dedellzyw comments

Repositories
Issues
Comments

Results 2 comments of


                                            dedellzyw

End-End Training

> The default.py training configuration says (Line 102 - 112): > > ``` > # if training 3 tasks end-to-end, set all parameters as True > # Alternating optimization >...

this is my loss: no smoothing: ![image](https://user-images.githubusercontent.com/31383717/173296291-d3658cfb-fc76-44ae-b95c-d42a8e71b8a8.png) smoothing: ![image](https://user-images.githubusercontent.com/31383717/173296387-32bcff85-8901-4458-8ee0-27f37e87db91.png)