SimpleNet
Lets Keep it simple, Using simple architectures to outperform deeper and more complex architectures
Docs
Access comprehensive developer documentation for PyTorch
Scriptable
Wide ResNet
Wide Residual Networks
vgg-nets
Award winning ConvNets from 2014 ImageNet ILSVRC challenge
SqueezeNet
Alexnet-level accuracy with 50x fewer parameters.
SNNMLP
Brain-inspired Multilayer Perceptron with Spiking Neurons
ShuffleNet v2
An efficient ConvNet optimized for speed and memory, pre-trained on ImageNet
ResNext
Next generation ResNets, more efficient and accurate
ResNet
Deep residual networks pre-trained on ImageNet
MobileNet v2
Efficient networks optimized for speed and memory, with residual blocks
Inception_v3
Also called GoogleNetv3, a famous ConvNet trained on ImageNet from 2015
GhostNet
Efficient networks by generating more features from cheap operations
FCN
Fully-Convolutional Network model with ResNet-50 and ResNet-101 backbones
Densenet
Dense Convolutional Network (DenseNet), connects each layer to every other layer in a feed-forward fashion.
AlexNet
The 2012 ImageNet winner achieved a top-5 error of 15.3%, more than 10.8 percentage points lower than that of the runner up.
Deeplabv3
DeepLabV3 models with ResNet-50, ResNet-101 and MobileNet-V3 backbones
YOLOv5
Ultralytics YOLOv5 🚀 for object detection, instance segmentation and image classification.
Silero Voice Activity Detector
Pre-trained Voice Activity Detector
Silero Text-To-Speech Models
A set of compact enterprise-grade pre-trained TTS Models for multiple languages
Silero Speech-To-Text Models
A set of compact enterprise-grade pre-trained STT Models for multiple languages.
Open-Unmix
Reference implementation for music source separation