Transfer Learning Resnet Pytorch. Im trying to use resnet 18 and 34 for transfer learning. Dataset i used cifar 100 as dataset and you can read the description below according to the docs.
Instructed by ali tourani at university of guilan. Pytorch is a library developed for python specializing in deep learning and natural language processing. There are two primary types of transfer learning.
Although my loss cross entropy is decreasing slowly the accuracy remains extremely low.
This method is super simple as it allows us to treat the pre trained cnn as a feature extractor and then pass those features. Check the architecture of your model in this case it is a densenet 161. Here we will freeze the weights for all of the network except that of the final fully. Although my loss cross entropy is decreasing slowly the accuracy remains extremely low.