1014 learning rate finder (#1454)

learning rate finder and corresponding test

Author: rijobro

monai/data/dataset.py

@@ -498,7 +498,13 @@ def _fill_cache(self) -> List:
             warnings.warn("tqdm is not installed, will not show the caching progress bar.")
         with ThreadPool(self.num_workers) as p:
             if has_tqdm:
-                return list(tqdm(p.imap(self._load_cache_item, range(self.cache_num)), total=self.cache_num))
+                return list(
+                    tqdm(
+                        p.imap(self._load_cache_item, range(self.cache_num)),
+                        total=self.cache_num,
+                        desc="Loading dataset",
+                    )
+                )
             return list(p.imap(self._load_cache_item, range(self.cache_num)))
 
     def _load_cache_item(self, idx: int):

monai/handlers/stats_handler.py

@@ -27,7 +27,7 @@
 DEFAULT_TAG = "Loss"
 
 
-class StatsHandler(object):
+class StatsHandler:
     """
     StatsHandler defines a set of Ignite Event-handlers for all the log printing logics.
     It's can be used for any Ignite Engine(trainer, validator and evaluator).

monai/optimizers/__init__.py

@@ -9,5 +9,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from .lr_finder import LearningRateFinder
 from .novograd import Novograd
 from .utils import generate_param_groups