shonenkov
commited on
Commit
•
3551c66
1
Parent(s):
63daf21
update readme
Browse files
README.md
CHANGED
@@ -21,7 +21,7 @@ The idea with freezing feedforward and self-attention layers in pretrained trans
|
|
21 |
Also, the model has a good chance for over-fitting text modality and lost generalization.
|
22 |
To deal with this problem is increased coefficient 10^3 in weighted cross-entropy loss for image codebooks part.
|
23 |
|
24 |
-
Full version of code is available on Kaggle [
|
25 |
|
26 |
|
27 |
### Examples of generated emojis
|
|
|
21 |
Also, the model has a good chance for over-fitting text modality and lost generalization.
|
22 |
To deal with this problem is increased coefficient 10^3 in weighted cross-entropy loss for image codebooks part.
|
23 |
|
24 |
+
Full version of training code is available on Kaggle: [![Kaggle](https://kaggle.com/static/images/open-in-kaggle.svg)](https://www.kaggle.com/shonenkov/emojich-rudall-e)
|
25 |
|
26 |
|
27 |
### Examples of generated emojis
|