Need help with social-lstm?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

219 Stars 172 Forks 23 Commits 16 Opened issues


Social LSTM implementation in PyTorch

Services available


Need anything else?

Contributors list

# 163,373
20 commits
# 362,132
1 commit

Social LSTM implementation in PyTorch

Project details

Semester project of Master of Computer Science in EPFL
Student name : Baran Nama
Advisor: Alexandre Alahi
Presentation :

Implementation details

Baseline implementation:
Detailed info about challange and datasets: Made improvements: Please see attached presentation


  • : Python script for generating artifical datasets
  • Python script includes various helper methods
  • Pyton script for random best parameter selection for a model
  • Bash script for creation of file structure
  • Python file includes Social LSTM model definition
  • Python file includes Occupancy LSTM model definition
  • Python script for training Occupancy LSTM model
  • Python script for model testing and getting output txt file for submission
  • Python script for training Social LSTM model
  • Python script for handling input train/test/validation data and batching it
  • Python script for externally evaluate a trained model by getting validation error
  • Python script for visualizing predicted trajectories during train/test/validation sessions
  • Python file includes Vanilla LSTM model definition
  • vlstm_train: Python script for training Vanilla LSTM model

How to deploy

  1. Fork the repository
  2. Start train a model >>> python train/olstm_train/ - -[Parameter set]
  3. If necesarry file structure is not exist (which is the initial situation), train script will run and this command will automatically create file structure
  4. Enjoy!


| Model name | Avarage error | Final error | Mean error | | :---: | :---: | :---: | :---: | | Social LSTM | 1.3865 | 2.098 | 0.675 | | Occupancy LSTM | 2.1105 | 3.12 | 1.101 | | Vanilla LSTM | 2.107 | 3.114 | 1.1 |


We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.