Add dataset description.
Browse filesIncluding data generation details.
README.md
CHANGED
|
@@ -51,3 +51,20 @@ configs:
|
|
| 51 |
- split: train
|
| 52 |
path: data/train-*
|
| 53 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
- split: train
|
| 52 |
path: data/train-*
|
| 53 |
---
|
| 54 |
+
|
| 55 |
+
### TaxisVIS1M
|
| 56 |
+
|
| 57 |
+
TaxisVIS1M is a subset of New York City’s 2015 Yellow Taxi trip records containing 1 million trips. It is sampled from the original TLC Trip Record Data to enable fast processing and visualization while preserving the temporal and spatial patterns of taxi mobility across the city.
|
| 58 |
+
|
| 59 |
+
- Source Data: NYC Taxi & Limousine Commission (TLC) [2015 Yellow Taxi Trip Records](https://data.cityofnewyork.us/Transportation/2015-Green-Taxi-Trip-Data/gi8d-wdg5/about_data) (TLC Trip Record Data
|
| 60 |
+
).
|
| 61 |
+
|
| 62 |
+
- Preprocessing:
|
| 63 |
+
|
| 64 |
+
- Download the original Parquet files from TLC.
|
| 65 |
+
|
| 66 |
+
- Load the data using pandas/geopandas.
|
| 67 |
+
|
| 68 |
+
- Slice the first 1 million rows for mana geable processing.
|
| 69 |
+
|
| 70 |
+
- Export the subset as CSV for use in urban analysis pipelines.
|