{"id":22919792,"url":"https://github.com/stanfordasl/trajectron","last_synced_at":"2025-05-12T20:19:11.698Z","repository":{"id":44383786,"uuid":"148928165","full_name":"StanfordASL/Trajectron","owner":"StanfordASL","description":"Code accompanying \"The Trajectron: Probabilistic Multi-Agent Trajectory Modeling with Dynamic Spatiotemporal Graphs\" by Boris Ivanovic and Marco Pavone.","archived":false,"fork":false,"pushed_at":"2020-12-09T20:12:29.000Z","size":136084,"stargazers_count":121,"open_issues_count":1,"forks_count":40,"subscribers_count":10,"default_branch":"master","last_synced_at":"2023-12-19T13:20:41.580Z","etag":null,"topics":["deep-learning","human-robot-interaction","human-trajectory-prediction"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/StanfordASL.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2018-09-15T18:17:48.000Z","updated_at":"2023-11-24T19:43:16.000Z","dependencies_parsed_at":"2022-07-14T14:01:03.098Z","dependency_job_id":null,"html_url":"https://github.com/StanfordASL/Trajectron","commit_stats":null,"previous_names":[],"tags_count":0,"template":null,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/StanfordASL%2FTrajectron","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/StanfordASL%2FTrajectron/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/StanfordASL%2FTrajectron/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/StanfordASL%2FTrajectron/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/StanfordASL","download_url":"https://codeload.github.com/StanfordASL/Trajectron/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":229673816,"owners_count":18105435,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["deep-learning","human-robot-interaction","human-trajectory-prediction"],"created_at":"2024-12-14T07:13:17.989Z","updated_at":"2024-12-14T07:13:18.547Z","avatar_url":"https://github.com/StanfordASL.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"**NOTE:** A new version of the Trajectron has been released! Check out [Trajectron++](https://github.com/StanfordASL/Trajectron-plus-plus)!\n\n\u003cp align=\"center\"\u003e\u003cimg width=\"100%\" src=\"img/Trajectron.png\"/\u003e\u003c/p\u003e\n\n# The Trajectron: Probabilistic Multi-Agent Trajectory Modeling with Dynamic Spatiotemporal Graphs\n\nThis repository contains the code for [The Trajectron: Probabilistic Multi-Agent Trajectory Modeling with Dynamic Spatiotemporal Graphs](https://arxiv.org/abs/1810.05993) by Boris Ivanovic and Marco Pavone.\n\n## Installation ##\n\nFirst, we'll create a conda environment to hold the dependencies.\n```\nconda create --name dynstg python=3.6 -y\nsource activate dynstg\npip install -r requirements.txt\n```\n\nThen, since this project uses IPython notebooks, we'll install this conda environment as a kernel.\n```\npython -m ipykernel install --user --name dynstg --display-name \"Python 3.6 (DynSTG)\"\n```\n\nNow, you can start a Jupyter session and view/run all the notebooks with\n```\njupyter notebook\n```\n\nWhen you're done, don't forget to deactivate the conda environment with\n```\nsource deactivate\n```\n\n## Scripts ##\nRun any of these with a `-h` or `--help` flag to see all available command arguments.\n* `code/train.py` - Trains a new Trajectron.\n* `code/test_online.py` - Replays a scene from a dataset and performs online inference with a trained Trajectron.\n* `code/evaluate_alongside_sgan.py` - Evaluates the performance of the Trajectron against Social GAN. This script mainly collects evaluation data, which can be visualized with `sgan-dataset/Result Analyses.ipynb`.\n* `code/compare_runtimes.py` - Evaluates the runtime of the Trajectron against Social GAN. This script mainly collects runtime data, which can be visualized with `sgan-dataset/Runtime Analysis.ipynb`.\n* `sgan-dataset/Qualitative Plots.ipynb` - Can be used to visualize predictions from the Trajectron alone, or against those from Social GAN.\n\n## Datasets ##\n\nThe preprocessed datasets are available in this repository, under `data/` folders (i.e. `sgan-dataset/data/`).\n\nIf you want the *original* ETH or UCY datasets, you can find them here: [ETH Dataset](http://www.vision.ee.ethz.ch/en/datasets/) and [UCY Dataset](https://graphics.cs.ucy.ac.cy/research/downloads/crowd-data).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fstanfordasl%2Ftrajectron","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fstanfordasl%2Ftrajectron","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fstanfordasl%2Ftrajectron/lists"}