Transfer Learning – Image Classification using Tensorflow
#deeplearning #transferlearning #imageclassification
For end to end image classification and object detection video you can check
End to End Image Classifier GPU - https://youtu.be/CZMGxRjzbHA
End to End Image Classifier TPU - https://www.youtube.com/watch?v=AuKu4BkzlU4
Transfer Learning - https://www.youtube.com/watch?v=QE0Ku_pVJDA
Object Detection - https://www.youtube.com/watch?v=vSwEbscJTKk
Transfer learning in machine learning focuses on storing knowledge gained while solving one problem and applying it to a different but related problem. Now take you have a classifier that can detect lemons not this knowledge can be transferred to detect oranges or sweet lime with some minimal fine tuning rather training a complete classifier from scratch
It also is a way when you have only limited samples of each class. If you have lot of data then you can train a classifier from scratch
Transfer learning only works if the initial and target problems of both models are similar enough. If the new task is too far from the data initial model was trained on, then the transfer learning might not work well
#deeplearning #transferlearning #imageclassification
For end to end image classification and object detection video you can check
End to End Image Classifier GPU – https://youtu.be/CZMGxRjzbHA
End to End Image Classifier TPU – https://www.youtube.com/watch?v=AuKu4BkzlU4
Transfer Learning – https://www.youtube.com/watch?v=QE0Ku_pVJDA
Object Detection – https://www.youtube.com/watch?v=vSwEbscJTKk
Transfer learning in machine learning focuses on storing knowledge gained while solving one problem and applying it to a different but related problem. Now take you have a classifier that can detect lemons not this knowledge can be transferred to detect oranges or sweet lime with some minimal fine tuning rather training a complete classifier from scratch
It also is a way when you have only limited samples of each class. If you have lot of data then you can train a classifier from scratch
Transfer learning only works if the initial and target problems of both models are similar enough. If the new task is too far from the data initial model was trained on, then the transfer learning might not work well
