This is the repo for the challenges of the Artificial Neural Networks and Deep Learning course at Polytechnic University of Milan, 2020/2021.
There are 3 kaggle competitions based on 3 different tasks that cover the most important topics learned during the course. We decided to use Google Colab to exploit the powerful GPU offered by Google. The main library used for building the neural network architectures is TensorFlow as explained throughout the course.
The goal of the challenge is to classify images of people wearing masks ( thanks Covid-19 ). The different labels are described below.
- All the people in the image are wearing a mask
- No person in the image is wearing a mask
- Someone in the image is not wearing a mask.
| Model from scratch | Transfer learning model |
|---|---|
| 0.895 | 0.955 |
Top 15% out of 192 participants. 🎉
Our main ideas and their implementation can be found here
The kaggle competition can be found here
The goal of the challenge is to perform precise automatic crop and weed segmentation for the agricoltural sector.
| Input Image | Expected segmentation |
|---|---|
 |
 |
Our main ideas and their implementation can be found here
The codalab competition can be found here
The goal of the challenge is to answer questions using the information provided by the corresponding image and question pair.
Q: Is the man's shirt blue? A: yes
Our main ideas and their implementation can be found here
The Kaggle competition can be found here
Top 9% out of 146 participants. 🎉