Computer Vision Task:
Build a YOLO model for document Analysis.
Features:
- Use streamlit for UI
- Users can upload pdf or images (PDF should be converted to images before passing to the model)
- The model should process tables , figures and text regions.
- Process the result and crop out those regions and save them as tables.png , figures.png , text.png
- If multiple use loops and index them (for example if 2 tables then tables_1.png and tables_2.png)
NLP task:
Build a chatbot creating agent using NLP.
Features:
- Use streamlit for UI
- Users can fill forms to give their information and choose number of epochs
- While training show progress bar UI
- After training a chatbot should appear where users will chat to their trained chatbot (The output should have voice like taught in class)
Please submit your task code with a video demo of the project on the forum and you can mail it to tejashkatuwal99@gmail.com.