This model performs car recognition on street shots of vehicles. It accepts up to 5000 x 5000 pixel RGB, JPEG or PNG files. It outputs a JSON file which includes the car class (model and year number), prediction probability, and explanation (in the form of temp and mask).
This model can be used to determine the makes and models of cars from different types of footage.
Create a Modzy account to get started →
Explainable – This model has a built-in explainability feature. What is model explainability?.
88% Recall – A higher recall score indicates that the model finds and predicts correct labels for the majority of the classes it is supposed to find.
This model was tested on a subset of the Wikipedia Language Identification Dataset and achieved an average recall of 0.95, an average precision of 0.95, and an average F1 score of 0.95. During evaluation, it was observed that the model’s selectance improved as the length of the text it processes increased.
This model was initially trained using the ResNet-152 weights for the ImageNet data, and then fine-tuned using the Stanford Cars dataset. The architecture allows the creation of very deep neural networks while eliminating the vanishing gradient problem. It does this by using “identity shortcut connections” which are shortcuts that are made available throughout the network to enable the gradient to propagate properly. The networks gradually restore the skipped layers at it learns the feature space.
This model was trained on the Stanford Cars dataset which contains 16,185 images of 196 classes of cars. The dataset was split into 8,144 training images and 8,041 testing images where each class was split in a roughly 50-50 split.
The performance of the model was tested on the test set of the Stanford Cars dataset. This contained 8,041 automobile images with their classification.
The input(s) to this model must adhere to the following specifications:
This model will output the following:
Get a video demo of Modzy, the ModelOps and MLOps software platform that businesses use to deploy, integrate, run, and monitor AI—anywhere.