Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
alpamayo-vision-v3 – AI Model by crypticvandal | AlphaNeural AI
You can deploy this model and start earning money today!
crypticvandal
/
alpamayo-vision-v3
like
0
tf-keras
autonomous-driving
vision-model
trajectory-prediction
tensorflow
nvidia-physical-ai
apache-2.0
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
Alpamayo Vision Model v3
Overview
Vision model trained on NVIDIA Physical AI Autonomous Vehicles dataset for trajectory prediction in autonomous driving scenarios.
Training Details
Dataset
: nvidia/PhysicalAI-Autonomous-Vehicles (streamed)
Training steps
: 1000
Batch size
: 4
Architecture
: CNN (4 conv layers) + GlobalAvgPool + Dense + 3 heads
Input
: 5-frame sequence, 256x256 RGB, camera_front_wide_120fov
Output
: trajectory (20x2), lane center offset, heading angle
Optimizer
: Adam (lr=1e-4)
Best trajectory error
: 15.74
Usage
