updated callback readme

luxonis_train/callbacks/README.md

@@ -9,6 +9,7 @@ List of all supported callbacks.
 - [LuxonisProgressBar](#luxonisprogressbar)
 - [MetadataLogger](#metadatalogger)
 - [TestOnTrainEnd](#testontrainend)
+- [UploadCheckpoint](#uploadcheckpoint)
 
 ## PytorchLightning Callbacks
 
@@ -51,3 +52,11 @@ Metadata include all defined hyperparameters together with git hashes of `luxoni
 ## TestOnTrainEnd
 
 Callback to perform a test run at the end of the training.
+
+## UploadCheckpoint
+
+Callback that uploads currently best checkpoint (based on validation loss) to specified cloud directory after every validation epoch.
+
+| Key         | Type        | Default value | Description                                                                                                             |
+| ----------- | ----------- | ------------- | ----------------------------------------------------------------------------------------------------------------------- |
+| upload_directory | str | /          | Path to cloud directory where checkpoints should be uploaded to. If you want to use current mlflow run set it to `mlflow://`. |

luxonis_train/callbacks/__init__.py

@@ -13,7 +13,7 @@
 from .metadata_logger import MetadataLogger
 from .module_freezer import ModuleFreezer
 from .test_on_train_end import TestOnTrainEnd
-from .upload_checkpoint_on_train_end import UploadCheckpointOnTrainEnd
+from .upload_checkpoint import UploadCheckpoint
 
 CALLBACKS.register_module(module=EarlyStopping)
 CALLBACKS.register_module(module=LearningRateMonitor)
@@ -28,5 +28,5 @@
     "MetadataLogger",
     "ModuleFreezer",
     "TestOnTrainEnd",
-    "UploadCheckpointOnTrainEnd",
+    "UploadCheckpoint",
 ]