalHingecPeZdZdZej j dffd ZxZS)CategoricalHingea%Computes the categorical hinge loss between `y_true` & `y_pred`. `loss = maximum(neg - pos + 1, 0)` where `neg=maximum((1-y_true)*y_pred) and pos=sum(y_true*y_pred)` Standalone usage: >>> y_true = [[0, 1], [0, 0]] >>> y_pred = [[0.6, 0.4], [0.4, 0.6]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> h = tf.keras.losses.CategoricalHinge() >>> h(y_true, y_pred).numpy() 1.4 >>> # Calling with 'sample_weight'. >>> h(y_true, y_pred, sample_weight=[1, 0]).numpy() 0.6 >>> # Using 'sum' reduction type. >>> h = tf.keras.losses.CategoricalHinge( ... reduction=tf.keras.losses.Reduction.SUM) >>> h(y_true, y_pred).numpy() 2.8 >>> # Using 'none' reduction type. >>> h = tf.keras.losses.CategoricalHinge( ... reduction=tf.keras.losses.Reduction.NONE) >>> h(y_true, y_pred).numpy() array([1.2, 1.6], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.CategoricalHinge()) ``` categorical_hingec2t|t||y)a?Initializes `CategoricalHinge` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction ption will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'categorical_hinge'. rvN)rXrrrws rrzCategoricalHinge.__init__xs& *Krrxrqs@rrrQs)#L%0055>> y_true = [[0., 1.], [0., 0.]] >>> y_pred = [[1., 1.], [0., 0.]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> p = tf.keras.losses.Poisson() >>> p(y_true, y_pred).numpy() 0.5 >>> # Calling with 'sample_weight'. >>> p(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy() 0.4 >>> # Using 'sum' reduction type. >>> p = tf.keras.losses.Poisson( ... reduction=tf.keras.losses.Reduction.SUM) >>> p(y_true, y_pred).numpy() 0.999 >>> # Using 'none' reduction type. >>> p = tf.keras.losses.Poisson( ... reduction=tf.keras.losses.Reduction.NONE) >>> p(y_true, y_pred).numpy() array([0.999, 0.], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.Poisson()) ``` poissonc2t|t||y)aInitializes `Poisson` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction ption will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'poisson'. rvN)rXrrrws rrzPoisson.__init__s tyArrxrqs@rrrs("H".!9!9!>!>YBBrrzkeras.losses.LogCoshcPeZdZdZej j dffd ZxZS)LogCoshaComputes the logarithm of the hyperbolic cosine of the prediction error. `logcosh = log((exp(x) + exp(-x))/2)`, where x is the error `y_pred - y_true`. Standalone usage: >>> y_true = [[0., 1.], [0., 0.]] >>> y_pred = [[1., 1.], [0., 0.]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> l = tf.keras.losses.LogCosh() >>> l(y_true, y_pred).numpy() 0.108 >>> # Calling with 'sample_weight'. >>> l(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy() 0.087 >>> # Using 'sum' reduction type. >>> l = tf.keras.losses.LogCosh( ... reduction=tf.keras.losses.Reduction.SUM) >>> l(y_true, y_pred).numpy() 0.217 >>> # Using 'none' reduction type. >>> l = tf.keras.losses.LogCosh( ... reduction=tf.keras.losses.Reduction.NONE) >>> l(y_true, y_pred).numpy() array([0.217, 0.], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.LogCosh()) ``` log_coshc2t|t||y)aInitializes `LogCosh` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction ption will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'log_cosh'. rvN)rXrrrws rrzLogCosh.__init__s$  Brrxrqs@rrrs(#L%0055JCCrrzkeras.losses.KLDivergencecPeZdZdZej j dffd ZxZS) KLDivergencea>Computes Kullback-Leibler divergence loss between `y_true` & `y_pred`. `loss = y_true * log(y_true / y_pred)` See: https://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence Standalone usage: >>> y_true = [[0, 1], [0, 0]] >>> y_pred = [[0.6, 0.4], [0.4, 0.6]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> kl = tf.keras.losses.KLDivergence() >>> kl(y_true, y_pred).numpy() 0.458 >>> # Calling with 'sample_weight'. >>> kl(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy() 0.366 >>> # Using 'sum' reduction type. >>> kl = tf.keras.losses.KLDivergence( ... reduction=tf.keras.losses.Reduction.SUM) >>> kl(y_true, y_pred).numpy() 0.916 >>> # Using 'none' reduction type. >>> kl = tf.keras.losses.KLDivergence( ... reduction=tf.keras.losses.Reduction.NONE) >>> kl(y_true, y_pred).numpy() array([0.916, -3.08e-06], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.KLDivergence()) ``` kl_divergencec2t|t||y)a7Initializes `KLDivergence` instance. Args: reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction ption will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'kl_divergence'. rvN)rXrrrws rrzKLDivergence.__init__+s& TYGrrxrqs@rrrs($N%0055OHHrrzkeras.losses.HubercReZdZdZdej j dffd ZxZS)HuberasComputes the Huber loss between `y_true` & `y_pred`. For each value x in `error = y_true - y_pred`: ``` loss = 0.5 * x^2 if |x| <= d loss = 0.5 * d^2 + d * (|x| - d) if |x| > d ``` where d is `delta`. See: https://en.wikipedia.org/wiki/Huber_loss Standalone usage: >>> y_true = [[0, 1], [0, 0]] >>> y_pred = [[0.6, 0.4], [0.4, 0.6]] >>> # Using 'auto'/'sum_over_batch_size' reduction type. >>> h = tf.keras.losses.Huber() >>> h(y_true, y_pred).numpy() 0.155 >>> # Calling with 'sample_weight'. >>> h(y_true, y_pred, sample_weight=[1, 0]).numpy() 0.09 >>> # Using 'sum' reduction type. >>> h = tf.keras.losses.Huber( ... reduction=tf.keras.losses.Reduction.SUM) >>> h(y_true, y_pred).numpy() 0.31 >>> # Using 'none' reduction type. >>> h = tf.keras.losses.Huber( ... reduction=tf.keras.losses.Reduction.NONE) >>> h(y_true, y_pred).numpy() array([0.18, 0.13], dtype=float32) Usage with the `compile()` API: ```python model.compile(optimizer='sgd', loss=tf.keras.losses.Huber()) ``` ? huber_lossc4t|t|||y)aInitializes `Huber` instance. Args: delta: A float, the point where the Huber loss function changes from a quadratic to linear. reduction: Type of `tf.keras.losses.Reduction` to apply to loss. Default value is `AUTO`. `AUTO` indicates that the reduction ption will be determined by the usage context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When used under a `tf.distribute.Strategy`, except via `Model.compile()` and `Model.fit()`, using `AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this custom training [tutorial]( https://www.tensorflow.org/tutorials/distribute/custom_training) for more details. name: Optional name for the instance. Defaults to 'huber_loss'. )rrdeltaN)rXrhuber)rrrrrs rrzHuber.__init__ms. TYeLrrxrqs@rrrAs,(X**//  MMrrz keras.metrics.mean_squared_errorzkeras.metrics.msezkeras.metrics.MSEzkeras.losses.mean_squared_errorzkeras.losses.msezkeras.losses.MSEctj|}tj||j}t j tj j||dS)a)Computes the mean squared error between labels and predictions. After computing the squared distance between the inputs, the mean value over the last dimension is returned. `loss = mean(square(y_true - y_pred), axis=-1)` Standalone usage: >>> y_true = np.random.randint(0, 2, size=(2, 3)) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.mean_squared_error(y_true, y_pred) >>> assert loss.shape == (2,) >>> assert np.array_equal( ... loss.numpy(), np.mean(np.square(y_true - y_pred), axis=-1)) Args: y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Mean squared error values. shape = `[batch_size, d0, .. dN-1]`. rr)r'convert_to_tensorcastdtypermeanmathsquared_differencer2r3s rrtrtsKB ! !& )F WWVV\\ *F <<2266B LLrc ddfd fd}t|tjs||jS|jj dd}t |dkDr"tj||j}n!tjg|j}||fDcgc]}|j}}|r1|D cgc] } t | } } | d| ddz k(r |dd d|d<tj|t |dkD } tj|} tj| 5t!j"| ||f| cd d d Scc}wcc} w#1swYy xYw) aApply a loss function on a per batch basis. Args: loss_fn: The loss function y_true: truth values (RaggedTensor) y_pred: predicted values (RaggedTensor) y_pred_extra_dim: whether y_pred has an additional dimension compared to y_true Returns: Loss-function result. A dense tensor if the output has a single dimension (per-batch loss value); a ragged tensor otherwise. cDtj|jDcgc]o}tjtjj tj |tjtjdgqc}Scc}w)zReturns true if this RaggedTensor has the same row_lengths across all ragged dimensions and thus can be converted to a dense tensor without loss of information. Args: rt: RaggedTensor. r) r' reduce_allnested_row_lengthsequalrreduce_variancerrfloatxconstant)rtrow_lenss rrt_is_equiv_densez4_ragged_tensor_apply_loss..rt_is_equiv_denses}}}!# 5 5 7  GG++'..*:;KK&     sA4Bc&td|DS)Nc3tK|]0}t|tjr|jn|2ywrL) isinstancer' RaggedTensor to_tensor).0rs r zG_ragged_tensor_apply_loss.._convert_to_dense..s0 )R__=BLLN2 E s68)tuple)inputss r_convert_to_densez4_ragged_tensor_apply_loss.._convert_to_denses    rc|}|r;t|tjs!tjj|}|S|s*t|tjr|j }|S)zAdapt the result to ragged or dense tensor according to the expected output type. This is done so that all the return values of the map operation have the same type. )rr'r from_tensorr)r ragged_outputrloss_fns r _call_lossz-_ragged_tensor_apply_loss.._call_losss\ V  Ar!?++A.A:a#A Arc\}}t|tjr(tj|fdfdSS)Nc"SrLr=)rrrrsrz=_ragged_tensor_apply_loss.._wrapper..s #4V#._wrapper..s 6=9r)rr'rcond)rrrr3rrrrs`` r_wrapperz+_ragged_tensor_apply_loss.._wrappersF 6 fboo .77!&)L9  rrr)shaperN)r)elemsr)rr'rrras_listlenRaggedTensorSpecr TensorSpecnested_row_splits functoolspartialr assert_splits_matchcontrol_dependenciesr map_fn)rr2r3y_pred_extra_dimrlshapespecrnested_splits_listslistrdimsrassertion_listrrrs` @@@r_ragged_tensor_apply_lossr si *    fboo .vv//122 \\ ! ! #Ab )F 6{Q""v||D}}2V\\::@&9IJ2"..JJ);<U<< 8uQx!| #$6q$9#2$> q !   xs6{Q GF 445GHN  0Q$$VFF3C4PQQK=QQs=E6E;FF c$tt||S)aImplements support for handling RaggedTensors. Args: y_true: RaggedTensor truth values. shape = `[batch_size, d0, .. dN]`. y_pred: RaggedTensor predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Mean squared error values. shape = `[batch_size, d0, .. dN-1]`. When the number of dimensions of the batch feature vector [d0, .. dN] is greater than one the return value is a RaggedTensor. Otherwise, a Dense tensor with dimensions [batch_size] is returned. )r rtrs r_ragged_tensor_msers %%7 HHrz!keras.metrics.mean_absolute_errorzkeras.metrics.maezkeras.metrics.MAEz keras.losses.mean_absolute_errorzkeras.losses.maezkeras.losses.MAEctj|}tj||j}t j tj ||z dS)aComputes the mean absolute error between labels and predictions. `loss = mean(abs(y_true - y_pred), axis=-1)` Standalone usage: >>> y_true = np.random.randint(0, 2, size=(2, 3)) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.mean_absolute_error(y_true, y_pred) >>> assert loss.shape == (2,) >>> assert np.array_equal( ... loss.numpy(), np.mean(np.abs(y_true - y_pred), axis=-1)) Args: y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Mean absolute error values. shape = `[batch_size, d0, .. dN-1]`. rr)r'rrrrrabsrs rr{r{sF< ! !& )F WWVV\\ *F <<v/b 99rc$tt||S)z-RaggedTensor adapter for mean_absolute_error.)r r{rs r_ragged_tensor_maer;s %%8&& IIrz,keras.metrics.mean_absolute_percentage_errorzkeras.metrics.mapezkeras.metrics.MAPEz+keras.losses.mean_absolute_percentage_errorzkeras.losses.mapezkeras.losses.MAPEcHtj|}tj||j}tj||z t j tj|t jz }dt j|dzS)aDComputes the mean absolute percentage error between `y_true` & `y_pred`. `loss = 100 * mean(abs((y_true - y_pred) / y_true), axis=-1)` Standalone usage: >>> y_true = np.random.random(size=(2, 3)) >>> y_true = np.maximum(y_true, 1e-7) # Prevent division by zero >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.mean_absolute_percentage_error(y_true, y_pred) >>> assert loss.shape == (2,) >>> assert np.array_equal( ... loss.numpy(), ... 100. * np.mean(np.abs((y_true - y_pred) / y_true), axis=-1)) Args: y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Mean absolute percentage error values. shape = `[batch_size, d0, .. dN-1]`. gY@rr) r'rrrrrmaximumepsilonr)r2r3diffs rrrAsuB ! !& )F WWVV\\ *F 66 &GOOBFF6NGOO>> y_true = np.random.randint(0, 2, size=(2, 3)) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.mean_squared_logarithmic_error(y_true, y_pred) >>> assert loss.shape == (2,) >>> y_true = np.maximum(y_true, 1e-7) >>> y_pred = np.maximum(y_pred, 1e-7) >>> assert np.allclose( ... loss.numpy(), ... np.mean( ... np.square(np.log(y_true + 1.) - np.log(y_pred + 1.)), axis=-1)) Args: y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Mean squared logarithmic error values. shape = `[batch_size, d0, .. dN-1]`. rrr) r'rrrrlogrrrrr)r2r3 first_log second_logs rrrrsF ! !& )F WWVV\\ *F GOOFGOO4EFLMIW__VW__5FG#MNJ << ""9j9 rc$tt||S)z.Implements support for handling RaggedTensors.)r rrs r_ragged_tensor_mslerrrctjd}tjd}tjtj||}fd}tjj j ||fd}|S)z!Converts binary labels into -1/1.rrcdzdz S)Nrrr=r2sr_convert_binary_labelsz5_maybe_convert_labels.._convert_binary_labelssV|c!!rcSrLr=r"srrz'_maybe_convert_labels..s6r)r'rr logical_orr* smart_cond)r2 are_zerosare_ones is_binaryr#updated_y_trues` r_maybe_convert_labelsr+sm#Ixx"H bmmIx@AI"__//::)>N rzkeras.metrics.squared_hingezkeras.losses.squared_hingec  tj|}tj||j}t |}t j tjtjd||zz ddS)anComputes the squared hinge loss between `y_true` & `y_pred`. `loss = mean(square(maximum(1 - y_true * y_pred, 0)), axis=-1)` Standalone usage: >>> y_true = np.random.choice([-1, 1], size=(2, 3)) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.squared_hinge(y_true, y_pred) >>> assert loss.shape == (2,) >>> assert np.array_equal( ... loss.numpy(), ... np.mean(np.square(np.maximum(1. - y_true * y_pred, 0.)), axis=-1)) Args: y_true: The ground truth values. `y_true` values are expected to be -1 or 1. If binary (0 or 1) labels are provided we will convert them to -1 or 1. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Squared hinge loss values. shape = `[batch_size, d0, .. dN-1]`. rrrr) r'rrrr+rrsquarerrs rrrse4 ! !& )F WWVV\\ *F "6 *F << "**S6F?2C89 rzkeras.metrics.hingezkeras.losses.hingectj|}tj||j}t |}t j tjd||zz ddS)aCComputes the hinge loss between `y_true` & `y_pred`. `loss = mean(maximum(1 - y_true * y_pred, 0), axis=-1)` Standalone usage: >>> y_true = np.random.choice([-1, 1], size=(2, 3)) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.hinge(y_true, y_pred) >>> assert loss.shape == (2,) >>> assert np.array_equal( ... loss.numpy(), ... np.mean(np.maximum(1. - y_true * y_pred, 0.), axis=-1)) Args: y_true: The ground truth values. `y_true` values are expected to be -1 or 1. If binary (0 or 1) labels are provided we will convert them to -1 or 1. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Hinge loss values. shape = `[batch_size, d0, .. dN-1]`. rrrr)r'rrrr+rrrrs rrrsW4 ! !& )F WWVV\\ *F "6 *F << 3&#8#>R HHrzkeras.losses.categorical_hingecRtj|}tj||j}tj||zd}tj d|z |zd}tjd|j}tj ||z dz|S)aComputes the categorical hinge loss between `y_true` & `y_pred`. `loss = maximum(neg - pos + 1, 0)` where `neg=maximum((1-y_true)*y_pred) and pos=sum(y_true*y_pred)` Standalone usage: >>> y_true = np.random.randint(0, 3, size=(2,)) >>> y_true = tf.keras.utils.to_categorical(y_true, num_classes=3) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.categorical_hinge(y_true, y_pred) >>> assert loss.shape == (2,) >>> pos = np.sum(y_true * y_pred, axis=-1) >>> neg = np.amax((1. - y_true) * y_pred, axis=-1) >>> assert np.array_equal(loss.numpy(), np.maximum(0., neg - pos + 1.)) Args: y_true: The ground truth values. `y_true` values are expected to be either `{-1, +1}` or `{0, 1}` (i.e. a one-hot-encoded tensor). y_pred: The predicted values. Returns: Categorical hinge loss values. rrrr)r'rrr reduce_sum reduce_maxr)r2r3posnegzeros rrrs6 ! !& )F WWVV\\ *F --b 1C --v/b 9C 773 %D ::cCi#ot ,,rzkeras.losses.huberc Rtj|tj}tj|tj}tj|tj}tj||}tj |}tj d|j}tjtj||k|tj|z||z|tj|zz dS)a!Computes Huber loss value. For each value x in `error = y_true - y_pred`: ``` loss = 0.5 * x^2 if |x| <= d loss = d * |x| - 0.5 * d^2 if |x| > d ``` where d is `delta`. See: https://en.wikipedia.org/wiki/Huber_loss Args: y_true: tensor of true targets. y_pred: tensor of predicted targets. delta: A float, the point where the Huber loss function changes from a quadratic to linear. Returns: Tensor with one scalar loss entry per sample. r?rr) r'rrrsubtractrrrrwherer-)r2r3rerror abs_errorhalfs rrrs,WWV7>>#3 4F WWV7>>#3 4F GGE!1 2E KK 'Eu I  9?? ;D <<    299U# # I ryy'7 7 7   rzkeras.losses.log_coshzkeras.losses.logcoshzkeras.metrics.log_coshzkeras.metrics.logcoshctj|}tj||j}d}t j |||z dS)aLogarithm of the hyperbolic cosine of the prediction error. `log(cosh(x))` is approximately equal to `(x ** 2) / 2` for small `x` and to `abs(x) - log(2)` for large `x`. This means that 'logcosh' works mostly like the mean squared error, but will not be so strongly affected by the occasional wildly incorrect prediction. Standalone usage: >>> y_true = np.random.random(size=(2, 3)) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.logcosh(y_true, y_pred) >>> assert loss.shape == (2,) >>> x = y_pred - y_true >>> assert np.allclose( ... loss.numpy(), ... np.mean(x + np.log(np.exp(-2. * x) + 1.) - tf.math.log(2.), ... axis=-1), ... atol=1e-5) Args: y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Logcosh error values. shape = `[batch_size, d0, .. dN-1]`. c|tjjd|zztjtjj d|j z S)Ngr)r'rsoftplusrrr)xs r_logcoshzlog_cosh.._logcoshgsC   * *RWWRWW[[5Eqww-O O rrr)r'rrrrr)r2r3rAs rrrAsJF ! !& )F WWVV\\ *F <<&1 ;;rz&keras.metrics.categorical_crossentropyz%keras.losses.categorical_crossentropyc ttrtddtt j t j jt j jjddk(r*tjdjdtd fd }tjjj|fd tj| S) aComputes the categorical crossentropy loss. Standalone usage: >>> y_true = [[0, 1, 0], [0, 0, 1]] >>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]] >>> loss = tf.keras.losses.categorical_crossentropy(y_true, y_pred) >>> assert loss.shape == (2,) >>> loss.numpy() array([0.0513, 2.303], dtype=float32) Args: y_true: Tensor of one-hot true targets. y_pred: Tensor of predicted targets. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in [0, 1]. If > `0` then smooth the labels. For example, if `0.1`, use `0.1 / num_classes` for non-target labels and `0.9 + 0.1 / num_classes` for target labels. axis: Defaults to -1. The dimension along which the entropy is computed. Returns: Categorical crossentropy loss value. -`axis` must be of type `int`. Received: axis= of type r6rrzIn loss categorical_crossentropy, expected y_pred.shape to be (batch_size, num_classes) with num_classes > 1. Received: y_pred.shape=B. Consider using 'binary_crossentropy' if you only have 2 classes. stacklevelctjtjj}dz z|z zS)Nrr'rrr) num_classesrrr3r2s r_smooth_labelsz0categorical_crossentropy.._smooth_labelssDggbhhv.t4fllC ./ k )  rcSrLr=r"srrz*categorical_crossentropy..r)rr)rboolrKtyper'rrrrwarningswarn SyntaxWarningr*r&rrr2r3rrrrLs`` `` rrrosB$ "V9T$ZL :   ! !& )F WWVV\\ *F**?&,,OO ||B1  <JO O    __ ' ' 2 2F  + +Kd rcVtjt|||}t|||S)aImplements support for handling RaggedTensors. Args: y_true: Tensor of one-hot true targets. y_pred: Tensor of predicted targets. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in [0, 1]. If > `0` then smooth the labels. For example, if `0.1`, use `0.1 / num_classes` for non-target labels and `0.9 + 0.1 / num_classes` for target labels. axis: The axis along which to compute crossentropy (the features axis). Defaults to -1. Returns: Categorical crossentropy loss value. Expected shape: (batch, sequence_len, n_classes) with sequence_len being variable per batch. Return shape: (batch, sequence_len). When used by CategoricalCrossentropy() with the default reduction (SUM_OVER_BATCH_SIZE), the reduction averages the loss over the number of elements independent of the batch. E.g. if the RaggedTensor has 2 batches with [2, 1] values respectively the resulting loss is the sum of the individual loss values divided by 3. rrr)rrrr r2r3rrrrYs r'_ragged_tensor_categorical_crossentropyrXs1<    '   B %R 88rz,keras.metrics.categorical_focal_crossentropyz+keras.losses.categorical_focal_crossentropyc t|trtd|dt|t j t j jt j jjddk(r*tjdjdtd fd }tjjj|fd tj|||| S) aEComputes the categorical focal crossentropy loss. Standalone usage: >>> y_true = [[0, 1, 0], [0, 0, 1]] >>> y_pred = [[0.05, 0.9, 0.05], [0.1, 0.85, 0.05]] >>> loss = tf.keras.losses.categorical_focal_crossentropy(y_true, y_pred) >>> assert loss.shape == (2,) >>> loss.numpy() array([2.63401289e-04, 6.75912094e-01], dtype=float32) Args: y_true: Tensor of one-hot true targets. y_pred: Tensor of predicted targets. alpha: A weight balancing factor for all classes, default is `0.25` as mentioned in the reference. It can be a list of floats or a scalar. In the multi-class case, alpha may be set by inverse class frequency by using `compute_class_weight` from `sklearn.utils`. gamma: A focusing parameter, default is `2.0` as mentioned in the reference. It helps to gradually reduce the importance given to simple examples in a smooth manner. When `gamma` = 0, there is no focal effect on the categorical crossentropy. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in [0, 1]. If > `0` then smooth the labels. For example, if `0.1`, use `0.1 / num_classes` for non-target labels and `0.9 + 0.1 / num_classes` for target labels. axis: Defaults to -1. The dimension along which the entropy is computed. Returns: Categorical focal crossentropy loss value. rCrDr6rrzIn loss categorical_focal_crossentropy, expected y_pred.shape to be (batch_size, num_classes) with num_classes > 1. Received: y_pred.shape=rErFrGctjtjdj}dz z|z zS)NrrrJ)rKrr3r2s rrLz6categorical_focal_crossentropy.._smooth_labels sDggbhhv.r2FLLA ./ k )  rcSrLr=r"srrz0categorical_focal_crossentropy.." rNr)targetoutputrrrr)rrOrKrPr'rrrrrQrRrSr*r&rr)r2r3rrrrrrLs`` ` rrrs^$ "V9T$ZL :   ! !& )F WWVV\\ *F**?&,,OO ||B1  <JO O    __ ' ' 2 2F  1 1   rcZtjt|||||}t|||S)aImplements support for handling RaggedTensors. Expected shape: (batch, sequence_len, n_classes) with sequence_len being variable per batch. Return shape: (batch, sequence_len). When used by CategoricalFocalCrossentropy() with the default reduction (SUM_OVER_BATCH_SIZE), the reduction averages the loss over the number of elements independent of the batch. E.g. if the RaggedTensor has 2 batches with [2, 1] values respectively the resulting loss is the sum of the individual loss values divided by 3. Args: alpha: A weight balancing factor for all classes, default is `0.25` as mentioned in the reference. It can be a list of floats or a scalar. In the multi-class case, alpha may be set by inverse class frequency by using `compute_class_weight` from `sklearn.utils`. gamma: A focusing parameter, default is `2.0` as mentioned in the reference. It helps to gradually reduce the importance given to simple examples in a smooth manner. When `gamma` = 0, there is no focal effect on the categorical crossentropy. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in [0, 1]. If > `0` then smooth the labels. For example, if `0.1`, use `0.1 / num_classes` for non-target labels and `0.9 + 0.1 / num_classes` for target labels. axis: Defaults to -1. The dimension along which the entropy is computed. Returns: Categorical focal crossentropy loss value. )rrrrr)rrrr )r2r3rrrrrrYs r-_ragged_tensor_categorical_focal_crossentropyr_/ s8R   &'   B %R 88rz-keras.metrics.sparse_categorical_crossentropyz,keras.losses.sparse_categorical_crossentropyc6tj|||||S)aComputes the sparse categorical crossentropy loss. Standalone usage: >>> y_true = [1, 2] >>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]] >>> loss = tf.keras.losses.sparse_categorical_crossentropy(y_true, y_pred) >>> assert loss.shape == (2,) >>> loss.numpy() array([0.0513, 2.303], dtype=float32) >>> y_true = [[[ 0, 2], ... [-1, -1]], ... [[ 0, 2], ... [-1, -1]]] >>> y_pred = [[[[1.0, 0.0, 0.0], [0.0, 0.0, 1.0]], ... [[0.2, 0.5, 0.3], [0.0, 1.0, 0.0]]], ... [[[1.0, 0.0, 0.0], [0.0, 0.5, 0.5]], ... [[0.2, 0.5, 0.3], [0.0, 1.0, 0.0]]]] >>> loss = tf.keras.losses.sparse_categorical_crossentropy( ... y_true, y_pred, ignore_class=-1) >>> loss.numpy() array([[[2.3841855e-07, 2.3841855e-07], [0.0000000e+00, 0.0000000e+00]], [[2.3841855e-07, 6.9314730e-01], [0.0000000e+00, 0.0000000e+00]]], dtype=float32) Args: y_true: Ground truth values. y_pred: The predicted values. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. axis: Defaults to -1. The dimension along which the entropy is computed. ignore_class: Optional integer. The ID of a class to be ignored during loss computation. This is useful, for example, in segmentation problems featuring a "void" class (commonly -1 or 255) in segmentation maps. By default (`ignore_class=None`), all classes are considered. Returns: Sparse categorical crossentropy loss value. rrr)rr)r2r3rrrs rrrc s'f  2 2!   rcZtjt|||}t|||dS)a%Implements support for handling RaggedTensors. Expected y_pred shape: (batch, sequence_len, n_classes) with sequence_len being variable per batch. Return shape: (batch, sequence_len). When used by SparseCategoricalCrossentropy() with the default reduction (SUM_OVER_BATCH_SIZE), the reduction averages the loss over the number of elements independent of the batch. E.g. if the RaggedTensor has 2 batches with [2, 1] values respectively, the resulting loss is the sum of the individual loss values divided by 3. raT)r)rrrr )r2r3rrrrYs r._ragged_tensor_sparse_categorical_crossentropyrc s3   '!   B %R$ OOrz!keras.metrics.binary_crossentropyz keras.losses.binary_crossentropycttj|}tj|jtj|jfd}tjj j |fdt jt j|||S)aComputes the binary crossentropy loss. Standalone usage: >>> y_true = [[0, 1], [0, 0]] >>> y_pred = [[0.6, 0.4], [0.4, 0.6]] >>> loss = tf.keras.losses.binary_crossentropy(y_true, y_pred) >>> assert loss.shape == (2,) >>> loss.numpy() array([0.916 , 0.714], dtype=float32) Args: y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in [0, 1]. If > `0` then smooth the labels by squeezing them towards 0.5 That is, using `1. - 0.5 * label_smoothing` for the target class and `0.5 * label_smoothing` for the non-target class. axis: The axis along which the mean is computed. Defaults to -1. Returns: Binary crossentropy loss value. shape = `[batch_size, d0, .. dN-1]`. r6c dz zdzzSNrr7r=rr2srrLz+binary_crossentropy.._smooth_labels ./#2GGGrcSrLr=r"srrz%binary_crossentropy.. rNr)rr) r'rrrr*r&rrrrTs` ` rrr s@ ! !& )F WWVV\\ *F**?&,,OOH__ ' ' 2 2F <<##FF L  rcVtjt|||}t|||S)aImplements support for handling RaggedTensors. Args: y_true: Tensor of one-hot true targets. y_pred: Tensor of predicted targets. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in [0, 1]. If > `0` then smooth the labels. For example, if `0.1`, use `0.1 / num_classes` for non-target labels and `0.9 + 0.1 / num_classes` for target labels. axis: Axis along which to compute crossentropy. Returns: Binary crossentropy loss value. Expected shape: (batch, sequence_len) with sequence_len being variable per batch. Return shape: (batch,); returns the per batch mean of the loss values. When used by BinaryCrossentropy() with the default reduction (SUM_OVER_BATCH_SIZE), the reduction averages the per batch losses over the number of batches. rV)rrrr rWs r"_ragged_tensor_binary_crossentropyrk s16   '   B %R 88rz'keras.metrics.binary_focal_crossentropyz&keras.losses.binary_focal_crossentropyc ztj|}tj|jtj|jfd}tjj j |fdt jt j||||||S)aComputes the binary focal crossentropy loss. According to [Lin et al., 2018](https://arxiv.org/pdf/1708.02002.pdf), it helps to apply a focal factor to down-weight easy examples and focus more on hard examples. By default, the focal tensor is computed as follows: `focal_factor = (1 - output)**gamma` for class 1 `focal_factor = output**gamma` for class 0 where `gamma` is a focusing parameter. When `gamma` = 0, there is no focal effect on the binary crossentropy loss. If `apply_class_balancing == True`, this function also takes into account a weight balancing factor for the binary classes 0 and 1 as follows: `weight = alpha` for class 1 (`target == 1`) `weight = 1 - alpha` for class 0 where `alpha` is a float in the range of `[0, 1]`. Standalone usage: >>> y_true = [[0, 1], [0, 0]] >>> y_pred = [[0.6, 0.4], [0.4, 0.6]] >>> loss = tf.keras.losses.binary_focal_crossentropy(y_true, y_pred, ... gamma=2) >>> assert loss.shape == (2,) >>> loss.numpy() array([0.330, 0.206], dtype=float32) Args: y_true: Ground truth values, of shape `(batch_size, d0, .. dN)`. y_pred: The predicted values, of shape `(batch_size, d0, .. dN)`. apply_class_balancing: A bool, whether to apply weight balancing on the binary classes 0 and 1. alpha: A weight balancing factor for class 1, default is `0.25` as mentioned in the reference. The weight for class 0 is `1.0 - alpha`. gamma: A focusing parameter, default is `2.0` as mentioned in the reference. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in `[0, 1]`. If higher than 0 then smooth the labels by squeezing them towards `0.5`, i.e., using `1. - 0.5 * label_smoothing` for the target class and `0.5 * label_smoothing` for the non-target class. axis: The axis along which the mean is computed. Defaults to `-1`. Returns: Binary focal crossentropy loss value. shape = `[batch_size, d0, .. dN-1]`. r6c dz zdzzSrfr=rgsrrLz1binary_focal_crossentropy.._smooth_labelsQ rhrcSrLr=r"srrz+binary_focal_crossentropy..U rNr)r\r]rrrrr) r'rrrr*r&rrr) r2r3rrrrrrrLs ` ` rrr s@ ! !& )F WWVV\\ *F**?&,,OOH__ ' ' 2 2F <<))"7#    rc \tjt||||||}t|||S)aImplements support for handling RaggedTensors. Expected shape: `(batch, sequence_len)` with sequence_len being variable per batch. Return shape: `(batch,)`; returns the per batch mean of the loss values. When used by BinaryFocalCrossentropy() with the default reduction (SUM_OVER_BATCH_SIZE), the reduction averages the per batch losses over the number of batches. Args: y_true: Tensor of one-hot true targets. y_pred: Tensor of predicted targets. apply_class_balancing: A bool, whether to apply weight balancing on the binary classes 0 and 1. alpha: A weight balancing factor for class 1, default is `0.25` as mentioned in the reference [Lin et al., 2018]( https://arxiv.org/pdf/1708.02002.pdf). The weight for class 0 is `1.0 - alpha`. gamma: A focusing parameter, default is `2.0` as mentioned in the reference. from_logits: Whether `y_pred` is expected to be a logits tensor. By default, we assume that `y_pred` encodes a probability distribution. label_smoothing: Float in `[0, 1]`. If > `0` then smooth the labels. For example, if `0.1`, use `0.1 / num_classes` for non-target labels and `0.9 + 0.1 / num_classes` for target labels. axis: Axis along which to compute crossentropy. Returns: Binary focal crossentropy loss value. )rrrrrr)rrrr ) r2r3rrrrrrrYs r(_ragged_tensor_binary_focal_crossentropyrpe s;T   !3'  B %R 88rzkeras.metrics.kl_divergencez)keras.metrics.kullback_leibler_divergencezkeras.metrics.kldzkeras.metrics.KLDzkeras.losses.kl_divergencez(keras.losses.kullback_leibler_divergencezkeras.losses.kldzkeras.losses.KLDctj|}tj||j}t j |t j d}t j |t j d}tj|tjj||z zdS)azComputes Kullback-Leibler divergence loss between `y_true` & `y_pred`. `loss = y_true * log(y_true / y_pred)` See: https://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence Standalone usage: >>> y_true = np.random.randint(0, 2, size=(2, 3)).astype(np.float64) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.kullback_leibler_divergence(y_true, y_pred) >>> assert loss.shape == (2,) >>> y_true = tf.keras.backend.clip(y_true, 1e-7, 1) >>> y_pred = tf.keras.backend.clip(y_pred, 1e-7, 1) >>> assert np.array_equal( ... loss.numpy(), np.sum(y_true * np.log(y_true / y_pred), axis=-1)) Args: y_true: Tensor of true targets. y_pred: Tensor of predicted targets. Returns: A `Tensor` with loss. Raises: TypeError: If `y_true` cannot be cast to the `y_pred.dtype`. rrr) r'rrrrcliprr0rrrs rrr sN ! !& )F WWVV\\ *F \\&'//"3Q 7F \\&'//"3Q 7F =="''++fvo">>R HHrzkeras.metrics.poissonzkeras.losses.poissonc  tj|}tj||j}t j ||tj j|t jzzz dS)a_Computes the Poisson loss between y_true and y_pred. The Poisson loss is the mean of the elements of the `Tensor` `y_pred - y_true * log(y_pred)`. Standalone usage: >>> y_true = np.random.randint(0, 2, size=(2, 3)) >>> y_pred = np.random.random(size=(2, 3)) >>> loss = tf.keras.losses.poisson(y_true, y_pred) >>> assert loss.shape == (2,) >>> y_pred = y_pred + 1e-7 >>> assert np.allclose( ... loss.numpy(), np.mean(y_pred - y_true * np.log(y_pred), axis=-1), ... atol=1e-5) Args: y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`. y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`. Returns: Poisson loss value. shape = `[batch_size, d0, .. dN-1]`. Raises: InvalidArgumentError: If `y_true` and `y_pred` have incompatible shapes. rr) r'rrrrrrrrrs rrr sa: ! !& )F WWVV\\ *F <<"''++fw/@&@AAA rkeras.losses.cosine_similarity)zkeras.metrics.cosine_proximityzkeras.metrics.cosinezkeras.losses.cosine_proximityzkeras.losses.cosinertctjj||}tjj||}tj||z| S)aQComputes the cosine similarity between labels and predictions. Note that it is a number between -1 and 1. When it is a negative number between -1 and 0, 0 indicates orthogonality and values closer to -1 indicate greater similarity. The values closer to 1 indicate greater dissimilarity. This makes it usable as a loss function in a setting where you try to maximize the proximity between predictions and targets. If either `y_true` or `y_pred` is a zero vector, cosine similarity will be 0 regardless of the proximity between predictions and targets. `loss = -sum(l2_norm(y_true) * l2_norm(y_pred))` Standalone usage: >>> y_true = [[0., 1.], [1., 1.], [1., 1.]] >>> y_pred = [[1., 0.], [1., 1.], [-1., -1.]] >>> loss = tf.keras.losses.cosine_similarity(y_true, y_pred, axis=1) >>> loss.numpy() array([-0., -0.999, 0.999], dtype=float32) Args: y_true: Tensor of true targets. y_pred: Tensor of predicted targets. axis: Axis along which to determine similarity. Returns: Cosine similarity tensor. r)r'linalg l2_normalizer0)r2r3rs rrr sORYY # #F # 6F YY # #F # 6F MM&6/ 5 55rct|txsKt|txr|jtk(xs$t |dxr|j dk(xs|dk(}|S)Nr r)rrrVrYrhasattrr )lossresults ris_categorical_crossentropyr|' sn401 0 t0 1 433 0 D* % < !;; 0 . .  Mrzkeras.losses.serializec|yt|ts"tjdt |d|rt j |St |S)a#Serializes loss function or `Loss` instance. Args: loss: A TF-Keras `Loss` instance or a loss function. use_legacy_format: Boolean, whether to use the legacy serialization format. Defaults to `False`. Returns: Loss configuration dictionary. NzzThe `keras.losses.serialize()` API should only be used for objects of type `keras.losses.Loss`. Found an instance of type z+, which may lead to improper serialization.)rrrQrRrPlegacy_serializationr)rzuse_legacy_formats r serializer7 s[ | dD !  NDzlE G #::4@@ !$ ''rzkeras.losses.deserializecv|r!tj|t|dSt|t|dS)aDeserializes a serialized loss class/function instance. Args: name: Loss configuration. custom_objects: Optional dictionary mapping names (strings) to custom objects (classes and functions) to be considered during deserialization. use_legacy_format: Boolean, whether to use the legacy serialization format. Defaults to `False`. Returns: A TF-Keras `Loss` instance or a loss function. z loss function)module_objectscustom_objectsprintable_module_name)r~rglobals)rrrs r deserializerP sE#<< "9)"1   $ y%-  rzkeras.losses.getc|yt|trt|}d|v}t||St|tr t|St |r|St d|)aRetrieves a TF-Keras loss as a `function`/`Loss` class instance. The `identifier` may be the string name of a loss function or `Loss` class. >>> loss = tf.keras.losses.get("categorical_crossentropy") >>> type(loss) >>> loss = tf.keras.losses.get("CategoricalCrossentropy") >>> type(loss) You can also specify `config` of the loss to this function by passing dict containing `class_name` and `config` as an identifier. Also note that the `class_name` must map to a `Loss` class >>> identifier = {"class_name": "CategoricalCrossentropy", ... "config": {"from_logits": True}} >>> loss = tf.keras.losses.get(identifier) >>> type(loss) Args: identifier: A loss identifier. One of None or string name of a loss function/class or loss configuration dictionary or a loss function or a loss class instance. Returns: A TF-Keras loss as a `function`/ `Loss` class instance. Raises: ValueError: If `identifier` cannot be interpreted. Nmodule)rz.Could not interpret loss function identifier: )rstrrrgcallablerK) identifierrs rrnrnn stD*c"_ $J6:9JKK*d#:&&   8 E rint32)F)r)Frr)rrFrr)FrN)FrrFrr)r)NF)hrOrQrrQtensorflow.compat.v2compatv2r' tf_keras.srcrtf_keras.src.savingrtf_keras.src.saving.legacyrr~%tf_keras.src.saving.serialization_librrrfrr tensorflow.python.ops.raggedr r tensorflow.python.utilr tensorflow.python.util.tf_exportr tensorflow.tools.docsrrrVrsrzr~rrrrrrrrrrrrrrr*add_dispatch_supportrtr dispatch_for_typesrrr{rrrrrr+rrrrrrrXrr_rrcrrkrrprrrbceBCEmseMSEmaeMAEmapeMAPEmsleMSLEkldKLDkullback_leibler_divergencelogcoshrr|rrrnrTr7sparse_softmax_cross_entropyLABEL_DTYPES_FOR_LOSSESr=rrrs   !! *LJH+'84+9.!"rr#rj="EM$MFM`-.8M*8M/8Mv./:N+:N0:Nz89B "5B :B J89= "5= := @/0i',i'1i'X45jF1jF6jFZ45U 1U 6U p9:HF#6HF;HFV:;U $7U <U p-.I *I /I X"#8@ 8@$8@v)*:H&:H+:Hz-.9L*9L/9Lx$%5B!5B&5Bp$%8C!8C&8Cv)*:H&:H+:Hz"#BM BM$BMJ&% ..M/M:VQr/AIBI"'& ..:/:40"//BJCJ 21 ..///@;R__MN21 .. / F;R__MN  +-IJ../K@#%9:..I/;I<./..-/0-B"r*..!/+!H  ..$</ $../?D$ .. 6/  6J c c c,,t,,t*777c '     &'(((0()*: !-"-bIILL44g#Wr