Detector Recipes: Supported Model Architectures
The following table presents the Detector Recipes that we currently support. The modular nature of LEIP recipes makes it easy to evaluate different architectures for speed and accuracy on your target hardware to find the model that best meets your application requirements.
Model | Default Size | recipe_name | architecture_name |
---|---|---|---|
YOLOv5 Large | 640 |
|
|
YOLOv5 Medium | 640 |
|
|
YOLOv5 Small | 640 |
|
|
Mobilenet V1 SSD | 300 |
|
|
Mobilenet V2 Lite SSD | 300 |
|
|
EfficientDet D0 | 512 |
|
|
TF EfficientDet D0 ** | 512 |
|
|
EfficientDet Q0 | 512 |
|
|
TF EfficientDet Lite3 | 512 |
|
|
EfficientDet D1 | 640 |
|
|
TF EfficientDet D1 ** | 640 |
|
|
EfficientDet Q1 | 640 |
|
|
TF EfficientDet Lite4 | 640 |
|
|
EfficientDet V2 DT | 768 |
|
|
EfficientDet Q2 | 768 |
|
|
TF EfficientDet D2 ** | 768 |
|
|
ResDet 50 | 640 |
|
|
** Known issue: The tf_effcientdet_d# recipes will currently not calibrate for Int8 optimization on Nvidia Ampere devices. If you want to run these models on Ampere devices, you will need to create the .activations
files on a pre-Ampere device.
Please contact Latent AI if you require larger to smaller models.