YubiAI

State-of-the-art models in AI,ML,NLP & Vision for FinTech community by Yubi's Data Science Team. <br> <br>

<details> <summary>What's new in NLP</summary> <p>

• Oct 2022
  • YubiTokenizer trained on FinTech multilingual data
  • YubiBERT Micro Encoder4
• Nov 2022
  • YubiBERT Small Encoder8
  • HuggingFace Supported YubiTokenizers
• Dec 2022
  • YuLan V1 - Yubi's Text Language Detection
  • TrueCaser Model v1
  • Character-2-Text generation Model v1
• Feb 2023
  • YuLan V2 - Yubi's Text Language Detection
• Jun 2023
  • NSFW Text Detection

</p> </details> </br> <details><summary>What's new in Vision</summary><p>

• January 2023
  • Image Augmentations (Random rotate & croppings)
  • Document Skew Detection v1
• March 2023
  • Fintech Document Segmentation model v1
• April 2023
  • Fintech Document Segmentation model v2
• Jun 2023
  • Document Image Detection
  • NSFW Image Detection

</p></details> </br> <br>

Download Model Locations

• You can download models directly from this google-drive shared folder

Environment

• Models files hosted on FTP server and are downloaded when constructor is called
• Curently tested on Ubuntu 20.04, MacOS 12.3 and python >=3.7

How to install package

• Clone the git repository or download zip/tar files and unzip

cd /parent/directory/path/of/YubiAI/

• Install only NLP dependencies

pip install ".[nlp]"

• Install only Vision dependencies

pip install ".[cv]"

• Install both NLP & Vision dependencies

pip install ".[nlp,cv]"

How to use package

• Clone the git repository or download zip/tar files and unzip
• You need to append the repo path using sys
• Remaining import & code would remain same.

import sys
sys.path.append("/parent/directory/path/of/yubiai/")

from yubiai.nlp.tokenizer.yubiTokenizer import YubiTokenizer
tokenizer = YubiTokenizer()

<br>