6biDxSrSSKrSSKJs Jr SSKJr SSKJ r \"S/S9"SS\ R55r g) z#Base class for Python-based metricsN) keras_export) base_metricz#keras.metrics.experimental.PyMetric)v1cX^\rSrSrSrS U4SjjrU4SjrS SjrSrSr Sr S r U=r $) PyMetricaH Metric which runs in Python, compiled outside of the TensorFlow graph. Args: name: (Optional) string name of the PyMetric instance. dtype: (Optional) data type of the PyMetric result. **kwargs: Additional layer keywords arguments. Usage of `PyMetric` is generally identical to `keras.metrics.Metric`. It can be used in isolation, or in tandem with the `compile()` API. For more information about the usage of `PyMetric`, see `keras.metrics.Metric`. Unlike regular metrics, `PyMetric` instances are outside-compiled with respect to the TensorFlow graph during training or evaluation. They have access to the same inputs of a standard in-graph metric, but they run in a Python interpreter on the host CPU. Any data stored in a `PyMetric` is located on the main memory of the host CPU, and any TensorFlow ops used in a PyMetric are run eagerly on the host CPU. As a result, `PyMetric` instances are generally not as performant as in-graph metrics, and should only be used in cases where computing the metric inside of the TensorFlow graph is either impossible or prohibitively expensive. **Note:** Due to the use of `tf.py_function`, PyMetrics are incompatible with XLA and therefore TPUs. Methods to be implemented by subclasses: * `update_state()`: Handles updates to internal state variables * `result()`: Computes and returns a scalar value or a dict of scalar values for the metric from the state variables. * `reset_state()`: Computes and returns a scalar value for the metric from the state variables. This subclass implementation is similar to that of `keras.metrics.Metric`, with two notable differences: * Inputs to `update_state()` in a `PyMetric` are eager tensors, and both `update_state()` and `result()` run outside of the TensorFlow graph, executing any TensorFlow ops eagerly. * `reset_state()` is also called at initialization time to initialize the Python state of the metric. * `result()` can only return a single scalar. It does not support returning a dictionary of results like `keras.metrics.Metric`. Example subclass implementation using sklearn's Jaccard Score: ```python from sklearn.metrics import jaccard_score import tensorflow as tf class JaccardScore(tf.keras.metrics.experimental.PyMetric): def __init__(self, name='jaccard_score', **kwargs): super().__init__(name=name, **kwargs) def update_state(self, y_true, y_pred, sample_weight=None): self.jaccard_sum += jaccard_score(y_pred, y_true, average="macro") self.count += 1 def reset_state(self): self.jaccard_sum = 0. self.count = 0. def result(self): return self.jaccard_sum / self.count ``` c J>[TU]"SXS.UD6 UR5 g)N)namedtype)super__init__ reset_state)selfr r kwargs __class__s X/home/james-whalen/.local/lib/python3.13/site-packages/tf_keras/src/metrics/py_metric.pyrPyMetric.__init__as$ :d:6: c6>^^^ [[RU]U5mTRm SU 4SjjnUTlSSjn[ R"UT5TlTRmU4SjnUTl U4Sjn[ R"UT5TlT$)Ncx>[R"S5 T"XU5sSSS5 $!,(df  g=fNz/cpu:0tfdevice)y_truey_pred sample_weightobj_update_states rupdate_state_on_cpu-PyMetric.__new__..update_state_on_cpuks%8$' F%$$s + 9cpX/nUbURU5 [R"URU/S9$)N)funcinpTout)appendr py_functionr )rrrr eager_inputss rupdate_state_fn)PyMetric.__new__..update_state_fnqs<"+L(##M2>>--[R"S5 T"5sSSS5 $!,(df  g=frr) obj_resultsrresult_on_host_cpu,PyMetric.__new__..result_on_host_cpu~s 8$!|%$$s) 7cX>[R"UR/TRS9$)N)r$r%)rr'r-r )robjs r result_fn#PyMetric.__new__..result_fns%>>''Rcii rN) r rMetric__new__ update_stater types MethodTyperesultr-) clsargsrr r)r-r1r0r,rrs @@@rr5PyMetric.__new__esK&&4S9++ G#6 !++OSAZZ  $"4  %%i5  rc[S5e)aAccumulates statistics for the metric. **Note:** This function is executed outside of the TensorFlow graph on the CPU host. This means: a) Inputs are eager tensors. b) Any TensorFlow ops run in this method are run eagerly. c) Any Tensors created are allocated to the CPU's main memory. Args: y_true: Target output y_pred: Predicted output sample_weight: (Optional) weights for the individual samples in `y_true` and `y_pred` z*Subclasses should implement `update_state`NotImplementedError)rrrrs rr6PyMetric.update_states$""NOOrc[S5e)zMerges the state from one or more metrics. `PyMetric` instances that intend to support merging state must override this method, as the default implementation in `keras.metrics.Metric` does not apply to `PyMetric`. z)Subclasses should implement `merge_state`r>)rmetricss r merge_statePyMetric.merge_states""MNNrc[S5e)zResets all of the metric state variables. This function is called between epochs when a metric is evaluated during training. It's also called when the metric is initialized. z)Subclasses should implement `reset_state`r>rs rrPyMetric.reset_states ""MNNrc[S5e)aYComputes and returns the scalar metric value. **Note:** This function is executed outside of the TensorFlow graph on the CPU host. This means any TensorFlow ops run in this method are run eagerly. Result computation is an idempotent operation that simply calculates the metric value using the state variables. Returns: A Python scalar. z$Subclasses should implement `result`r>rFs rr9PyMetric.results""HIIrr )NNr3) __name__ __module__ __qualname____firstlineno____doc__rr5r6rCrr9__static_attributes__ __classcell__)rs@rrrs3DL&PP(OO J Jrr) rNr7tensorflow.compat.v2compatv2r tensorflow.python.util.tf_exportrtf_keras.src.metricsrr4rr rrrVsH* !!9,3;eJ{!!eJ<eJr