File size: 2,463 Bytes
9df973e
 
 
 
 
 
 
 
 
63daf21
9df973e
 
 
 
 
 
 
 
 
 
 
 
 
3551c66
9df973e
 
 
 
1f6a895
 
 
 
 
 
9df973e
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
# Emojich
![](./pics/emojich_rgba_100.png)
### generate emojis from text

Model was trained by [Sber AI](https://github.com/sberbank-ai)
* Task: `text2image generation`
* Num Parameters: `1.3 B`
* Training Data Volume: `120 million text-image pairs` & [`2749 text-emoji pairs`](https://www.kaggle.com/shonenkov/russian-emoji)

[![Telegram](https://img.shields.io/badge/Telegram-Stickers-blue?style=for-the-badge&logo=data:image/svg%2bxml;base64,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)](https://telegram.me/addstickers/SberAI_ruDALLE)

### Model Description
😋 Emojich is a 1.3 billion params model from the family GPT3-like, it generates emoji-style images with the brain of ◾ Malevich.


### Fine-tuning stage:

The main goal of fine-tuning is trying to keep the generalization of [ruDALL-E Malevich (XL)](https://huggingface.co/sberbank-ai/rudalle-Malevich)
model on text to emoji tasks. ruDALL-E Malevich is a multi-modality big pretrained transformer, that uses images and texts.
The idea with freezing feedforward and self-attention layers in pretrained transformer is demonstrated high performance in changing different modalities.
Also, the model has a good chance for over-fitting text modality and lost generalization. 
To deal with this problem is increased coefficient 10^3 in weighted cross-entropy loss for image codebooks part. 

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)


### Examples of generated emojis

All examples are generated automatically (without manual cherry-picking) with hyper-parameters: 
seed 42, batch size 16, top-k 2048, top-p 0.995, temperature 1.0, GPU A100.
For making better generative emojis should use more attempts (~512) and select the best one manually.

*Remember, the great art makers became "great" after creating just only one masterpiece.*

![](./pics/examples.png)