RE: https://mastodon.social/@xlth/116144192667591833
Not the ideal conditions for geospatial applications of VLMs π
#GIScience #VLM #spatiotemporal #MobilityDataScience #SpatialDataScience
RE: https://mastodon.social/@xlth/116144192667591833
Not the ideal conditions for geospatial applications of VLMs π
#GIScience #VLM #spatiotemporal #MobilityDataScience #SpatialDataScience
The #MobilityDB team just announced v1.3.0-alpha featuring
New temporal types:
π tgeometry & tgeography that can represent the temporal evolution of any geometry type (polygon, multipoint, etc.)
π temporal circular buffer (tcbuffer)
π temporal pose (tpose) type, storing the evolution of a pose
(point position + orientation)
and more ...
https://github.com/MobilityDB/MobilityDB/releases/tag/v1.3.0-alpha
If you're at #AGIT2025, I'd love to see you at our #MobilityDataScience workshop focused on #PublicTransport tomorrow morning π
09:00-10:15 Session 1A: WS: Mobility Data Science & KI β Potenziale fΓΌr den ΓΆffentlichen Verkehr
π€© #MobilityDataAnalytics & #GIScience all around:
Attending the @emeraldseu GA today. Presented progress on #Trajectools and our #explainableAI & #activeLearning for #MobilityDataScience. While simultaneously traveling to #AGIT2025 π
π New @movingpandas 0.22 and #Trajectools 2.6 just landed
Over 50% runtime reduction for many #MovementDataAnalytics tasks π
The latest releases of MovingPandas and Trajectools come with many βunder the hoodβ changes that aim to make your movement analytics faster: Instead of immediately creating a GeoPandas β¦
And here's @martinfleis et al.'s new street network generalization tool:
Another interesting new paper on street network analysis, HT @gboeing
π― it's surprisingly hard to count intersections
Direct links for convenience ...
π Paper: https://www.mdpi.com/2071-1050/17/8/3634
Code: https://github.com/plus-mobilitylab/netascore/tree/v0.9.0
#Walkability #HumanMobility #MobilityDataScience #GIScience #GISchat #OSM
π₯³ New versions of @movingpandas and #Trajectools have just landed:
π’ππ²π¦ #MovingPandas 0.21.3 https://github.com/movingpandas/movingpandas/releases/tag/v0.21.3
Trajectools 2.5 https://codeberg.org/movingpandas/trajectools/releases/tag/v2.5
π Please update your MovingPandas install to get all the improvements in #QGIS Trajectools https://plugins.qgis.org/plugins/processing_trajectory/#plugin-versions
#GISChat #MovementDataAnalytics #MobilityDataScience #Python
The quest for a fair TimeGPT benchmark
At the end of yesterday's #TimeGPT for mobility post, we concluded that TimeGPT's trainingset probably included a copy of the popular BikeNYC timeseries dataset and that, therefore, we were not looking at a fair comparison ...
http://anitagraser.com/2025/03/29/the-quest-for-a-fair-timegpt-benchmark/