![nn top 100 models nn top 100 models](https://0.s3.envato.com/files/121847239/pv_590.jpg)
ImageNet are also often accurate when you apply them to other natural image data sets Measure the accuracy of networks trained on ImageNet. The classification accuracy on the ImageNet validation set is the most common way to The area of each marker is proportional to the size of Tesla ® P100) and a mini-batch size of 128. The plot displays the classificationĪccuracy versus the prediction time when using a modern GPU (an NVIDIA ® The exact prediction and training iteration times depend on the hardwareĪ good network has a high accuracy and is fast.
![nn top 100 models nn top 100 models](https://s3.amazonaws.com/media.ai-ap.com/dam/cropped/2017/02/22/dsc00323-ps1sh-worked-best_1haunnN.jpg)
The plot above only shows an indication of the relative speeds of the different Time required to make a prediction using the network. Use the plot below to compare the ImageNet validation accuracy with the Choosing a network is generally a tradeoff between theseĬharacteristics. The most important characteristics are network accuracy, Pretrained networks have different characteristics that matter when choosing a network Try more pretrained networks, see Train Deep Learning Network to Classify New Images. For a simple example, see Get Started with Transfer Learning. For more information, see Transfer Learning. Take layers from a network trained on a large data set andįine-tune on a new data set. For an example, see Extract Image Features Using Pretrained Network. For more information, see Feature Extraction. Train another machine learning model, such as a support vector machine You can use these activations as features to Use a pretrained network as a feature extractor by using the layerĪctivations as features. Network for classification, see Classify Image Using GoogLeNet.
#NN TOP 100 MODELS HOW TO#
For an example showing how to use a pretrained Apply pretrained networks directly to classification problems.