This Repository refers to the [Third Homework] of the course Advanced Machine Learning (AML) at University Sapienza of Rome 2022/2023.
The first part of the task involves implementing the fundamental building blocks of a transformer architecture. In the other section, however, we applied that architecture to predict human trajectories. In this part also we "played" with the various hyperparameters to analyze changes in the output of the model.
- Carolina Romani
- Giacomo Scarponi
- Francesco Sciarra
- Alessandro Sottile
TF4AML_theory.ipynb
notebook contains exercises regarding the implementation of Transformers fundamental blocks;TF4AML_practice.ipynb
notebook contains exercises for a Computer Vision research application. See paper [Under the Hood of Transformer Networks for Trajectory Forecasting];Data
folder has all the ETH/UCY data used as trajectory forecasting benchmark;transformer
folder includes all the necessary file to build TF and experiment with the benchmark. The code is inspired by the original Hugging code of 2018.