You can run the code for this section in this jupyter notebook link. Rather, it saves a path to the file containing the class, which is used during load time. Let us run an experiment to get a more quantitative view of the execution time. Accuracy : To save a DataParallel model generically, save the model.
In this experiment, we train ModelParallelResNet50 and the existing torchvision. To save multiple components, organize them in a dictionary and use torch. After the training, the models will not produce any useful predictions, but we can get a reasonable understanding of the execution times. FC 1 Parameters print list model. When saving a model comprised of multiple torch. If you wish to resuming training, call model.
