issues
search
rhysdg
/
vision-at-a-clip
Low-latency ONNX and TensorRT based zero-shot classification and detection with contrastive language-image pre-training based prompts
16
stars
1
forks
source link
Feat - example notebook
#3
Closed
rhysdg
closed
3 months ago
rhysdg
commented
3 months ago
Adding an example Notebook with installation and basic inference
Environment
Ubuntu 22.04 - RTX 3080, 8-core
Incoming Changes
:
Numpy only SigLIP tokenization
Environment
Incoming Changes :