Skip to content

Getting Started

1. Access the Data

Data type Source
Raw data (AVL, AFC, GTFS, LTI) Mendeley Data
Processed data (Boarding, Alighting, OD) Hugging Face
Code & models GitHub

2. Clone the Repository

git clone https://github.com/LabIA-UFBA/SUNT.git
cd SUNT

3. Install Dependencies

pip install -r requirements.txt

4. Load Processed Data (Hugging Face)

from datasets import load_dataset

boarding  = load_dataset("labiaufba/PublicTransportationSunt", "boarding")
alighting = load_dataset("labiaufba/PublicTransportationSunt", "alighting")
od        = load_dataset("labiaufba/PublicTransportationSunt", "od")

5. Load Data Locally (Pandas)

import pandas as pd

avl_lines    = pd.read_csv("data/raw/avl_lines.csv")
avl_vehicles = pd.read_csv("data/raw/avl_vehicles.csv")
afc          = pd.read_csv("data/raw/afc.csv")
lti          = pd.read_csv("data/raw/lti.csv")

boarding  = pd.read_csv("data/processed/boarding.csv")
alighting = pd.read_csv("data/processed/alighting.csv")
od        = pd.read_csv("data/processed/od.csv")

6. Load the Graph

import pandas as pd
import networkx as nx

nodes = pd.read_csv("data/graph/nodes.csv")
edges = pd.read_csv("data/graph/edges.csv")

G = nx.from_pandas_edgelist(
    edges,
    source="src",
    target="dst",
    edge_attr=["distance", "average_speed", "trip_time", "loading"],
    create_using=nx.DiGraph()
)

7. Explore the Sample Notebook

jupyter notebook docs/dataloader_sample.ipynb

Next Steps