ResNet Model trained on ImageNet
The ResNet50 model pretrained on imagenet for TensorFlow.js as a layers model.
On ImageNet, this model gets to a top-1 validation accuracy of 0.749 and a top-5 validation accuracy of 0.921.
The default input size for this model is 224x224.
This model has been converted, using the tfjs-converter.
The base model and weights were taken from keras.
Install npm install resnet_imagenet
How to use
; const tabbyCatURI = 'https://upload.wikimedia.org/wikipedia/commons/6/66/An_up-close_picture_of_a_curious_male_domestic_shorthair_tabby_cat.jpg'; const run = async { const predictor = await ResNetPredictor; const prediction = await predictor; return prediction;}
To try the model you can just load it using:
ResNetURL = 'https://raw.githubusercontent.com/paulsp94/tfjs_resnet_imagenet/master/ResNet50/model.json';const ResNet = await tf;