Expand evaluation doc (#1396)

Author: alextrott16

docs/source/trainer/evaluation.rst

@@ -17,6 +17,26 @@ specified by the the :class:`.Trainer` parameter ``eval_interval``.
     )
 
 The metrics should be provided by :meth:`.ComposerModel.metrics`.
+For more information, see the "Metrics" section in :doc:`/composer_model`.
+
+To provide a deeper intuition, here's pseudocode for the evaluation logic that occurs every ``eval_interval``:
+
+.. code:: python
+
+    metrics = model.metrics(train=False)
+
+    for batch in eval_dataloader:
+        outputs, targets = model.validate(batch)
+        metrics.update(outputs, targets)  # implements the torchmetrics interface
+
+    metrics.compute()
+
+- The trainer iterates over ``eval_dataloader`` and passes each batch to the model's :meth:`.ComposerModel.validate` method.
+- Outputs of ``model.validate`` are used to update ``metrics`` (a :class:`torchmetrics.Metric` or :class:`torchmetrics.MetricCollection` returned by :meth:`.ComposerModel.metrics <model.metrics(train=False)>`).
+- Finally, metrics over the whole validation dataset are computed.
+
+Note that the tuple returned by :meth:`.ComposerModel.validate` provide the positional arguments to ``metrics.update``.
+Please keep this in mind when using custom models and/or metrics.
 
 Multiple Datasets
 -----------------
@@ -25,7 +45,7 @@ If there are multiple validation datasets that may have different metrics,
 use :class:`.Evaluator` to specify each pair of dataloader and metrics.
 This class is just a container for a few attributes:
 
-- ``label``: a user-specified name for the metric.
+- ``label``: a user-specified name for the evaluator.
 - ``dataloader``: PyTorch :class:`~torch.utils.data.DataLoader` or our :class:`.DataSpec`.
     See :doc:`DataLoaders</trainer/dataloaders>` for more details.
 - ``metric_names``: list of names of metrics to track.
@@ -55,3 +75,5 @@ can be specified as in the following example:
         eval_dataloader=[glue_mrpc_task, glue_mnli_task],
         ...
     )
+
+Note that `metric_names` must be a subset of the metrics provided by the model in :meth:`.ComposerModel.metrics`.