Evaluates given model with provided evaluation data. Stop conditions - we evaluate on the given input data until one of the following: - If `steps` is provided, and `steps` batches of size `batch_size` are processed. - If `input_fn` is provided, and it raises an end-of-input
(self,
x=None,
y=None,
input_fn=None,
feed_fn=None,
batch_size=None,
steps=None,
metrics=None,
name=None,
checkpoint_path=None,
hooks=None)
| 44 | |
| 45 | @abc.abstractmethod |
| 46 | def evaluate(self, |
| 47 | x=None, |
| 48 | y=None, |
| 49 | input_fn=None, |
| 50 | feed_fn=None, |
| 51 | batch_size=None, |
| 52 | steps=None, |
| 53 | metrics=None, |
| 54 | name=None, |
| 55 | checkpoint_path=None, |
| 56 | hooks=None): |
| 57 | """Evaluates given model with provided evaluation data. |
| 58 | |
| 59 | Stop conditions - we evaluate on the given input data until one of the |
| 60 | following: |
| 61 | - If `steps` is provided, and `steps` batches of size `batch_size` are |
| 62 | processed. |
| 63 | - If `input_fn` is provided, and it raises an end-of-input |
| 64 | exception (`OutOfRangeError` or `StopIteration`). |
| 65 | - If `x` is provided, and all items in `x` have been processed. |
| 66 | |
| 67 | The return value is a dict containing the metrics specified in `metrics`, as |
| 68 | well as an entry `global_step` which contains the value of the global step |
| 69 | for which this evaluation was performed. |
| 70 | |
| 71 | Args: |
| 72 | x: Matrix of shape [n_samples, n_features...] or dictionary of many |
| 73 | matrices |
| 74 | containing the input samples for fitting the model. Can be iterator that |
| 75 | returns |
| 76 | arrays of features or dictionary of array of features. If set, |
| 77 | `input_fn` must |
| 78 | be `None`. |
| 79 | y: Vector or matrix [n_samples] or [n_samples, n_outputs] containing the |
| 80 | label values (class labels in classification, real numbers in |
| 81 | regression) or dictionary of multiple vectors/matrices. Can be iterator |
| 82 | that returns array of targets or dictionary of array of targets. If set, |
| 83 | `input_fn` must be `None`. Note: For classification, label values must |
| 84 | be integers representing the class index (i.e. values from 0 to |
| 85 | n_classes-1). |
| 86 | input_fn: Input function returning a tuple of: |
| 87 | features - Dictionary of string feature name to `Tensor` or `Tensor`. |
| 88 | labels - `Tensor` or dictionary of `Tensor` with labels. |
| 89 | If input_fn is set, `x`, `y`, and `batch_size` must be `None`. If |
| 90 | `steps` is not provided, this should raise `OutOfRangeError` or |
| 91 | `StopIteration` after the desired amount of data (e.g., one epoch) has |
| 92 | been provided. See "Stop conditions" above for specifics. |
| 93 | feed_fn: Function creating a feed dict every time it is called. Called |
| 94 | once per iteration. Must be `None` if `input_fn` is provided. |
| 95 | batch_size: minibatch size to use on the input, defaults to first |
| 96 | dimension of `x`, if specified. Must be `None` if `input_fn` is |
| 97 | provided. |
| 98 | steps: Number of steps for which to evaluate model. If `None`, evaluate |
| 99 | until `x` is consumed or `input_fn` raises an end-of-input exception. |
| 100 | See "Stop conditions" above for specifics. |
| 101 | metrics: Dict of metrics to run. If None, the default metric functions |
| 102 | are used; if {}, no metrics are used. Otherwise, `metrics` should map |
| 103 | friendly names for the metric to a `MetricSpec` object defining which |
no outgoing calls