Fix min/max logging default value (#11310)

Co-authored-by: Carlos Mocholi <[email protected]>

Author: 2 people

CHANGELOG.md

@@ -10,6 +10,7 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 ### Fixed
 
 - Fixed `LightningCLI` race condition while saving the config ([#11199](https://github.com/PyTorchLightning/pytorch-lightning/pull/11199))
+- Fixed the default value used with `log(reduce_fx=min|max)` ([#11310](https://github.com/PyTorchLightning/pytorch-lightning/pull/11310))
 - Fixed data fetcher selection ([#11294](https://github.com/PyTorchLightning/pytorch-lightning/pull/11294))
 - Fixed a race condition that could result in incorrect (zero) values being observed in prediction writer callbacks ([#11288](https://github.com/PyTorchLightning/pytorch-lightning/pull/11288))
 - Fixed dataloaders not getting reloaded the correct amount of times when setting `reload_dataloaders_every_n_epochs` and `check_val_every_n_epoch` ([#10948](https://github.com/PyTorchLightning/pytorch-lightning/pull/10948))

pytorch_lightning/trainer/connectors/logger_connector/result.py

@@ -208,8 +208,14 @@ def __init__(self, metadata: _Metadata, is_tensor: bool) -> None:
         self.meta = metadata
         self.has_reset = False
         if is_tensor:
+            if metadata.is_max_reduction:
+                default = float("-inf")
+            elif metadata.is_min_reduction:
+                default = float("inf")
+            else:
+                default = 0.0
             # do not set a dtype in case the default dtype was changed
-            self.add_state("value", torch.tensor(0.0), dist_reduce_fx=torch.sum)
+            self.add_state("value", torch.tensor(default), dist_reduce_fx=torch.sum)
             if self.meta.is_mean_reduction:
                 self.add_state("cumulated_batch_size", torch.tensor(0), dist_reduce_fx=torch.sum)
 

tests/core/test_metric_result_integration.py

@@ -575,14 +575,14 @@ def test_metric_result_respects_dtype(floating_dtype):
     assert rm.cumulated_batch_size.dtype == fixed_dtype
 
     # two fixed point numbers - should be converted
-    value, batch_size = torch.tensor(2), torch.tensor(3)
+    value, batch_size = torch.tensor(2), 3
     assert value.dtype == fixed_dtype
     with pytest.warns(
         UserWarning, match=rf"`self.log\('bar', ...\)` in your `foo` .* Converting it to {floating_dtype}"
     ):
         rm.update(value, batch_size)
     # floating and fixed
-    rm.update(torch.tensor(4.0), torch.tensor(5))
+    rm.update(torch.tensor(4.0), 5)
 
     total = rm.compute()
 
@@ -591,3 +591,12 @@ def test_metric_result_respects_dtype(floating_dtype):
 
     # restore to avoid impacting other tests
     torch.set_default_dtype(torch.float)
+
+
+@pytest.mark.parametrize(["reduce_fx", "expected"], [(max, -2), (min, 2)])
+def test_result_metric_max_min(reduce_fx, expected):
+    metadata = _Metadata("foo", "bar", reduce_fx=reduce_fx)
+    metadata.sync = _Sync()
+    rm = ResultMetric(metadata, is_tensor=True)
+    rm.update(torch.tensor(expected), 1)
+    assert rm.compute() == expected