Apply suggestions from code review

Author: vfdev-5

examples/contrib/detection/README.md

@@ -14,6 +14,7 @@ In this example, we show how to use _Ignite_ to train a image detection model wi
 - [aim](https://github.com/aimhubio/aim): `pip install aim`
 - [albumentations](https://github.com/albumentations-team/albumentations): `pip install albumentations`
 - [pycocotools](https://cocodataset.org/): `pip install pycocotools`
+- [python-fire](https://google.github.io/python-fire/guide/): `pip install fire`
 
 Alternatively, install the all requirements using `pip install -r requirements.txt`.
 

examples/contrib/detection/main.py

@@ -161,7 +161,7 @@ def update_model(engine, batch):
         scaler.update()
 
         loss_items = {k: v.item() for k, v in loss_dict.items()}
-        loss_items["loss_average"] = loss.item() / 4
+        loss_items["loss_average"] = loss.item() / len(loss_dict)
 
         return loss_items