easy5 pts · first answer countsfine-tuningtransfer-learning
Training only a new head
You freeze every layer of a pretrained image model and train only a new final classification layer on your data. What is this approach usually called?
Discussion 0
Join free to post in the discussion.
Related questions
- Preprocessing for a pretrained ResNetMachine Learning Engineer
- 800 labelled imagesMachine Learning Engineer
- What full fine-tuning changesMachine Learning Engineer
- Why begin from a pretrained modelMachine Learning Engineer
- Learning rate for fine-tuningMachine Learning Engineer
- Shape of SFT training dataMachine Learning Engineer
- Reading train and validation lossMachine Learning Engineer
- Meaning of PEFTMachine Learning Engineer